Laboratory Medicine Practice Guidelines

Guidelines and Recommendations for Laboratory

Analysis in the Diagnosis and Management of
Diabetes Mellitus
Edited by David B. Sacks
ii
 The National Academy of Clinical Biochemistry
Presents

LABORATORY MEDICINE PRACTICE GUIDELINES

Guidelines and Recommendations for

Laboratory Analysis in the Diagnosis
and Management of Diabetes Mellitus

EDITED BY
David B. Sacks

David B. Sacks M. Sue Kirkman

Department of Laboratory Medicine, National Institutes of American Diabetes Association, Alexandria, VA
Health, Bethesda, MD
Ake Lernmark
Mark Arnold Department of Clinical Sciences, Lund University/CRC,
Department of Chemistry, University of Iowa, Iowa City, IA Skane University Hospital Malm, Malm, Sweden

George L. Bakris Boyd E. Metzger

Department of Medicine, Hypertensive Disease Unit, Section Division of Endocrinology, Northwestern University,
of Endocrinology, Diabetes and Metabolism, University of Feinberg School of Medicine, Chicago, IL
Chicago, Chicago, IL
David M. Nathan
David E. Bruns Massachusetts General Hospital and Harvard Medical School,
Department of Pathology, University of Virginia Medical Diabetes Center, Boston, MA, and Pathobiology, University
School, Charlottesville, VA of Toronto, Ontario, Canada

Andrea Rita Horvath

Screening and Test Evaluation Program, School of Public
Health, University of Sydney, SEALS Department of Clinical
Chemistry, Prince of Wales Hospital, Sydney, Australia
Copyright 2011 by the American Association for Clinical Chemistry, Inc and the American Diabetes Association. All rights reserved.

Single copies for personal use may be printed from authorized Internet sources such as the
NACBs home page (http://www.aacc.org/members/nacb/LMPG/Pages/default.aspx), provided it is printed in its entirety, including this notice.
Printing of selected portions of the document is also permitted for personal use, provided the user also prints and attaches the title page and
cover pages to the selected reprint or otherwise clearly identifies the reprint as having been produced by the NACB. Otherwise, this document
may not be reproduced in whole or in part, stored in a retrieval system, translated into another language, or transmitted in any form without
express written permission of the National Academy of Clinical Biochemistry. Such permission may be requested from NACB, 1850 K Street,
Suite 625, Washington, DC, 20006-2213. Permission will ordinarily be granted, provided the NACB logo and the following notice appear
prominently at the front of the document:
Reproduced (translated) with permission of the National Academy of Clinical Biochemistry, Washington, DC.

This document (PID 6278) was approved by the National Academy of Clinical Biochemistry Board of Directors in January 2011. The NACB
is the Academy of the American Association for Clinical Chemistry.
Table of Contents

Preamble vi

Chapter 1. Introduction 1

Chapter 2. Glucose 3

Chapter 3. Glucose Meters 9

Chapter 4. Continuous Minimally Invasive Glucose Analysis 15

Chapter 5. Noninvasive Glucose Analysis 17

Chapter 6. Gestational Diabetes Mellitus 19

Chapter 7. Urinary Glucose 21

Chapter 8. Ketone Testing 23

Chapter 9. Hb A1c 25

Chapter 10. Genetic Markers 31

Chapter 11. Autoimmune Markers 35

Chapter 12. Albuminuria (formerly microalbuminuria) 39

Chapter 13. Miscellaneous Potentially Important Analytes 43

References 45

Acknowledgment 57

Appendix 59
Preamble

Methods for Updating the NACB Diabetes Mellitus Laboratory Medicine

Practice Guidelines

The National Academy of Clinical Biochemistry (NACB) has developed evidence-based guidelines on topics related to the practice
of laboratory medicine. These guidelines are updated approximately every 5 years and are available on the NACB Web site (http://
www.aacc.org/members/nacb). The NACB issued its Guidelines and Recommendations for Laboratory Analysis in the Diagnosis
and Management of Diabetes Mellitus in 2002 (1). These recommendations were reviewed and updated via an evidence-based
approach, especially in areas in which new evidence has emerged since the 2002 publication. The process of updating guideline rec-
ommendations followed the standard operating procedures for preparing, publishing, and editing NACB laboratory medicine practice
guidelines. The key steps are summarized in Fig. 1 and are explained below. The guideline-updating process was designed to fulfill
the methodological quality criteria of the Appraisal of Guidelines for Research and Evaluation (AGREE) II Instrument (2).

Figure 1: Process of updating the NACB Diabetes Mellitus guideline.

STEP 1: Determine the scope and key topics of the guideline
STEP 2: Determine the target group of the guideline and establish a multidisciplinary guideline team
STEP 3: Identify key areas for revisions and define the structure and methodology of the updated guideline
STEP 4: Define and prioritize key questions
STEP 5: Search the literature systematically for high priority questions and select relevant key publications
STEP 6: Subject selected key publications to critical expert review Extract data into evidence tables
STEP 7: Define the quality of evidence underlying each recommendation
STEP 8: Release the first draft of the guideline for public comments
STEP 9: Incorporate comments, grade recommendations and prepare the second draft of the guideline
STEP 10: Release the second draft of the guideline for public comments and submit the final draft to NACB for review and approval

STEP 1: Determine the Scope and Key Topics of the Guideline

The scope and purpose of this guideline is primarily to focus on the laboratory aspects of testing in the contexts of type 1 and type
2 diabetes mellitus (DM). It does not deal with any issues related to the clinical management of DM that are already covered in
the American Diabetes Association (ADA) or WHO guidelines. In January of each year, the ADA publishes in Diabetes Care a
supplement entitled Clinical Practice Recommendations. This supplement, a compilation of all ADA position statements related
to clinical practice, is an important resource for healthcare professionals who care for people with DM. The intention of the
NACB guideline is to supplement the ADA guidelines and to avoid duplication or repetition of information. Therefore, it focuses
on practical aspects of care to assist in making decisions related to the use or interpretation of laboratory tests during screening,
diagnosing, or monitoring of patients with DM.

STEP 2: Determine the Target Group of the Guideline and Establish a Multidisciplinary
Guideline Team
The primary target of these recommendations includes general practitioners, physicians, nurses, and other healthcare practitioners
directly involved in the care of diabetic patients, as well as laboratory professionals. The guidelines can be used by patients where
relevant (e.g., self-monitoring of blood glucose), policy makers, and payers for healthcare, as well as by researchers. In addition,
the guidelines may advise industry/manufacturers on how to use or develop assays for the laboratory management of DM.
The guideline committee included representatives of key stakeholders to whom the recommendations are meant to apply primar-
ily. Experts of the guideline team are listed in the guideline (3) and represented the NACB (D.B. Sacks, D.E. Bruns) and the ADA
(M.S. Kirkman). The guideline committee included clinical experts (G.L. Bakris, A. Lernmark, B.E. Metzger, D.M. Nathan) and
laboratory experts (D.B. Sacks, D.E. Bruns, M. Arnold, A.R. Horvath) whose key area of research and practice is DM. Some members

vi
of the committee provided additional support in evidence-based guideline-development methodology (D.E. Bruns, A.R. Horvath,
D.B. Sacks). Members of the guideline committee were mostly from the US. The perspectives and views of various international
and national organizations representing the wider laboratory and clinical professions and practice settings, as well as other potential
stakeholders (including other healthcare providers, patients, policy makers, regulatory bodies, health insurance companies, research-
ers, and industry) were taken into account during the public-consultation process (see steps 8 and 10; see Appendix Table 1).
The guideline committee received no sponsorship, honoraria, or other direct funding related to the development of this guideline.
The NACB supported the development process by providing funds to cover the expenses of meetings and consensus conferences and
provided administrative support. The views of the NACB officers and staff have not influenced the content of the guideline.
All authors who contributed to the development of the recommendations of this guideline have declared (via the official dis-
closure form of the NACB) any financial, personal, or professional relationships that might constitute conflicts of interest with this
guideline. These disclosures are part of the guideline document published on the NACB Web site.

STEP 3: Identify Key Areas for Revisions and Define the Structure and Methodology of the
Updated Guideline
The chairman of the guideline committee (D.B. Sacks) acted as editor and assigned lead authors to each section. Authors reviewed
the 2002 edition of the NACB DM guideline (1) and identified key areas for revisions and updating. The guideline team discussed
the scope and methods of the updating process at a face-to-face meeting, which was followed by numerous teleconferences
and e-mail exchanges among authors that were coordinated by the editor and the NACB. The guideline group decided that the
structure of the guideline would remain the same as the 2002 document and that it would cover virtually all key analytes that are
used primarily in the diagnosis and management of individuals with DM. As before, the testing of lipids and related cardiovascular
risk factors is not covered in this update but is addressed in a separate NACB guideline (4). For each area of testing discussed, the
guideline highlights the clinical use and rationale for the test or tests; the preanalytical, analytical, and interpretive aspects of each
test; and, where relevant, emerging considerations for future research.

STEP 4: Define and Prioritize Key Questions

The lead authors used the review process outlined above to define specific key questions to enter on a standard form developed for
this process. These questions were sent to all members of the guideline committee for independent review and prioritization, a process
that used preset criteria related to the relationship between testing and outcomes (see Appendix Table 2). Authors used the categories
and explanatory notes provided (see Appendix Table 2) to document the rationale for prioritization or individually provided their own
reasoning. Authors assigned priority scores on a scale of 1 to 4 (most important, important, moderately important, or least important,
respectively). The independent replies collected from all authors were the basis for drafting a consensus priority list. Final key ques-
tions with priority scores and categories of reasoning are presented in the evidence tables (see Appendix Table 3).

STEP 5: Search the Literature Systematically for High-Priority Questions and Select Relevant Key
Publications
Key questions that earned the highest priority score were covered by a more systematic approach during the search and evalua-
tion of the evidence currently available in the literature. Other topics that were considered less important were dealt with in a less
rigorous way. Because this guideline is an update of the 2002 version, authors limited their searches to the period beginning in
January 2002. Guidelines related to the topic were searched in the Agency for Healthcare Research and Quality National Guideline
Clearinghouse database (http://www.guideline.gov/). Systematic reviews and metaanalyses were searched by using the Clinical
QueriesFind Systematic Reviews function of PubMed. If no such publications were found, PubMed, Embase, and other data-
bases were used to search the primary literature. Because the group of authors included leading experts in their fields, the authors
personal files, communications with experts, and unpublished or ongoing-trial data were also made available to be used in the
guideline-updating process. Additional literature citations were added during the comment periods (see below).
Authors selected relevant key publications for updating each section, and the editor of the guideline (D.B. Sacks) and lead
authors of other sections (D.E. Bruns, M.S. Kirkman, D.M. Nathan) acted as independent expert reviewers to avoid biased selec-
tion of papers. When the guideline team retrieved and agreed with existing guideline recommendations that had already covered
the key question comprehensively and had reached concordant conclusions, the guideline team simply adopted and referenced the
published recommendations in order to avoid duplicate publication.

STEP 6: Subject Selected Key Publications to Critical Expert Review; Extract Data into
Evidence Tables
Critical review of selected key publications formed the basis for establishing the level and quality of the evidence underlying each
recommendation (see step 7 for details). Section authors and a methodology expert (A.R. Horvath) extracted data into evidence

vii
tables (see Appendix for Table 3). These tables list all key questions together with their priority scores (step 4). Related recommen-
dations and their grades from the 2002 guideline were aligned with those of the new updated recommendations (see columns 1 and
2 in Appendix Table 3). In the updated recommendation, authors highlighted changes to the original text in boldface and provided
explanation for the changes where necessary (column 3). Key references supporting the new recommendation were listed (column 4).

STEP 7: Define the Quality of Evidence Underlying Each Recommendation

To our knowledge, no uniformly accepted grading scheme exists for rating the quality of evidence and the strength recom-
mendations when questions related to laboratory testing for the screening, diagnosis, prognosis, and monitoring of a condition
are addressed (5). The guideline group agreed that the grading scheme of the ADA, which was used in the 2002 version of this
guideline (1), is applicable predominantly to therapeutic recommendations and that its use in this diagnostic guideline was thus
impracticable. Therefore, we developed a grading system by adapting the key elements of evidence-rating frameworks employed
by various international guideline agencies, the US Preventive Services Task Force, and the Grading of Recommendations
Assessment, Development and Evaluation (GRADE) Working Group (612). In this system, the overall quality of the body of
evidence (step 7) and the strength of recommendations (step 9) are graded separately. Rating the quality of the body of evidence
is based on (a) the level of evidence of individual studies defined by their study design and methodological quality; (b) the consis-
tency of results across various studies; (c) the directness of comparisons; and (d) the precision-of-effect estimates. Table 1 provides
a detailed explanation of evidence-level categories and these elements of the rating scheme for the quality of evidence.

Table 1. Grading the quality of evidence.

THE QUALITY OF THE BODY OF EVIDENCE IS BASED ON:

Level of evidence: This refers to the detailed study methods and the quality of their execution, i.e., the methodological quality
of individual studies. The level of evidence can be:
High: if the study has an appropriate design for the question being asked and if it is well conducted in representative
populations and is free from design-related biases.
Moderate: if the study has an appropriate design for the question being asked but suffers from some design-related
biases that might influence the conclusions to a certain extent but would not affect patient-important outcomes or conclu-
sions significantly.
Low: if the study is wrongly designed and conducted and there is a high likelihood that its conclusions are grossly biased
and misleading.
Consistency of results across various studies: i.e., when results are heterogeneous across studies, inconsistency of results
lowers the strength of evidence.
Directness of comparisons: Indirectness applies and lowers quality when, for example:
Evidence is indirectly related to the actual question;
The study population differs from that to which the study results would be applied in practice;
The test in the study differs (e.g., in its analytical performance, or a new generation of the same test has emerged) from
the one commonly used or recommended in practice;
The outcome of interest for the guideline differs from the one studied in the trial.
Precision-of-effect estimates: If the study is relatively small and includes few patients or events, the confidence interval
around the effect estimate is relatively large, and imprecision of results leads to downgrading the quality of evidence.

RATING SCALE FOR THE OVERALL QUALITY OF THE BODY OF EVIDENCE:

High: Further research is very unlikely to change our confidence in the estimate of effect. The body of evidence comes from high-
level individual studies that are sufficiently powered and provide precise, consistent, and directly applicable results in a relevant
population.
Moderate: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the
estimate and the recommendation. The body of evidence comes from high-/moderate-level individual studies that are sufficient
to determine effects, but the strength of the evidence is limited by the number, quality, or consistency of the included studies; by
the generalizability of results to routine practice; or indirect nature of the evidence.
Low: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change
the estimate and the recommendation. The body of evidence is of low level and comes from studies with serious design flaws or
with evidence that is indirect.
Very low: Any estimate of effect is very uncertain. Recommendation may change when higher-quality evidence becomes avail-
able. Evidence is insufficient to assess the effects on health outcomes because of limited number or power of studies, important
flaws in their design or conduct, gaps in the chain of evidence, or lack of information.

viii
 Members of the guideline committee received detailed explanations and guidance, as well as methodological support,
on how to use the grading scheme. At this stage of the guideline-development process, section authors indicated the study
design (see column 5 in Appendix Table 3) and the level of evidence (column 6) of all individual studies listed in the evidence
tables. The quality of the totality of the evidence underlying each recommendation was established by means of the criteria
mentioned above (column 7).

STEP 8: Release the First Draft of the Guideline for Public Comments
The first draft of the guideline was released on the NACB Web site for soliciting of public review and feedback. The still non-
graded draft recommendations were sent to a number of external organizations (see Appendix Table 1) for peer review and expert
comments that could be submitted either via the NACB Web site or by mail. The draft guideline was also presented at the Arnold
O. Beckman consensus conference in 2007, and the discussions at this conference were recorded.

STEP 9: Incorporate Comments, Grade Recommendations, and Prepare the Second Draft of
the Guideline
The guideline team reviewed and discussed the comments that were received and made many changes to the first draft to reflect the
views of external peers, organizations, or individuals. The amended draft of the guideline was also presented at the 2009 AACC
annual meeting and used for grading recommendations.
The grade or strength of recommendation refers to the extent of collective confidence that the desirable effects of a recommen-
dation outweigh the potential undesirable effects. Desirable effects of a recommendation may include improved health-related,
organizational, or economic outcomes or aspects of care. The quality of evidence (step 7, Table 1) is only one element in making
recommendations for practice. Scientific evidence was supplemented with considered judgment that balanced the potential clinical
benefits and harms with perceived patients preferences, bioethical considerations, and organizational and economic impacts of
testing (5, 6, 912). Considered judgment therefore may have upgraded or downgraded a recommendation. Categories for grading
recommendations are shown in Table 2.
During the considered-judgment process, the guideline committee was primarily driven by two core bioethical values
beneficence and nonmalevolence. The guideline group also observed the first principle of bioethics, i.e., respect for patients
autonomy and the decision-making capacities of individuals to make their own choices. The guideline group assumes that the
target users will also deal with this core bioethical principle when using these guidelines in practice (13). The guideline committee
acknowledges that it was not able to cover universally other bioethical principles, such as justice and equity. As mentioned above,
the members of the guideline team, as well as individuals who commented on the recommendations, were mostly from North
America and other developed countries. Their views and experiences therefore unavoidably affected the considered-judgment
and consensus processes involved in formulating recommendations. The guideline team also could not consider explicitly the cost
implications of the recommendations in various resource settings, although recommendations were formulated in a generic way
and in a cost-conscious manner.
Recommendations in diagnostic guidelines frequently are supported primarily by expert consensus. This reflects the often
poor quality of evidence, or the lack or indirectness of evidence that the intervention is relevant to patient outcomes. To avoid
the influence of dominant personalities and overrepresentation of the individual opinions or views of experts, the guideline
team reached consensus when the evidence base was inconsistent, weak, or lacking. The matrix in Table 3 assisted in the
assignment of final grades to recommendations. The methodology expert pregraded recommendations by using the informa-
tion in columns 5, 6, and 7 of the evidence tables provided by committee members (see Appendix Table 3). Authors reviewed
these grades and returned the amended evidence tables to the methodology expert for completion. Committee members added
comments or explanatory notes when necessary (column 8) to enhance the transparency and reproducibility of the considered-
judgment and consensus process of grading and to address the adaptability and applicability of the final recommendations. All
sections were reviewed by the ADA representative (M.S. Kirkman), a clinical expert (D.M. Nathan), and a methodology expert
(A.R. Horvath) and were edited by the chairman of the guideline committee (D.B. Sacks).

STEP 10: Release the Second Draft of the Guideline for Public Comments and Submit the Final
Draft to the NACB for Review and Approval
The second draft of the guideline with graded recommendations was posted on the NACB Web site for a last call for public
comments. The guideline recommendations were also reviewed by the Professional Practice Committee of the ADA. Several
comments were received and incorporated, and the final guideline draft was submitted for review by the joint Evidence-Based
Laboratory Medicine Committee of the AACC and the NACB. After addressing the reviewers comments, the guideline committee
referred the guideline to the NACB Board of Directors, which approved it before its official release for publication.

ix
Table 2. Grading the strength of recommendations.

A. THE NACB STRONGLY RECOMMENDS ADOPTION

Strong recommendations for adoption are made when:
There is high-quality evidence and strong or very strong agreement of experts that the intervention improves important
health outcomes and that benefits substantially outweigh harms; or
There is moderate-quality evidence and strong or very strong agreement of experts that the intervention improves impor-
tant health outcomes and that benefits substantially outweigh harms.
Strong recommendations against adoption are made when:
There is high-quality evidence and strong or very strong agreement of experts that the intervention is ineffective or that
benefits are closely balanced with harms, or that harms clearly outweigh benefits; or
There is moderate-quality evidence and strong or very strong agreement of experts that the intervention is ineffective or
that benefits are closely balanced with harms, or that harms outweigh benefits.

B. THE NACB RECOMMENDS ADOPTION

Recommendations for adoption are made when:
There is moderate-quality evidence and level of agreement of experts that the intervention improves important health
outcomes and that benefits outweigh harms; or
There is low-quality evidence but strong or very strong agreement and high level of confidence of experts that the inter-
vention improves important health outcomes and that benefits outweigh harms; or
There is very lowquality evidence but very strong agreement and very high level of confidence of experts that the inter-
vention improves important health outcomes and that benefits outweigh harms.
Recommendations against adoption are made when:
There is moderate-quality evidence and level of agreement of experts that the intervention is ineffective or that benefits
are closely balanced with harms, or that harms outweigh benefits; or
There is low-quality evidence but strong or very strong agreement and high level of confidence of experts that the inter-
vention is ineffective or that benefits are closely balanced with harms, or that harms outweigh benefits; or
There is very lowquality evidence but very strong agreement and very high level of confidence of experts that the inter-
vention is ineffective or that benefits are closely balanced with harms, or that harms outweigh benefits.

C. THE NACB CONCLUDES THAT THERE IS INSUFFICIENT INFORMATION TO MAKE A RECOMMENDATION

Grade C is applied in the following circumstances:
Evidence is lacking, scarce, or of very low quality, the balance of benefits and harms cannot be determined, and there is
no or very low level of agreement of experts for or against adoption of the recommendation.
At any level of evidenceparticularly if the evidence is heterogeneous or inconsistent, indirect, or inconclusiveif there
is no agreement of experts for or against adoption of the recommendation.

GPP. THE NACB RECOMMENDS IT AS GOOD PRACTICE POINT

Good practice points (GPPs) are recommendations mostly driven by expert consensus and professional agreement and are
based on the information listed below and/or professional experience, or widely accepted standards of best practice. This
category applies predominantly to technical (e.g., preanalytical, analytical, postanalytical), organizational, economic, or quality-
management aspects of laboratory practice. In these cases, evidence often comes from observational studies, audit reports,
case series or case studies, nonsystematic reviews, guidance or technical documents, nonevidence-based guidelines, per-
sonal opinions, expert consensus, or position statements. Recommendations are often based on empirical data, usual practice,
quality requirements, and standards set by professional or legislative authorities or accreditation bodies, etc.

Table 3. Matrix for the assignment of grades to guideline recommendations.

Strength of recommendation Quality of evidence Agreement of experts

(Table 2) (Table 1)
A: Strongly recommended High Strongvery strong
Moderate
B: Recommended Moderate Moderate
Low Strongvery strong
Very low Very strong
C: Insufficient information to make rec- Very low No agreement or very weak
ommendation Low, moderate, high
GPP: Good practice point Expert consensus on best practice

x
Implementation and Review
To assist implementation, key recommendations of the guideline and their grades are summarized below. Key diagnostic and risk-
assessment criteria are presented in tables, and a diagnostic algorithm is provided for urinary albumin testing. Most recommendations
are worded to represent standards of care and thus can be easily converted to key performance indicators for local audit purposes.
Although recommendations have been developed for national and international use and are intended to be generic, certain
elements of this guideline will not reflect views that are universally held, and other elements may have limited applicability in
healthcare settings that lack sufficient resources for adopting the recommendations. The guideline committee advises users to
adapt recommendations to their local settings. During such adaptation processes, the evidence tables provided (see Appendix
Table 3) might assist users in making informed decisions.
The next review of this guideline is planned in 5 years, unless substantial new evidence emerges earlier for high-priority areas
in the laboratory management of patients with DM.

Nonstandard Abbreviations
Nonstandard abbreviations throughout this document are as follows: IDDM, insulin-dependent diabetes mellitus; GDM, gesta-
tional diabetes mellitus; FPG, fasting plasma glucose; NHANES, National Health and Nutrition Examination Survey; OGTT,
oral glucose tolerance test; NACB, National Academy of Clinical Biochemistry; ADA, American Diabetes Association; GPP,
good practice point; IDF, International Diabetes Federation; Hb A1c, hemoglobin A1c; QALY, quality- adjusted life-year; UKPDS,
United Kingdom Prospective Diabetes Study; DCCT, Diabetes Control and Complications Trial; CAP, College of American
Pathologists; DKA, diabetic ketoacidosis; ICU, intensive care unit; SMBG, self-monitoring of blood glucose; GHb, glycated
hemoglobin; DiGEM, Diabetes Glycaemic Education and Monitoring (trial); ISO, International Organization for Standardization;
CGM, continuous glucose monitoring; FDA, US Food and Drug Administration; IADPSG, International Association of Diabetes
and Pregnancy Study Groups; HAPO, Hyperglycemia and Adverse Pregnancy Outcome (study); AcAc, acetoacetate; HBA,
-hydroxybutyric acid; NGSP, National Glycohemoglobin Standardization Program; eAG, estimated average glucose; ADAG,
A1c-Derived Average Glucose (study); ACCORD, Action to Control Cardiovascular Risk in Diabetes (study); HEDIS, Healthcare
Effectiveness Data and Information Set; MODY, maturity-onset diabetes of the young; ICA, autoantibody to islet cell cytoplasm;
HNF, hepatocyte nuclear factor; VNTR, variable nucleotide tandem repeat; IAA, insulin autoantibody; GAD65A, autoantibody
to 65-kDa isoform of glutamic acid decarboxylase; IA-2A, autoantibody to insulinoma antigen 2; IA-2A, autoantibody to insuli-
noma antigen 2; ZnT8A, autoantibody to zinc transporter 8; LADA, latent autoimmune diabetes of adulthood; DPT-1, Diabetes
Prevention Trial of Type 1 Diabetes; DASP, Diabetes Autoantibody Standardization Program; JDF, Juvenile Diabetes Foundation;
JNC, Joint National Committee; NKF, National Kidney Foundation; eGFR, estimated glomerular filtration rate.

Table 4. Key Recommendations

Recommendation Grade
Glucose
When glucose is used to establish the diagnosis of diabetes, it should be measured in venous A (high)
plasma.
When glucose is used for screening of high-risk individuals, B (moderate)
it should be measured in venous plasma.
Plasma glucose should be measured in an accredited laboratory when used for diagnosis of or GPP
screening for diabetes.
Outcome studies are needed to determine the effectiveness of screening. C (moderate)
Routine measurement of plasma glucose concentrations in an accredited laboratory is not recom- B (low)
mended as the primary means of monitoring or evaluating therapy in individuals with diabetes.
Blood for fasting plasma glucose analysis should be drawn in the morning after the individual has B (low)
fasted overnight (at least 8 h).
To minimize glycolysis, one should place the sample tube immediately in an icewater slurry, and B (moderate)
plasma should be separated from the cells within 30 min. If that cannot be achieved, a tube
containing a rapidly effective glycolysis inhibitor, such as citrate buffer, should be used for
collecting the sample. Tubes with only enolase inhibitors, such as sodium fluoride, should not
be relied on to prevent glycolysis.

xi
Table 4. Key Recommendations (Cont'd)

Recommendation Grade
On the basis of biological variation, glucose measurement should have an analytical imprecision B (low)
2.9%, a bias 2.2%, and a total error 6.9%. To avoid misclassification of patients, the goal
for glucose analysis should be to minimize total analytical error, and methods should be without
measurable bias.
Glucose Meters
There are insufficient published data to support a role for portable meters and skin-prick (finger- C (moderate)
stick) blood samples in the diagnosis of diabetes or for population screening.
The imprecision of the results, coupled with the substantial differences among meters, precludes A (moderate)
the use of glucose meters from the diagnosis of diabetes and limits their usefulness in screen-
ing for diabetes.
SMBG is recommended for all insulin-treated patients with diabetes. A (high)
In patients with type 2 diabetes treated with diet and oral agents, SMBG may help achieve better C (high)
control, particularly when therapy is initiated or changed. Data are insufficient, however, to
claim an associated improvement of health outcomes. The role of SMBG in patients with stable
type 2 diabetes controlled by diet alone is not known.
Patients should be instructed in the correct use of glucose meters, including quality control. B (moderate)
Comparison between SMBG and concurrent laboratory glucose analysis should be performed
at regular intervals to evaluate the performance of the meters in the patients hands.
Multiple performance goals for portable glucose meters have been proposed. These targets vary C (low)
widely and are highly controversial. Manufacturers should work to improve the imprecision of
current meters, with an intermediate goal of limiting total error for 95% of samples to 15% at
glucose concentrations 5.6 mmol/L (100 mg/dL) and to<0.8 mmol/L (15 mg/dL) at glucose
concentrations <5.6 mmol/L (100 mg/dL). Lower total error would be desirable and may prove
necessary in tight glucose-control protocols and for avoiding hypoglycemia in all settings.
Meters should measure and report plasma glucose concentrations to facilitate comparison with GPP
assays performed in accredited laboratories.
Studies are needed to determine the analytical goals (quality specifications) for glucose meters in C (moderate)
SMBG and in intensive care units.
Recommendations for future research: Important end-points in studies of SMBG should include, GPP
at a minimum, Hb A1c and frequency of hypoglycemic episodes to ascertain whether improved
meters enable patients to achieve better glucose control.For studies of meter use in intensive
or critical care, important end points include mean blood glucose, frequency of hypoglycemia
and variation of glucose control. Ideally, outcomes (e.g., long-term complications) should also
be examined.
Continuous Minimally Invasive Glucose Analyses
Real-time CGM in conjunction with intensive insulin regimens can be a useful tool to lower Hb A1c A (high)
in selected adults (age >25 years) with type 1 diabetes.
Although the evidence for lowering Hb A1c is not as strong for children, teens, and younger adults, B (moderate)
real-time CGM may be helpful in these groups. Success correlates with adherence to ongoing
use of the device.
Real-time CGM may be a supplemental tool to SMBG in individuals with hypoglycemia unaware- B (low)
ness and/or frequent episodes of hypoglycemia.
Patients require extensive training in using the device. Available devices must be calibrated with GPP
SMBG readings, and the latter are recommended for making treatment changes.
Noninvasive Glucose Analysis
No noninvasive sensing technology is currently approved for clinical glucose measurements of C (very low)
any kind. Major technological hurdles must be overcome before noninvasive sensing technol-
ogy will be sufficiently reliable to replace existing portable meters, implantable biosensors, or
minimally invasive technologies.
Gestational Diabetes Mellitus
All pregnant women not previously known to have diabetes should undergo testing for gestational A (high)
diabetes mellitus at 2428 weeks of gestation.

xii
Table 4. Key Recommendations

Recommendation Grade
Gestational diabetes mellitus should be diagnosed by a 75-g oral glucose tolerance test according A (moderate)
to the IADPSG criteria derived from the HAPO study.
Urinary Glucose
Semiquantitative urine glucose testing is not recommended for routine care of patients with diabe- B (low)
tes mellitus.
Ketone Testing
Ketones measured in urine or blood in the home setting by patients with diabetes and in the clinic/ GPP
hospital setting should be considered only an adjunct to the diagnosis of diabetic ketoacidosis.
Urine ketone measurements should not be used to diagnose or monitor the course of diabetic GPP
ketoacidosis.
Blood ketone determinations that rely on the nitroprusside reaction should be used only as an ad- B (moderate)
junct to diagnose diabetic ketoacidosis and should not be used to monitor diabetic ketoacidosis
treatment. Specific measurement of beta-hydroxybutyric acid in blood can be used for diagno-
sis and monitoring of diabetic ketoacidosis.
Hb A1c
Hb A1c should be measured routinely in all patients with diabetes mellitus to document their degree A (moderate)
of glycemic control.
Laboratories should use only Hb A1c assay methods that are certified by the NGSP as traceable to GPP
the DCCT reference. The manufacturers of Hb A1c assays should also show traceability to the
IFCC reference method.
Laboratories that measure Hb A1c should participate in a proficiency-testing program, such as the GPP
College of American Pathologists Hb A1c survey, that uses fresh blood samples with targets set
by the NGSP Laboratory Network.
Laboratories should be aware of potential interferences, including hemoglobinopathies, that may GPP
affect Hb A1c test results, depending on the method used. In selecting assay methods, labo-
ratories should consider the potential for interferences in their particular patient population. In
addition, disorders that affect erythrocyte turnover may cause spurious results, regardless of
the method used.
Desirable specifications for Hb A1c measurement are an intralaboratory CV<2% and an interlabora- B (low)
tory CV <3.5%. At least 2 control materials with different mean values should be analyzed as
an independent measure of assay performance.
Samples with Hb A1c results below the lower limit of the reference interval or >15% Hb A1c should B (low)
be verified by repeat testing.
Hb A1c values that are inconsistent with the clinical presentation should be investigated further. GPP
Treatment goals should be based on American Diabetes Association recommendations, which in- A (high)
clude generally maintaining Hb A1c concentrations at <7% and more-stringent goals in selected
individual patients if they can be achieved without significant hypoglycemia or other adverse
treatment effects. Somewhat higher intervals are recommended for children and adolescents
and may be appropriate for patients with limited life expectancy, extensive comorbid illnesses,
a history of severe hypoglycemia, or advanced complications (note that these values are ap-
plicable only if the NGSP has certified the assay method as traceable to the DCCT reference).
Hb A1c testing should be performed at least biannually in all patients and quarterly for patients B (low)
whose therapy has changed or who are not meeting treatment goals.
Hb A1c may be used for the diagnosis of diabetes, with values 6.5% being diagnostic. An NGSP A (moderate)
certified method should be performed in an accredited laboratory. Analogous to its use in the
management of diabetes, factors that interfere with or adversely affect the Hb A1c assay will
preclude its use in diagnosis.
Point-of-care Hb A1c assays are not sufficiently accurate to use for the diagnosis of diabetes. B (moderate)
Genetic Markers
Routine measurement of genetic markers is not of value at this time for the diagnosis or manage- A (moderate)
ment of patients with type 1 diabetes. For selected diabetic syndromes, including neonatal
diabetes, valuable information can be obtained with definition of diabetes-associated mutations.

xiii
Table 4. Key Recommendations (Cont'd)

Recommendation Grade
There is no role for routine genetic testing in patients with type 2 diabetes. These studies should A (moderate)
be confined to the research setting and evaluation of specific syndromes.
Autoimmune Markers
Islet cell autoantibodies are recommended for screening nondiabetic family members who wish to B (low)
donate part of their pancreas for transplantation into a relative with end-stage type 1 diabetes.
Islet cell autoantibodies are not recommended for routine diagnosis of diabetes, but standardized B (low)
islet cell autoantibody tests may be used for classification of diabetes in adults and in prospec-
tive studies of children at genetic risk for type 1 diabetes after HLA typing at birth.
Screening patients with type 2 diabetes for islet cell autoantibodies is not recommended at pres- B (low)
ent. Standardized islet cell autoantibodies are tested in prospective clinical studies of type 2
diabetes patients to identify possible mechanisms of secondary failures of treatment of type 2
diabetes.
Screening for islet cell autoantibodies in relatives of patients with type 1 diabetes or in persons B (low)
from the general population is not recommended at present. Standardized islet cell autoanti-
bodies are tested in prospective clinical studies.
There is currently no role for measurement of islet cell autoantibodies in the monitoring of patients B (low)
in clinical practice. Islet cell autoantibodies are measured in research protocols and in some
clinical trials as surrogate end points.
It is important that islet cell autoantibodies be measured only in an accredited laboratory with an GPP
established quality-control program and participation in a proficiency-testing program.
Albuminuria (formerly microalbuminuria)
Annual testing for albuminuria in patients without clinical proteinuria should begin in pubertal or B (moderate)
postpubertal individuals 5 years after diagnosis of type 1 diabetes and at the time of diagnosis
of type 2 diabetes, regardless of treatment.
Urine albumin at concentrations 30 mg/g creatinine should be considered as a continuous risk B (moderate)
marker for cardiovascular events.
The analytical CV of methods to measure albuminuria should be <15%. B (moderate)
Semiquantitative or qualitative screening tests should be positive in >95% of patients with albu- GPP
minuria to be useful for screening. Positive results must be confirmed by analysis in an accred-
ited laboratory.
Currently available dipstick tests do not have adequate analytical sensitivity to detect albuminuria. B (moderate)
Acceptable samples to test for increased urinary albumin excretion are timed collections (e.g., 12 or B (moderate)
24 h) for the measurement of albumin concentration and timed or untimed samples for measure-
ment of the albumincreatinine ratio.
The optimal time for spot urine collection is the early morning. All collections should be at the same GPP
time of day to minimize variation. The patient should not have ingested food within the preceding
2 h, but should be well hydrated (i.e., not volume depleted).
Low urine albumin concentrations (i.e., <30 mg/g creatinine) are not associated with high cardio- A (moderate)
vascular risk if the eGFR is >60 mL min21 (1.73 m2)21 and the patient is normotensive. If the
eGFR is <60 min21 (1.73 m2)21 and/or the level of albuminuria is 30 mg/g creatinine on a spot
urine sample, a repeat measurement should be taken within the year to assess change among
people with hypertension.
Miscellaneous Potentially Important Analytes
There is no role for routine testing for insulin, C-peptide, or proinsulin in most patients with diabetes. B (moderate)
Differentiation between type 1 and type 2 diabetes may be made in most cases on the basis of
the clinical presentation and the subsequent course. These assays are useful primarily for re-
search purposes. Occasionally, C-peptide measurements may help distinguish type 1 from type
2 diabetes in ambiguous cases, such as patients who have a type 2 phenotype but present in
ketoacidosis.
There is no role for measurement of insulin concentration in the assessment of cardiometabolic risk, B (moderate)
because knowledge of this value does not alter the management of these patients.

xiv
Table 4. Key Recommendations

Recommendation Grade
Because current measures of insulin are poorly harmonized, a standardized insulin assay should be GPP
developed to encourage the development of measures of insulin sensitivity that will be practical
for clinical care.
There is no published evidence to support the use of insulin antibody testing for routine care of C (very low)
patients with diabetes.
Abbreviations: GPP, good practice point; SMBG, self-monitoring of blood glucose; Hb A1c, hemoglobin A1c; NGSP, National Glycohemoglobin
Standardization; DCCT, Diabetes Control and Complications Trial; CGM, continuous glucose monitoring; IADPSG, International Association
of Diabetes and Pregnancy Study Groups; HAPO, Hyperglycemia and Adverse Pregnancy Outcome; eGFR, estimated glomerular filtration
rate.

References

1. Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory
analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2002;48:43672.
2. Appraisal of Guidelines for Research & Evaluation II. AGREE II instrument. The AGREE Next Steps Consortium, May 2009,
56 p. http://www.agreetrust.org/resource-centre/agree-ii (Accessed December 2010).
3. Sacks DB, Arnold M, Bakris GL, Bruns DE, Horvath AR, Kirkman MS, et al. Guidelines and recommendations for laboratory
analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2011;57:e1e47.
4. AACC. National Academy of Clinical Biochemistry laboratory medicine practice guidelines. Emerging biomarkers for
primary prevention of cardiovascular disease and stroke. Myers GL, ed. 70 p. http://www.aacc.org/members/nacb/LMPG/
OnlineGuide/PublishedGuidelines/risk/Pages/toc.aspx (Accessed December 2010).
5. Horvath AR. Grading quality of evidence and strength of recommendations for diagnostic tests and strategies. Clin Chem
2009;55:853-5.
6. Scottish Intercollegiate Guidelines Network. SIGN 50: a guideline developers handbook. Edinburgh, January 2008. 112 p.
http://www.sign.ac.uk/guidelines/fulltext/50/index.html (Accessed December 2010).
7. West S, King V, Carey TS, Lohr KN, McKoy N, Sutton SF, Lux L (Research Triangle Institute University of North Carolina
Evidence-Based Practice Center, Research Triangle Park, NC). Systems to rate the strength of scientific evidence. Rockville,
MD: Agency for Healthcare Research and Quality; 2002. Publication no. 02-E016 (evidence reports/technology assessment,
no. 47). Contract no. 209-97-0011. p 5163.
8. Agency for Healthcare Research and Quality. National Guideline Clearinghouse. http://www.guidelines.gov/. (Accessed
December 2010).
9. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading quality of evidence and strength of recommen-
dations. GRADE Working Group. BMJ 2004;328:1490.
10. Australian Government National Health and Medical Research Council. NHMRC levels of evidence and grades for recom-
mendations for developers of guidelines. December 2009. http://www.nhmrc.gov.au/_files_nhmrc/file/guidelines/evidence_
statement_form.pdf (Accessed December 2010).
11. Schnemann HJ, Fretheim A, Oxman AD. Improving the use of research evidence in guideline development: 9. Grading evi-
dence and recommendations. Health Res Policy Syst 2006;4:21.
12. Schnemann HJ, Oxman AD, Brozek J, Glasziou P, Jaeschke R, Vist GE, et al. Grading quality of evidence and strength of
recommendations for diagnostic tests and strategies. BMJ 2008;336:110610.
13. Watine J. What sort of bioethical values are the evidence-based medicine and the GRADE approaches willing to deal with?
J Med Ethics 2010;37:1846.

xv
xvi
Chapter 1
Introduction

Diabetes mellitus is a group of metabolic disorders of car- Table 5. Classification of diabetes mellitus.a
bohydrate metabolism in which glucose is under-utilized I. Type 1 diabetes
and overproduced, causing hyperglycemia. The disease is
A. Immune mediated
classified into several categories. The revised classification,
B. Idiopathic
published in 1997 (1), is presented in Table 5. Type 1 diabe-
II. Type 2 diabetes
tes mellitus, formerly known as insulin-dependent diabetes
III. Other specific types
mellitus (IDDM)10 or juvenile-onset diabetes mellitus, is
A. Genetic defects of beta-cell function
usually caused by autoimmune destruction of the pancreatic
islet beta cells, rendering the pancreas unable to synthesize B. Genetic defects in insulin action
and secrete insulin (2). Type 2 diabetes mellitus, formerly C. Diseases of the exocrine pancreas
known as nonIDDM or adult-onset diabetes, is caused by D. Endocrinopathies
a combination of insulin resistance and inadequate insulin E. Drug or chemical induced
secretion (3, 4). Gestational diabetes mellitus (GDM), which F. Infections
resembles type 2 diabetes more than type 1, develops during G. Uncommon forms of immune-mediated diabetes
approximately 7% (range, 5%15%) of pregnancies, usu- H. Other genetic syndromes sometimes associated with
ally remits after delivery, and constitutes a major risk factor diabetes
for the development of type 2 diabetes later in life. Other IV. GDM
types of diabetes are rare. Type 2 is the most common form, a
From the ADA (378).
accounting for 85%95% of diabetes in developed countries.
Some patients cannot be clearly classified as type 1 or type
2 diabetes (5).
Diabetes is a common disease. The current worldwide 2008 to 9 billion/year). The high costs of diabetes are attrib-
prevalence is estimated to be approximately 250 x 106, and utable to care for both acute conditions (such as hypoglyce-
it is expected to reach 380 x 106 by 2025 (6). The prevalence mia and ketoacidosis) and debilitating complications (12).
of diabetes [based on fasting plasma glucose (FPG) results] The latter include both microvascular complicationspre-
in US adults in 19992002 was 9.3%, of which 30% of the dominantly retinopathy, nephropathy, and neuropathyand
cases were undiagnosed (7). The most recent data, which macrovascular complications, particularly stroke and coro-
were derived from the 20052006 National Health and Nutri- nary artery disease. Together, they make diabetes the fourth
tion Examination Survey (NHANES) with both FPG and most common cause of death in the developed world (13).
2-h oral glucose tolerance test (OGTT) results, show a prev- About 3.8 x 106 people worldwide were estimated to have
alence of diabetes in US persons 20 years old of 12.9% died from diabetes-related causes in 2007 (6).
(approximately 40 x 106) (8). Of these individuals, 40% The National Academy of Clinical Biochemistry (NACB)
(approximately 16 million) are undiagnosed. The prevalence issued its Guidelines and Recommendations for Laboratory
of diabetes has also increased in other parts of the world. For Analysis in the Diagnosis and Management of Diabetes Mel-
example, recent estimates suggest 110 x 106 diabetic indi- litus in 2002 (14). These recommendations were reviewed
viduals in Asia in 2007 (9), but the true number is likely to and updated with an evidence-based approach, especially
be substantially greater, because China alone was thought to in key areas in which new evidence has emerged since the
have 92.4 x 106 adults with diabetes in 2008 (10). 2002 publication. The process of updating guideline recom-
The worldwide costs of diabetes were approximately mendations followed the standard operating procedures for
$232 billion in 2007 and are likely to be $302 billion by 2025 preparing, publishing, and editing NACB laboratory medi-
(6). In 2007, the costs of diabetes in the US were estimated to cine practice guidelines, and the key steps and the grading
be $174 billion (11). The mean annual per capita healthcare scheme are detailed in the Preamble.
costs for an individual with diabetes are approximately 2.3- This guideline focuses primarily on the laboratory aspects
fold higher than those for individuals who do not have diabe- of testing in diabetes.
tes (11). Similarly, diabetes in the UK accounts for roughly To facilitate comprehension and assist the reader, we
10% of the National Health Service budget (equivalent in divide each analyte into several headings and subheadings

1
2 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

(in parentheses), which are: use (diagnosis, screening, moni- (including frequency of measurement and turnaround time);
toring, and prognosis); rationale (diagnosis and screening); and, where applicable, emerging considerations, which alert
analytical considerations (pre-analytical, including reference the reader to ongoing studies and potential future aspects rel-
intervals; and analytical, such as methods), interpretation evant to that analyte.
Chapter 2
Glucose

1. USE for patients who have unequivocal hyperglycemia, i.e., .11.1

mmol/L (200 mg/dL) with symptoms consistent with hypergly-
cemia]. The WHO and the International Diabetes Federation
Recommendation (IDF) recommend either an FPG test or a 2-h postload glucose
When glucose is used to establish the diagnosis of diabetes, test that uses the same cutoffs as the ADA (19) (Table 7). In
it should be measured in venous plasma. 2009, the International Expert Committee (20), which com-
A (high) prised members appointed by the ADA, the European Associa-
tion for the Study of Diabetes, and the IDF, recommended that
diabetes be diagnosed by measurement of hemoglobin A1c (Hb
Recommendation A1c), which reflects long-term blood glucose concentrations
(see Hb A1c section below). The ADA (21) and the WHO have
When glucose is used for screening of high-risk individuals,
endorsed the use of Hb A1c for diagnosis of diabetes.
it should be measured in venous plasma.
Testing to detect type 2 diabetes in asymptomatic people,
B (moderate)
previously controversial, is now recommended for those at risk
of developing the disease (21, 22). The ADA proposes that
Recommendation all asymptomatic people 45 years of age be screened in a
healthcare setting. An Hb A1c, FPG, or 2-h OGTT evaluation is
Plasma glucose should be measured in an accredited labora- appropriate for screening (21). The IDF recommends that the
tory when used for diagnosis of or screening for diabetes. health service in each country decide whether to implement
Good Practice Point (GPP) screening for diabetes (23). FPG is the suggested test. In con-
trast, the International Expert Committee and the ADA have
Recommendation recommended that Hb A1c can be used for screening for diabe-
tes (20, 21, 24) (see section on Hb A1c below). If an FPG result
Outcome studies are needed to determine the effectiveness is ,5.6 mmol/L (100 mg/dL) and/or a 2-h plasma glucose
of screening. concentration is ,7.8 mmol/L (140 mg/dL), testing should be
C (moderate) repeated at 3-year intervals. Screening should be considered at
a younger age or be carried out more frequently in individuals
who are overweight (body mass index 25 kg/m2) or obese
and who have a least 1 additional risk factor for diabetes [see
A. Diagnosis/screening. The diagnosis of diabetes is established
(21) for conditions associated ith increased risk]. Because of
by identifying the presence of hyperglycemia. For many years
the increasing prevalence of type 2 diabetes in children, screen-
the only method recommended for diagnosis was a direct dem-
ing of children is now advocated (25). Starting at age 10 years
onstration of hyperglycemia by measuring increased glucose (or at the onset of puberty if puberty occurs at a younger age),
concentrations in the plasma (15, 16). In 1979, a set of criteria testing should be performed every 3 years in over-weight
based on the distribution of glucose concentrations in high-risk individuals who have 2 other risk factorsnamely family
populations was established to standardize the diagnosis (15). history, a race/ethnicity recognized to increase risk, signs of
These recommendations were endorsed by the WHO (16). In insulin resistance, and a maternal history of diabetes or GDM
1997, the diagnostic criteria were modified (1) to better iden- during the childs gestation (25). Despite these recommenda-
tify individuals at risk of retinopathy and nephropathy (17, 18). tions and the demonstration that interventions can delay and
The revised criteria comprised: (a) an FPG value 7.0 mmol/L sometimes prevent the onset of type 2 diabetes in individuals
(126 mg/dL); (b) a 2-h postload glucose concentration 11.1 with impaired glucose tolerance (26, 27), there is as yet no pub-
mmol/L (200 mg/dL) during an OGTT; or (c) symptoms of dia- lished evidence that treatment based on screening has an effect
betes and a casual (i.e., regardless of the time of the preceding on long-term complications. In addition, the published litera-
meal) plasma glucose concentration 11.1 mmol/L (200 mg/ ture lacks consensus as to which screening procedure (FPG,
dL) (Table 6) (1). If any one of these 3 criteria is met, confir- OGTT, and/or Hb A1c) is the most appropriate (20, 2830). On
mation by repeat testing on a subsequent day is necessary to the basis of an evaluation of NHANES III data, a strategy has
establish the diagnosis [note that repeat testing is not required been proposed to use FPG to screen whites 40 years and other

3
4 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

accepted by all organizations (19, 38). The rationale is based

Table 6. Criteria for the diagnosis of diabetes.a
on data that individuals with FPG values between 5.6 mmol/L
Any one of the following is diagnostic: (100 mg/dL) and 6.05 mmol/L (109 mg/dL) are at increased
1. Hb A1c 6.5% (48 mmol/mol)b risk for developing type 2 diabetes (39, 40). More-recent evi-
OR dence indicates that FPG concentrations even lower than 5.6
2. FPG 7.0 mmol/L (126 mg/dL)c mmol/L (100 mg/dL) are associated with a graded risk for type
OR 2 diabetes (41). Data were obtained from 13 163 men between
3. 2-h Plasma glucose 11.1 mmol/L (200 mg/dL) during 26 and 45 years of age who had FPG values ,5.55 mmol/L
an OGTTd (100 mg/dL) and were followed for a mean of 5.7 years. Men
OR with FPG values of 4.835.05 mmol/L (8791 mg/dL) have a
significantly increased risk of type 2 diabetes, compared with
4. Symptoms of hyperglycemia and casual plasma glu-
cose 11.1 mmol/L (200 mg/dL)e
men with FPG values ,4.5 mmol/L (81 mg/dL). Although the
prevalence of diabetes is low at these glucose concentrations,
a
In the absence of unequivocal hyperglycemia, these criteria should the data support the concept of a continuum between FPG and
be confirmed by repeat testing. From the ADA (378).
b
The test should be performed in a laboratory that is NGSP certified the risk of diabetes.
and standardized to the DCCT assay. Point-of-care assays should
not be used for diagnosis.
c
Fasting is defined as no caloric intake for at least 8 h. Recommendation
d
The OGTT should be performed as described by the WHO, with
a glucose load containing the equivalent of 75 g of anhydrous Routine measurement of plasma glucose concentrations in
glucose dissolved in water. an accredited laboratory is not recommended as the primary
e
Casual is defined as any time of day without regard to time since means of monitoring or evaluating therapy in individuals
previous meal. The classic symptoms of hyperglycemia include
polyuria, polydipsia, and unexplained weight loss.
with diabetes.
B (low)

populations 30 years of age (31). The cost-effectiveness of

screening for type 2 diabetes has been estimated. The incre- B. Monitoring/prognosis. There is a direct relationship
mental cost of screening all persons 25 years of age has been between the degree of chronic plasma glucose control and
estimated to be $236 449 per life-year gained and $56 649 per the risk of late renal, retinal, and neurologic complications.
quality-adjusted life-year (QALY) gained (32). Interestingly, This correlation has been documented in epidemiologic stud-
screening was more cost-effective at ages younger than the 45 ies and clinical trials for both type 1 (42) and type 2 (43)
years currently recommended. In contrast, screening targeted diabetes. The important causal role of hyperglycemia in the
to individuals with hypertension reduces the QALY from $360 development and progression of complications has been doc-
966 to $34 375, with ages between 55 and 75 years being the umented in clinical trials. Persons with type 1 diabetes who
most cost-effective (33). Modeling run on 1 3 106 individuals maintain lower mean plasma glucose concentrations exhibit
suggests considerable uncertainty as to whether screening for a significantly lower incidence of microvascular complica-
diabetes would be cost-effective (34). By contrast, the results tionsnamely diabetic retinopathy, nephropathy, and neu-
of a more recent modeling study imply that screening com- ropathy (44). Although intensive insulin therapy reduced
mencing at 30 or 45 years is highly cost-effective (,$11 000 hypercholesterolemia by 34%, the risk of macrovascular dis-
per QALY gained) (35). Longterm outcome studies are neces- ease was not significantly decreased in the original analysis
sary to provide evidence to resolve the question of the efficacy (44). Longer follow-up documented a significant reduction
of diabetes screening (36). in cardiovascular disease in patients with type 1 diabetes
In 2003, the ADA lowered the threshold for normal FPG treated with intensive glycemic control (45). The effects of
from ,6.1 mmol/L (110 mg/dL) to ,5.6 mmol/L (100 mg/ tight glycemic control on microvascular complications in
dL) (37). This change has been contentious and has not been patients with type 2 diabetes (46) are similar to those with

Table 7. WHO criteria for interpreting 2-h OGTT.a

2-h OGTT result, mmol/L (mg/dL)

0h 2h
Impaired fasting glucoseb .6.1 (110 to ,7.0 (126) ,7.8 (140)
Impaired glucose tolerancec ,7.0 (126) .7.8 (140) to ,11.1 (200)
Diabetesd .7.0 (126) .11.1 (200)
a
Values are for venous plasma glucose using a 75 g oral glucose load. From the WHO (19).
b
If 2-h glucose is not measured, status is uncertain as diabetes or impaired glucose tolerance cannot be excluded.
c
Both fasting and 2-h values need to meet criteria.
d
Either fasting or 2-h measurement can be used. Any single positive result should be repeated on a separate day.
Glucose 5

type 1 diabetes, given the differences in glycemia achieved thermore, it is estimated that 40% of people in the US with
between the active-intervention and control groups in the type 2 diabetes are undiagnosed (8). Notwithstanding this
various trials. Intensive plasma glucose control significantly recommendation, there is no published evidence that popu-
reduced microvascular complications in patients with type 2 lation screening for hyperglycemia provides any long-term
diabetes. Although metaanalyses have suggested that inten- benefit. Outcome studies examining the potential long-term
sive glycemic control reduces cardiovascular disease in indi- benefits of screening are ongoing.
viduals with type 2 diabetes (47, 48), clinical trials have not
consistently demonstrated a reduction in macrovascular dis-
ease (myocar-dial infarction or stroke) with intensive therapy
3. ANALYTICAL CONSIDERATIONS
aimed at lowering glucose concentrations in type 2 diabetes.
Long-term follow-up of the United Kingdom Prospective
Diabetes Study (UKPDS) population supported a benefit of
Recommendation
intensive therapy on macrovascular disease (49), but 3 other
recent trials failed to demonstrate a significant difference To minimize glycolysis, one should place the sample tube
in macrovascular disease outcomes between very intensive immediately in an icewater slurry, and the plasma should
treatment strategies, which achieved Hb A1c concentrations of be separated from the cells within 30 min. If that can-
approximately 6.5% (48 mmol/mol), and the control groups, not be achieved, a tube containing a rapidly effective gly-
which had Hb A1c concentrations 0.8%1.1% higher (5052). colysis inhibitor, such as citrate buffer, should be used for
One study even observed higher cardiovascular mortality in collecting the sample. Tubes with only enolase inhibitors,
the intensive-treatment arm (50). In both the Diabetes Control such as sodium fluoride, should not be relied on to prevent
and Complications Trial (DCCT) and the UKPDS, patients glycolysis.
in the intensive-treatment group maintained lower median B (moderate)
plasma glucose concentrations; however, analyses of the out-
comes were linked to Hb A1c, which was used to evaluate gly-
cemic control, rather than glucose concentration. Moreover, Recommendation
most clinicians use the recommendations of the ADA and
Blood for fpg analysis should be drawn in the morning
other organizations, which define a target Hb A1c concentra-
after the individual has fasted overnight (at least 8 h).
tion as the goal for optimum glycemic control (21, 53).
B (low)
Neither random nor fasting glucose concentrations should
be measured in an accredited laboratory as the primary means
of routine outpatient monitoring of patients with diabetes.
A. Preanalytical. Blood should be drawn in the morning after
Laboratory plasma glucose testing can be used to supple-
an overnight fast (no caloric intake for at least 8 h), during
ment information from other testing, to test the accuracy of
which time the individual may consume water ad libitum (1).
self-monitoring (see below), or to adjust the dosage of oral
Published evidence reveals diurnal variation in FPG, with the
hypoglycemic agents (22, 54). In addition, individuals with
mean FPG being higher in the morning than in the afternoon,
well-controlled type 2 diabetes who are not on insulin therapy
indicating that many diabetes cases would be missed in patients
can be monitored with periodic measurement of the FPG con-
seen in the afternoon (57).
centration, although analysis need not be done in an accredited
Loss of glucose from sample containers is a serious and
laboratory (54, 55).
underappreciated problem (58). Decreases in glucose concen-
trations in whole blood ex vivo are due to glycolysis. The rate
of glycolysisreported to average 5%7%/h [approximately
2. RATIONALE 0.6 mmol/L (10 mg/dL)] (59)varies with the glucose concen-
tration, temperature, leukocyte count, and other factors (60).
A. Diagnosis. The disordered carbohydrate metabolism that Such decreases in glucose concentration will lead to missed
underlies diabetes manifests as hyperglycemia. Therefore, diabetes diagnoses in the large proportion of the population
measurement of either plasma glucose or Hb A1c is the diagnos- who have glucose concentrations near the cut-points for diag-
tic criterion. This strategy is indirect, because hyperglycemia nosis of diabetes.
reflects the consequence of the metabolic derangement, not the The commonly used glycolysis inhibitors are unable
cause; however, until the underlying molecular pathophysiol- to prevent short-term glycolysis. Glycolysis can be attenu-
ogy of the disease is identified, measurement of glycemia is ated by inhibiting enolase with sodium fluoride (2.5 mg/mL
likely to remain an essential diagnostic modality. of blood) or, less commonly, lithium iodoacetate (0.5 mg/
B. Screening. Screening is recommended for several mL of blood). These reagents can be used alone or, more
reasons. The onset of type 2 diabetes is estimated to occur commonly, with such anticoagulants as potassium oxalate,
approximately 47 years (or more) before clinical diagnosis EDTA, citrate, or lithium heparin. Unfortunately, although
(56), and epidemiologic evidence indicates that complica- fluoride helps to maintain long-term glucose stability, the
tions may begin several years before clinical diagnosis. Fur- rates of decline in the glucose concentration in the first hour
6 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

after sample collection are virtually identical for tubes with are significantly higher than those in venous blood [mean, 1.7
and without fluoride, and glycolysis continues for up to 4 h in mmol/L (30 mg/dL), which is equivalent to 20%25% higher
samples containing fluoride (59). After 4 h, the concentration (68)], probably owing to glucose consumption in the tissues.
of glucose in whole blood in the presence of fluoride remains In contrast, the mean difference in fasting samples is only 0.1
stable for 72 h at room temperature (59) (leukocytosis will mmol/L (2 mg/dL) (68, 69).
increase glycolysis even in the presence of fluoride if the
leukocyte count is very high). Reference intervals. Glucose concentrations vary with age
Few effective and practical methods are available for in healthy individuals. The reference interval for children is
prompt stabilization of glucose in whole-blood samples. Loss 3.35.6 mmol/L (60100 mg/dL), which is similar to the adult
of glucose can be minimized in 2 classic ways: (a) immediate interval of 4.16.1 mmol/L (74 110 mg/dL) (70). Note that the
separation of plasma from blood cells after blood collection ADA and WHO criteria (19, 21), not the reference intervals,
(the glucose concentration is stable for 8 h at 25 C and 72 h at are used for the diagnosis of diabetes. Moreover, the threshold
4 C in separated, nonhemolyzed, sterile serum without fluo- for the diagnosis of hypoglycemia is variable. Reference inter-
ride(61)); and (b) placing the blood tube in an icewater slurry vals are not useful for diagnosing these conditions. In adults,
immediately after blood collection and separating the plasma the mean FPG concentration increases with increasing age
from the cells within 30 min (19, 62). These methods are not from the third to the sixth decade (71) but does not increase
always practical and are not widely used. significantly after 60 years of age (72, 73). By contrast, glu-
A recent study showed that acidification of blood cose concentrations after a glucose challenge are substantially
with citrate buffer inhibits in vitro glycolysis far more higher in older individuals (72, 73). The evidence for an asso-
effectively than fluoride (62). The mean glucose con- ciation between increasing insulin resistance and age is incon-
centration in samples stored at 37 C decreased by only sistent (74). Aging appears to influence glucose homeostasis,
0.3% at 2 h and 1.2% at 24 h when blood was drawn into tubes and visceral obesity seems to be responsible for the reported
containing citrate buffer, sodium fluoride, and EDTA. The continuous decrease in glucose tolerance that begins in middle
use of these blood-collection tubes, where they are available, age (75).
appears to offer a practical solution to the glycolysis problem.
Glucose can be measured in whole blood, serum, or Recommendation
plasma, but plasma is recommended for diagnosis [note
that although both the ADA and WHO recommend venous On the basis of biological variation, glucose measure-
plasma, the WHO also accepts measurement of glucose in ment should have an analytical imprecision 2.9%, A bias
capillary blood (19, 21)]. The molality of glucose (i.e., the 2.2%, and a total error 6.9%. To avoid misclassification
amount of glucose per unit water mass) in whole blood is of patients, the goal for glucose analysis should be to mini-
identical to that in plasma. Although erythrocytes are essen- mize total analytical error, and methods should be without
tially freely permeable to glucose (glucose is taken up by measurable bias.
facilitated transport), the concentration of water (in kilo- B (low)
grams per liter) in plasma is approximately 11% higher
than in whole blood. Therefore, glucose concentrations are
approximately 11% higher in plasma than in whole blood if B. Analytical. Glucose is measured almost exclusively by enzy-
the hematocrit is normal. Glucose concentrations in heparin- matic methods. An analysis of proficiency surveys conducted
ized plasma were reported in 1974 to be 5% lower than in by the College of American Pathologists (CAP) reveals that
serum (63). The reasons for the difference are not apparent hexokinase or glucose oxidase is used in virtually all analyses
but have been attributed to the shift in fluid from erythrocytes performed in the US (70). A very few laboratories (,1%) use
to plasma caused by anticoagulants. In contrast, some more glucose dehydrogenase. Enzymatic methods for glucose analysis
recent studies found that glucose concentrations are slightly are relatively well standardized. At a plasma glucose concentra-
higher in plasma than in serum. The observed differences tion of approximately 7.5 mmol/L (135 mg/dL), the impreci-
were approximately 0.2 mmol/L (3.6 mg/dL) (64), or approx- sion (CV) among laboratories that used the same method was
imately 2% (65), or 0.9% (62). Other studies have found that 2.6% (70). Similar findings have been reported for glucose
glucose values measured in serum and plasma are essentially analyses of samples from patients. The method of glucose mea-
the same (66, 67). Given these findings, it is unlikely that surement does not influence the result. A comparison of results
values for plasma and serum glucose will be substantially from approximately 6000 clinical laboratories reveals that the
different when glucose is assayed with current instruments, mean glucose concentrations measured in serum samples by
and any differences will be small compared with the day-to- the hexokinase and glucose oxidase methods are essentially the
day biological variation of glucose. Clinical organizations do same (76). Compared with a reference measurement procedure,
not recommend the measurement of glucose in serum (rather significant bias (P ,0.001) was observed for 40.6% of the peer
than plasma) for the diagnosis of diabetes (19, 21). Use of groups (76). If similar biases occur with plasma, patients near
plasma allows samples to be centrifuged promptly to prevent the diagnostic threshold could be misclassified.
glycolysis without waiting for the blood to clot. The glucose No consensus has been achieved on the goals for glucose
concentrations in capillary blood obtained during an OGTT analysis. Numerous criteria have been proposed to establish
Glucose 7

analytical goals. These criteria include expert opinion (con- the 95% CI is approximately 12.88%. Thus, the 95% CI for
sensus conferences), the opinion of clinicians, regulation, the a fasting glucose concentration of 7.0 mmol/L (126 mg/dL)
state of the art, and biological variation (77). A rational and would be 7.0 mmol/L 6.4% (126 mg/dL 6.4%), i.e., 6.1
realistic recommendation that has received some support is to 7.9 mmol/L (110142 mg/dL). Use of an assay CV of 3%
use biological criteria as the basis for analytical goals. It has only (excluding biological variation) would yield a 95% CI of
been suggested that imprecision should not exceed one-half of 6.67.4 mmol/L (118134 mg/dL) among laboratories, for a
the within-individual biological CV (78, 79). For plasma glu- true glucose concentration of 7.0 mmol/L (126 mg/dL). Per-
cose, a CV 2.2% has been suggested as a target for impreci- forming the same calculations at the cutoff for impaired fasting
sion, with a 0% bias (79). Although this recommendation was glucose yields a 95% CI of 5.6 mmol/L 6.4% (100 mg/dL
proposed for within-laboratory error, it would be desirable to 6.4%), i.e., 4.96.3 mmol/L (87113 mg/dL). One should
achieve this goal for interlaboratory imprecision to minimize bear in mind that these intervals include 95% of the results and
differences among laboratories in the diagnosis of diabetes in that the remaining 5% will be outside this interval. Thus, the
individuals with glucose concentrations close to the threshold biological variation is substantially greater than the analytical
value. Therefore, the goal for glucose analysis should be to variation. Using biological variation as the basis for deriving
minimize total analytical error, and methods should be without analytical performance characteristics (77), Westgard proposed
measurable bias. A national or international program that uses the following desirable specifications for glucose (86): analyti-
commutable samples (e.g., fresh frozen plasma) to eliminate cal imprecision, 2.9%; bias, 2.2%; and total error, 6.9%.
matrix effects and has accuracy-based grading with values
derived with a reference measurement procedure should be A. Turnaround time. A short turnaround time for glucose analy-
developed to assist in achieving this objective. sis is not usually necessary for diagnosis of diabetes. In some
clinical situations, such as acute hyper- or hypoglycemic epi-
sodes in the emergency department or treatment of diabetic
ketoacidosis (DKA), rapid analysis is desirable. A turnaround
4. INTERPRETATION time of 30 min has been proposed (87). This value is based on
the suggestions of clinicians, however, and no outcome data
Despite the low analytical imprecision at the diagnostic deci- that validate this time interval have been published. Inpatient
sion limits of 7.0 mmol/L (126 mg/dL) and 11.1 mmol/L (200 management of diabetic patients on occasion may require a
mg/dL), classification errors may occur. Knowledge of intra- rapid turnaround time (minutes, not hours). Similarly, for pro-
individual (within-person) variation in FPG concentrations tocols with intensive glucose control in critically ill patients
is essential for meaningful interpretation of patient values (88), rapid glucose results are required in order to calculate
(although total biological variation includes within-person the insulin dose. Bedside monitoring with glucose meters (see
and between-person variation, most discussions focus on the below) has been adopted by many as a practical solution.
within-person variation). An early study, which repeated the
OGTT in 31 nondiabetic adults at a 48-h interval, revealed that B. Frequency of measurement. The frequency of measure-
the FPG concentration varied between the 2 values by ,10% ment of plasma glucose is dictated by the clinical situation.
in 22 participants (77%) and by ,20% in 30 participants (97%) The ADA, WHO, and IDF recommend that an increased FPG
(80). A careful evaluation of healthy individuals over several or an abnormal OGTT result must be confirmed to establish
consecutive days revealed that the biological variation in FPG the diagnosis of diabetes (19, 89). Screening by FPG is recom-
[mean glucose, 4.9 mmol/L (88 mg/dL)] exhibited within- mended every 3 years, beginning at 45 years of age and more
and between-individual CVs of 4.8%6.1% and 7.5%7.8%, frequently in high-risk individuals; however, the frequency of
respectively (8183). Larger studies have revealed intraindi- analysis has not been specified for the latter group. Monitoring
vidual CVs of 4.8% and 7.1% for FPG in 246 healthy individu- is performed by patients who measure their glucose themselves
als and 80 previously undiagnosed individuals with diabetes, with meters and by assessment of Hb A1c in an accredited labo-
respectively (83). Similar findings were obtained from an ratory (see below). The appropriate interval between glucose
analysis of 685 adults from NHANES III, in which the mean measurements in acute clinical situations (e.g., patients admit-
within-person variation in FPG measured 24 weeks apart ted to a hospital, patients with DKA, neonatal hypoglycemia,
was 5.7% (95% CI, 5.3%6.1%) (84). An analysis of larger and so forth) is highly variable and may range from 30 min to
numbers of individuals from the same NHANES III database 24 h or more.
yielded within- and between-person CVs of 8.3% and 12.5%,
respectively, at a glucose concentration of approximately 5.1
mmol/L (92 mg/dL) (85). If a within-person biological CV of
5.7% is applied to a true glucose concentration of 7.0 mmol/L 5. EMERGING CONSIDERATIONS
(126 mg/dL), the 95% CI would encompass glucose concen-
trations of 6.27.8 mmol/L (112140 mg/dL). If the analyti- Continuous minimally invasive and noninvasive analysis of
cal CV of the glucose assay (approximately 3%) is included, glucose is addressed below.
8
Chapter 3
Glucose Meters

Portable meters for the measurement of blood glucose concentra- with portable metersalthough attractive because of conve-
tions are used in 3 major settings: (a) in acute- and chronic-care nience, ease, and accessibilitywould generate many false
facilities, including intensive care units (ICUs); (b) in physicians positives and false negatives.
offices; and (c) by patients at home, work, and school. Measure-
ment in the last setting, self-monitoring of blood glucose (SMBG),
Recommendation
was performed at least once per day by 40% and 26% of individu-
als with type 1 and type 2 diabetes, respectively, in the US in 1993 SMBG is recommended for all insulin-treated patients with
(90). The overall rate of daily SMBG among adults with diabe- diabetes.
tes in the US increased to 40.6% in 1997 and to 63.4% in 2006 A (high)
(91). The ADA summarized the uses of SMBG as early as 1987
[see (92) and references therein] and currently recommends that
SMBG be carried out 3 times daily by patients who use multiple Recommendation
insulin injections or insulin pump therapy (92, 93). It is recom-
mended that most individuals with diabetes attempt to achieve and In patients with type 2 diabetes treated with diet and oral
maintain blood glucose concentrations as close to those in nondia- agents, SMBG may help achieve better control, particularly
betic individuals as is safely possible. when therapy is initiated or changed. Data are insufficient,
however, to claim an associated improvement of health out-
comes. The role of SMBG in patients with stable type 2 dia-
betes controlled by diet alone is not known.
1. USE C (high)

Recommendation B. Monitoring/prognosis. SMBG is recommended for all

insulin-treated patients with diabetes. Intensive glycemic con-
There are insufficient published data outcome to support a
trol can decrease microvascular complications in individuals
role for portable meters and skin-prick (finger-stick) blood
with type 1 (44) or type 2 (46) diabetes. In the DCCT, patients
samples in the diagnosis of diabetes or for population
with type 1 diabetes achieved intensive glycemic control by
screening.
performing SMBG at least 4 times per day (44). Therapy in
C (moderate)
patients with type 2 diabetes in the UKPDS (46) was adjusted
according to FPG concentration; SMBG was not evaluated.
The role of SMBG in individuals with type 2 diabetes has
Recommendation generated considerable controversy (94, 95). Faas et al. (96)
reviewed 11 studies published between 1976 and 1996 that
The imprecision of the results, coupled with the substan- evaluated SMBG in patients with type 2 diabetes. Only one of
tial differences among meters, precludes the use of glucose the published studies reported that SMBG produced a significant
meters from the diagnosis of diabetes and limits their useful- improvement in glycated Hb (GHb). The reviews authors con-
ness in screening for diabetes. cluded that the efficacy of SMBG in type 2 diabetes is question-
A (moderate) able (96). Similar conclusions were drawn in an early (2000)
metaanalysis (97) of a sample of patients with type 2 diabetes
A. Diagnosis/screening. The glucose-based criteria for the in the NHANES (98) and the Freemantle Diabetes Study (99).
diagnosis of diabetes are based on outcome data (the risk of Two early randomized trials assessed the use of glucose meters
micro- and macrovascular disease) correlated with plasma in individuals with type 2 diabetes (100, 101). One of these tri-
glucose concentrationsboth fasting and 2 h after a glucose als (100) had statistical power to detect a 0.5% reduction in Hb
loadassayed in an accredited laboratory (1). Whole blood is A1c but reported only a modest decrease (0.3%) in Hb A1c among
used in portable meters. Although most portable meters have poorly controlled patients treated with oral agents. The second
been programmed to report a plasma glucose concentration, study (101) failed to demonstrate a significant difference in Hb
the imprecision of the current meters (see below) precludes A1c in patients who were assigned to use meters, compared with
their use from the diagnosis of diabetes. Similarly, screening those who were not.

9
10 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

For individuals with type 2 diabetes, cross-sectional and diabetes, as an aid in determining appropriate insulin doses at
longitudinal observational studies in several countries have different times of the day (92). Patients adjust the amount of
failed to demonstrate an improvement in glycemic control (as insulin according to their plasma or blood glucose concentra-
measured by mean Hb A1c concentration) associated with the tion. Frequent SMBG is particularly important for tight glyce-
use of SMBG (102 104). This lack of effect was seen in indi- mic control in type 1 diabetes.
viduals treated with insulin, oral agents, or both. Frequency of Hypoglycemia is a major, potentially life-threatening com-
meter use did not predict Hb A1c. plication of the treatment of diabetes. The risk of hypoglyce-
A 2005 Cochrane review (105, 106) of self-monitoring in mia is seen primarily in patients treated with insulin or insulin
individuals with type 2 diabetes not using insulin concluded that secretagogues, and it increases substantially when pharma-
SMBG might be effective in improving glucose control. There cologic therapy is directed towards maintaining the glycemic
was insufficient evidence to evaluate whether it was beneficial concentrations as close to those found in nondiabetic indi-
in improving quality of life, improving well-being or patient sat- viduals as is safely possible (44, 46). The incidence of major
isfaction, or decreasing the number of hypoglycemic episodes. hypoglycemic episodesrequiring third-party help or medical
The randomized controlled Diabetes Glycaemic Educa- interventionwas 2- to 3-fold higher in the intensive-treat-
tion and Monitoring (DiGEM) trial (107) studied people with ment group than in the conventional group in clinical trials of
type 2 diabetes, a third of whom were treated with diet alone. patients with type 1 and type 2 diabetes (44, 46). Furthermore,
In 2007, the investigators reported, Evidence is not convinc- many patients with diabetes, particularly those with type 1, lose
ing of an effect of self monitoring blood glucose ... in improv- the autonomic warning symptoms that normally precede neu-
ing glycaemic control [as assessed by Hb A1c] compared with roglycopenia (hypoglycemic unawareness) (113), increasing
usual care in reasonably well controlled non-insulin treated the risk of hypoglycemia. SMBG can be useful for detecting
patients with type 2 diabetes. A cost-effectiveness analysis asymptomatic hypoglycemia and allowing patients to avoid
of data from the DiGEM trial concluded, Self monitoring major hypoglycemic episodes.
of blood glucose with or without additional training in incor-
porating the results into self care was associated with higher
costs and lower quality of life in patients with non-insulin
treated type 2 diabetes. In light of this, and no clinically sig- 3. ANALYTICAL CONSIDERATIONS
nificant differences in other outcomes, self monitoring of
blood glucose is unlikely to be cost effective in addition to
standardised usual care (108). Recommendation
The later ESMON study (109), a randomized controlled
trial of SMBG in newly diagnosed people with diabetes not Patients should be instructed in the correct use of glucose
treated with insulin, found no benefit of SMBG on glycemic meters, including quality control. Comparison between
control but did find higher scores on a depression subscale. SMBG and concurrent laboratory glucose analysis should be
Two recent systematic reviews of randomized controlled performed at regular intervals to evaluate the performance of
studies of SMBG in people with type 2 diabetes not treated with the meters in the patients hands.
insulin reported small but significantly greater decreases in Hb B (moderate)
A1c among patients using SMBG than in controls (110, 111).
In the first review (110), SMBG was associated with a larger
A. Preanalytical. Numerous factors can interfere with
reduction in Hb A1c compared with non-SMBG (weighted mean
glucose analysis with portable meters. Several of these fac-
difference, 20.31%; 95% CI, 20.44 to 20.17). In the second tors, such as improper application, timing, and removal of
study (111), the relative decrease in Hb A1c was 20.24% (95% excess blood (61), have been mitigated or eliminated by
CI, 20.34% to 20.14%). The effect of SMBG was limited to advances in technology. Important variables that may influ-
patients with Hb A1c values 8% (64 mmol/mol). ence the results of bedside glucose monitoring include
A 2009 review of studies of patients with type 2 diabetes changes in hematocrit (114), altitude, environmental temper-
(112) addressed recent large randomized trials of tight glyce- ature or humidity, hypo-tension, hypoxia and high triglyc-
mic control, a major rationale for SMBG use in these patients.
eride concentrations (115), and various drugs. Furthermore,
It concluded that tight glycemic control burdens patients with
most meters are inaccurate at very high or very low glu-
complex treatment programs, hypoglycemia, weight gain, and
cose concentrations. Another important factor is variation
costs and offers uncertain benefits in return, thus raising addi-
in results among different glucose meters. Different assay
tional uncertainty about the use of SMBG in people with type
methods and architectures lead to a lack of correlation among
2 diabetes.
meters, even from a single manufacturer. In fact, 2 meters of
the same brand have been observed to differ substantially
in accuracy (116, 117). Patient factors are also important,
2. RATIONALE particularly adequate training. Recurrent education at clinic
visits and comparison of SMBG with concurrent laboratory
Knowledge of ambient plasma or blood glucose concentrations is glucose analysis improved the accuracy of patients blood
used by insulin-requiring patients, particularly those with type 1 glucose readings (118). Thus, it is important to evaluate the
Glucose Meters 11

patients technique at regular intervals (21). In addition to tial performance as judged by performance in the hands of
these technical issues, the anatomic site where skin-puncture skilled medical technologists (124).
samples are obtained influences results. Testing blood from Numerous analytical goals have been proposed for the
so-called alternative sites may introduce a temporal lag in performance of glucose meters. The rationale for these goals
changes in measured blood glucose. is not always clear. In 1987, the ADA recommended a goal
of total error (user plus analytical) of ,10% at glucose con-
centrations of 1.722.2 mmol/L (30400 mg/dL) 100% of the
Recommendation
time (125). In addition, the ADA proposed that values should
Multiple performance goals for portable glucose meters have differ by 15% from those obtained by a laboratory refer-
been proposed. These targets vary widely and are highly con- ence method. The recommendation was modified in response
troversial. Manufacturers should work to improve the impre- to the significant reduction in complications obtained by tight
cision of current meters, with an intermediate goal of limiting glucose control in the DCCT. A revised performance goal,
total error for 95% of samples to 15% at glucose concen- published in 1996 (92), was for a total analytical error of
trations 5.6 mmol/l (100 mg/dl) and to ,0.8 mmol/l (15 ,5%. To our knowledge, there are no published studies of
mg/dl) at glucose concentrations ,5.6 mmol/l (100 mg/dl). diabetes patients achieving the goal of an analytical error of
Lower total error would be desirable and may prove neces- ,5% with any glucose meters.
sary in tight glucose-control protocols and for avoiding hypo- The less stringent CLSI (formerly NCCLS) recommenda-
glycemia in all settings. tions are that, for 95% of the samples, the difference between
C (low) meter and laboratory measurements of glucose be (a) ,20%
when the laboratory glucose value is .5.5 mmol/L (100 mg/
dL) and (b) ,0.83 mmol/L (15 mg/dL) of the laboratory glu-
cose value when the glucose concentration is 5.5 mmol/L
Recommendation (100 mg/dL) (126). The 2003 International Organization for
Meters should measure and report plasma glucose concen- Standardization (ISO) recommendations (127) propose that
trations to facilitate comparison with assays performed in for test readings .4.2 mmol/L (75 mg/ dL), the discrepancy
accredited laboratories. between meters and an accredited laboratory should be ,20%;
GPP for glucose readings 4.2 mmol/L (75 mg/dL), the discrep-
ancy should not exceed 0.83 mmol/L (15 mg/dL) in 95% of
the samples. In both the CLSI and ISO guidelines, 5% of these
B. Analytical. Virtually all glucose meters use strips that con- results can be substantially outside these limits. At the time of
tain enzymes, such as glucose oxidase or glucose dehydroge- writing, both the CLSI and ISO recommendations were under-
nase. A drop of whole blood is applied to a strip that contains going revision.
all the reagents necessary for the assay. Some meters have a These criteria serve as de facto minimal quality require-
porous membrane that separates erythrocytes, and analysis is ments for manufacturers wishing to sell meters. With these cri-
performed on the resultant plasma. Meters can be calibrated to teria, a concentration of 2.5 mmol/L (45 mg/dL) may be read
report plasma glucose values, even when the sample is whole as 1.7 mmol/L (30 mg/dL) or 3.3 mmol/L (60 mg/dL) and be
blood. An IFCC working group recommended that glucose considered acceptable. Such errors do not appear to be accept-
meters report the plasma glucose concentration, irrespective of able for reliably detecting hypoglycemia. Similarly, errors of
the sample type or technology (119, 120). This approach can 20% can lead to errors in insulin dosing, which, when com-
improve harmonization and allow comparison with laboratory- bined with other factors, can lead to hypoglycemia.
generated results (121). The meters use reflectance photometry Others have proposed different approaches to establish-
or electrochemistry to measure the rate of the reaction or the ing quality requirements. Clarke et al. (128) developed an
final concentration of the products, and they provide digital error grid that attempts to define clinically important errors
readouts of glucose concentration. Manufacturers claim report- by identifying fairly broad target ranges. In another approach,
able concentration ranges as large as 33.3 mmol/L (600 mg/ 201 patients with longstanding type 1 diabetes were ques-
dL), e.g., 033.3 mmol/L (0600 mg/dL). tioned to estimate quality expectations for glucose meters
Several important technological advances decrease (129). On the basis of patients perceptions of their needs
operator error. These improvements include automatic com- and their reported actions in response to changes in measured
mencement of timing when both the sample and the strip glucose concentrations, a goal for analytical quality at hypo-
are in the meter, smaller sample-volume requirements, an glycemic concentrations was a CV of 3.1%. With hypoglyce-
error signal if the sample volume is inadequate, lock out mia excluded, the analytical CV to meet the expectations of
if controls are not assayed, and bar code readers to identify 75% of the patients was 6.4% to 9.7%. The authors recom-
the lot of the strips. Moreover, meters store up to several mended an analytical CV of 5% with a bias 5% (129). A
hundred results that can subsequently be downloaded for third approach used simulation modeling of errors in insulin
analysis. Together, these improvements have improved the dose (130). The results revealed that meters that achieve both
performance of new meters (122, 123 ). Nonetheless, meter a CV and a bias ,5% rarely lead to major errors in insulin
performance in the hands of patients does not equal poten- dose. To provide the intended insulin dosage 95% of the time,
12 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

however, the bias and CV needed to be ,1%2%, depending by finger stick. The integrity of results obtained with finger-
on the dosing schedule for insulin and the intervals of glucose stick samples can be compromised by such factors as shock,
concentrations for the individual patient (130). No meters have hypoxia, and low hematocrit, which are common in these
been shown to achieve CVs of 1%2% in routine use in the settings (138). Moreover, the error of glucose meters may
hands of patients. compound the problem and compromise the ability to control
The lack of consensus on quality goals for glucose meters blood glucose and avoid hypoglycemia. Simulation model-
reflects the absence of agreed objective criteria. With the same ing studies have demonstrated that errors in glucose measure-
biological-variation criteria described above for glucose anal- ment (which include errors related to sample type and sample
ysis in accredited laboratories (section 4, Interpretation), a collection) lead to marked degradation of glycemic control in
biological goal would be a total error 6.9% with an impreci- tight glucose-control protocols (139). In this study, frequen-
sion (as the CV of measurements over several days or weeks) cies of both hyperglycemia and hypoglycemia were increased
2.9% and a bias 2.2% (86). Additional studies, however, with increasing assay imprecision. In a 2005 study of ICU
are necessary to define a goal that is related to medical needs. patients (140), the agreement of meter results with accredited
Current meters exhibit performance superior to prior gen- laboratory results was poor: Among 767 paired results, the
erations of meters (122, 123). A variety of studies of newer 95% limits of agreement were 12.4 to 21.5 mmol/L (143.1
analyzers have documented CVs of about 2% in the hands of to 227.2 mg/dL). Hoedemaekers et al. (141), in a study of
trained workers. Nonetheless, there is room for improvement. 197 arterial blood samples from ICU patients, reported that
In a study conducted under carefully controlled conditions the evaluated meter did not meet the ISO total-error criteria.
in which a single medical technologist performed all of the They also demonstrated that the total error of meters used in
assays, about 50% of the analyses met the 1996 ADA crite- ICU patients was greater than in non-ICU patients. A later
rion of ,5% deviation from reference intervals (122). Another report, which also studied arterial blood from ICU patients,
study that evaluated meter performance in 226 hospitals with measured glucose in 239 samples by a portable meter and by
split samples analyzed simultaneously on meters and labora- a laboratory method and found that the meter results did not
tory glucose analyzers revealed that 45.6%, 25%, and 14% of meet the CLSI/ISO criteria (142). Similarly, a 2005 study of
the split samples differed from each other by .10%, .15%, arterial, venous, and capillary samples from a mixed medical/
and .20%, respectively (131). In another study, none of the surgical ICU of a tertiary care hospital in Canada found that
meters met the 1996 ADA criterion (132). In an evaluation meters did not meet proposed CLSI goals but that a blood gas
in which all testing was performed by trained study staff in analyzer did (143).
an inpatient Clinical Research Center setting, only 81% of
results with a meter that used a hexokinase method were within
10% of results obtained from an accredited laboratory (133). Recommendation
We are aware of no studies that document patient-generated Studies are needed to determine the analytical goals (quality
results that meet the 1996 ADA criteria. Moreover, an analy- specifications) for glucose meters in SMBG and in ICUs.
sis of published studies of glucose meters demonstrated that C (moderate)
the studies suffered from deficiencies in study design, meth-
odology, and reporting (134), raising the possibility that the
reported total error underestimates the true total error of the
meters. A standardized method for evaluating meters has been Recommendations
developed in Norway (134), and the Norwegian health authori-
For future research: important end points in studies of SMBG
ties have decided that all SMBG instruments marketed in Nor-
should include, at a minimum, Hb A1c and frequency of hypo-
way should be examined by a similar procedure (135). Results
glycemic episodes to ascertain whether improved meters
of evaluations of 9 brands of meters according to this method
enable patients to achieve better glucose control. For studies
showed that 3 of 9 meters did not meet the ISO criteria, and
of meter use in intensive or critical care, important end points
none met the 1996 ADA criteria in the hands of patients (135).
include mean blood glucose, frequency of hypoglycemia, and
Glucose meters are also used to support tight control of
variation of glucose control. Ideally, outcomes (e.g., long-
glucose in patients in ICU settings. A 2001 report of a seminal
term complications) should also be examined.
randomized controlled trial by van den Berghe et al. described
GPP
a 34% reduction in mortality in surgical ICU patients man-
aged according to a tight glucose-control protocol (88). A
metaanalysis of multiple randomized controlled trials of tight
glucose control conducted 7 years later failed to identify any 4. INTERPRETATION
improved outcomes but did find an increased incidence of
hypoglycemia (136). A Clinical Chemistry Perspective arti- A. Frequency of measurement. SMBG should be performed at
cle (137) pointed out that the study of van den Berghe et al. least 3 times per day in patients with type 1 diabetes. Moni-
used a precise and accurate glucose analyzer and collected toring less frequently than 3 times per day leads to deterio-
arterial blood samples, whereas subsequent studies often ration in glycemic control (92, 144, 145). Patients perform
used glucose meters and capillary blood samples obtained self-monitoring much less frequently than recommended. Data
Glucose Meters 13

from NHANES III collected between 1988 and 1994 reveal that with GDM according to the results of post-prandial, rather than
SMBG was performed at least once a day by 39% of patients preprandial, plasma glucose concentrations improved glyce-
taking insulin and by 5%6% of patients treated with oral mic control and reduced the risk of neonatal complications
agents or diet alone (98). Moreover, 29% and 65% of patients (146). The optimal frequency of SMBG for patients with type
treated with insulin and oral agents, respectively, monitored 2 diabetes is unknown.
their blood glucose less than once per month; however, no The ADA recommends that patients treated with multiple
evaluation has been performed to verify that 3 times per day is daily injections of insulin perform SMBG 3 times per day (21)
ideal or whether a different frequency would improve glycemic and states that SMBG is useful in achieving glycemic goals
control. For example, adjustment of insulin therapy in women in other patients. The last statement is based on expert opinion.
14
Chapter 4
Continuous Minimally Invasive Glucose Analyses

1. USE but rather one the patient wears for 3 days and then returns to
the providers office for its data to be downloaded for trend
analyses. More recently, a number of real-time devices that
Recommendation allow patients to read both current glucose concentrations
and trends have become commercially available. In the US,
Real-time continuous glucose monitoring (CGM) in conjunc- these devices include the Guardian Real-Time (Medtronic
tion with intensive insulin regimens can be a useful tool to Diabetes), the Seven Plus System (DexCom), and the Free-
lower Hb A1c in selected adults (age >25 years) with type 1 style Navigator (Abbott Laboratories). CGM devices require
diabetes. calibration and confirmation of accuracy with conventional
A (high) SMBG, and the FDA advises using the latter for treatment
decisions, such as calculating premeal insulin doses.
The clinical studies of these devices, generally in highly
Recommendation selected populations, had primarily been limited to assess-
ments of their accuracy or to short-term trials demonstrating
Although the evidence for lowering Hb A1c is not as strong
reductions in the time patients spend within hypo- and hyper-
for children, teens, and younger adults, real-time CGM may
glycemic intervals (148). A systematic review of trials of the
be helpful in these groups. Success correlates with adherence
nonreal-time CGM system device suggests that it does not
to ongoing use of the device.
lead to significantly lower Hb A1c values compared with SMBG
B (moderate)
(149). In 2008, a large 26-week randomized trial of 322 type
1 diabetes patients showed that adults 25 years of age who
used intensive insulin therapy and real-time CGM experienced
Recommendation a 0.5% reduction in Hb A1c, from approximately 7.6% to 7.1%
Real-time CGM may be a supplemental tool to smbg in (approximately 60 to 54 mmol/mol), compared with the usual
individuals with hypoglycemia unawareness and/or frequent intensive insulin therapy with SMBG (150). Sensor use in
episodes of hypoglycemia. children, teens, and adults to 24 years of age did not lower
B (low) Hb A1c significantly, and there was no significant difference in
hypoglycemia for any group. The greatest predictor of Hb A1c
reduction in this study among all age groups was frequency of
sensor use, which was lower in younger-age groups. Although
Recommendation CGM is an evolving technology, the emerging data suggest
Patients require extensive training in using the device. Avail- that it may offer benefit in appropriately selected patients who
able devices must be calibrated with SMBG readings, and the are motivated to wear it most of the time. CGM may be par-
latter are recommended for making treatment changes. ticularly useful for patients with hypoglycemia unawareness
GPP and/or frequent episodes of hypoglycemia; studies in this area
are ongoing.
The development of a device for continuous in vivo moni-
toring of glucose concentrations in blood has become a very
high priority as patients are required to control their plasma 2. RATIONALE
glucose more closely (21, 44, 147). The first device approved
by the US Food and Drug Administration (FDA) for mini- The first goal for developing a reliable in vivo continuous
mally invasive interstitial fluid glucose sensing, the transcu- glucose sensor is to detect unsuspected hypoglycemia. The
taneous GlucoWatch Biographer, is no longer on the market. importance of this goal has been increasingly appreciated
Several implanted-catheter systems have subsequently been with the recognition that strict glucose control is accompanied
approved. The initial device in the latter category is the Con- by a marked increase in the risk of hypoglycemia (44, 147).
tinuous Glucose Monitoring System (CGMS) (Medtronic), Therefore, a sensor designed to detect severe hypoglycemia
a system that does not provide real-time data to the patient, alone would be of value. In contrast, a full-range, reliable

15
16 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

continuous in vivo glucose monitor is a prerequisite for the glucose readings was 15.0% (12.2%) for 735 paired samples,
development of a closed-loop pump or artificial pancreas whereas the GlucoDay microdialysis device (Menarini) had a
that would measure blood glucose concentrations and auto- mean absolute difference of 13.6% (10.2%) for 1156 paired
matically adjust insulin administration. samples (158). For both devices, accuracy was lowest in the
hypoglycemic ranges. Approximately 97% of the values for
both devices were within zones A and B of a Clarke error
3. ANALYTICAL CONSIDERATIONS grid, with none falling in zone E (158). A study of 91 insulin-
treated patients using the DexCom device showed that 95% of
The methods to sample biological fluids in a continuous and 6767 paired glucose values fell within Clarke error grid zones
minimally invasive way vary among test systems. The under- A and B, with a mean absolute difference of 21.2% (148).
lying fundamental concept is that the concentration of glu- Currently, there are no analytical goals for noninvasive
cose in the interstitial fluid correlates with blood glucose. The and minimally invasive glucose analyses. Such standards will
implanted sensors use multiple detection systems, including clearly need to be different for different proposed uses. For
enzyme- (usually glucose oxidase), electrode-, and fluores- example, the reliability, precision, and accuracy requirements
cence-based techniques. Alternatives to enzymes, including for a glucose sensor that is linked to a system that automatically
artificial glucose receptors, as glucose-recognition mol- adjusts insulin doses will be much more stringent than those
ecules are being developed (151, 152). Fluorescence tech- for a sensor designed to trigger an alarm in cases of apparent
nologies include the use of engineered molecules that exhibit extreme hyper- or hypoglycemia. It seems intuitively obvious
altered fluorescence intensity or spectral characteristics on that a larger imprecision can be tolerated in instruments that
binding glucose, or the use of competitive-binding assays make frequent readings during each hour than in an instrument
that use 2 fluorescent molecules in the fluorescent resonance used only 2 or 3 times per day to adjust a major portion of a
energy transfer technique (153157). persons daily insulin dose.

4. INTERPRETATION 5. EMERGING CONSIDERATIONS

The subcutaneous sensors are generally worn for a number With FDA approval of several self-monitoring continuous glu-
of days and require calibration with SMBG readings several cose sensors, it is anticipated that there will be renewed efforts
times per day. A few small studies have examined their accu- to bring other technologies forward into clinical studies. Ulti-
racy compared with SMBG and/or plasma glucose assays. mately, we shall see improved methods for noninvasive or min-
For the Medtronic CGMS System Gold device, the mean imally invasive glucose measurements that will complement
(SD) absolute difference between sensor readings and blood current glucose self-monitoring techniques.
Chapter 5
Noninvasive Glucose Analysis

1. USE 2. RATIONALE

Indirect and direct methods are being developed for noninva-

Recommendation sive glucose sensing. Indirect methods rely on the effect of in
vivo glucose concentrations on a measurable parameter. The
No noninvasive sensing technology is currently approved for
classic example of this approach is the effect of blood glu-
clinical glucose measurements of any kind. Major techno-
cose concentrations on the scattering properties of skin (163).
logical hurdles must be overcome before noninvasive sens-
Changes in blood glucose substantially affect the difference in
ing technology will be sufficiently reliable to replace existing
refractive index between skin cells and the surrounding inter-
portable meters, implantable biosensors, or minimally
stitial fluid and thereby alter the scattering coefficient of skin.
invasive technologies.
This parameter can be measured in a number of ways, including
C (very low)
ocular coherent tomography. Skin impedance and the aggrega-
tion properties of erythrocytes are other indirect approaches.
Noninvasive glucose-sensing technologies represent a group Direct methods measure a property of the glucose mol-
of potential analytical methods for measuring blood glucose ecule itself. Vibrational spectroscopy is the primary direct
concentrations without implanting a probe or collecting a method and generally involves mid-infrared, near-infrared,
sample of any type. The most commonly explored methods photoacoustic, or Raman scattering spectroscopy. The basis of
involve passing a selected band of nonionizing electromag- these measurements is the unique spectral signature of glucose
netic radiation (light) through a vascular region of the body relative to the background tissue matrix.
and then determining the in vivo glucose concentration from Selectivity is the primary factor that must be addressed for
an analysis of the resulting light or spectrum. The distin- either indirect or direct approaches. The lack of an isolated sample
guishing feature of this approach is a lack of physical con- precludes the use of physical separations or chemical reactions to
tact between the sample matrix and a measurement probe. enhance measurement selectivity. All of the analytical informa-
The only functional interaction is the light passing through tion must originate from the noninvasive signal. Ultimately, the
the sample. success of any approach demands a full understanding of the fun-
A truly noninvasive method would be painless in operation damental basis of selectivity. To this end, basic research efforts are
and capable of continuous readings over time. In addition, non- paramount to establish such a level of understanding.
invasive sensing technology may be less expensive to imple-
ment than existing technologies that demand either a fresh test
strip for each measurement or a new implantable probe that 3. ANALYTICAL CONSIDERATIONS
requires multiple daily calibration measurements with fresh
test strips. Furthermore, most noninvasive strategies offer the It should no longer be acceptable to publish results that simply
potential for measuring multiple analytes from a single nonin- demonstrate the ability to follow glucose transients during simple
vasive measurement. The development of this technology is glucose tolerance tests (164). This ability is well established in
driven by the features of both low cost and painless, continuous the literature for numerous approaches, both indirect and direct. In
operation with no reagents or waste for disposal. fact, it is rather easy to monitor optical changes that correlate with
Reports in the peer-reviewed literature describe noninva- in vivo glucose concentrations during glucose tolerance tests. It
sive measurements based on a variety of techniques, such as is considerably more difficult, however, to demonstrate that such
absorption spectroscopy, photoacoustic spectroscopy, Raman measurements are reliable and selective. Reliability and selectivity
scattering, static light scattering, polarimetry, and optical must be the focus of the next generation of research. Indeed, the
coherent tomography (159162). Potential applications include FDA considers all noninvasive sensing technologies to be high-
discrete home glucose testing, continuous home glucose moni- risk medical devices, and premarket approval documentation will
toring, nocturnal hypoglycemia alarm, measurements in a be required for commercialization in the US (165).
physicians office, point-of-care monitoring, screening for dia- Many reports of attempts to measure glucose noninva-
betes, and control of hyperglycemia in critically ill patients. To sively lack sufficient information to judge the likelihood that
date, none of these applications has been realized. glucose is actually being measured. The interpretation of such

17
18 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

clinical data is complicated by the common use of multivari- Despite the limitations noted above, real progress is
ate statistical methods, such as partial least squares regression being made to further the development of noninvasive glu-
and artificial neural networks. These multivariate methods are cose-sensing technologies (171, 172). Rigorous testing of
prone to spurious correlations that can generate apparently noninvasive technologies must be continued in concert with
functional glucose measurements in the complete absence of efforts to understand the underlying chemical basis of selec-
glucose-specific analytical information (166, 167). Given this tivity. Issues of calibration stability must also be investigated.
known limitation of these multivariate methods, care must be Overall progress demands advances in both instrumentation
used in their implementation. Tests for spurious correlations and methods of data analysis. For each, meaningful bench-
(168170) must be developed and implemented with all future marks must be established to allow rigorous inter- and intral-
clinical data to avoid reports of false success. aboratory comparisons.
Chapter 6
Gestational Diabetes Mellitus

1. USE 2. RATIONALE

The ADA states that because of the risks of GDM to the mother
Recommendation and the neonate, screening and diagnosis are warranted (21).
The screening and diagnostic criteria for GDM have recently
All pregnant women not previously known to have diabetes been modified extensively. The Hyperglycemia and Adverse
should undergo testing for GDM at 2428 weeks of gestation. Pregnancy Outcome (HAPO) study was a large (approximately
A (high) 25 000 pregnant women) prospective, multinational epidemio-
logic study to assess adverse outcomes as a function of maternal
glycemia (176). The study revealed strong, graded, predomi-
GDM has been defined as any degree of glucose intoler-
nantly linear associations between maternal glycemia and pri-
ance with onset or first recognition occurring during pregnancy
mary study outcomes, i.e., birth weight >90th percentile, delivery
(1). After recent discussions, the International Association
by cesarean section, clinical neonatal hypoglycemia, and cord
of Diabetes and Pregnancy Study Groups (IADPSG) recom-
serum insulin (C-peptide) concentrations >90th percentile of
mended that high-risk women who have diabetes established
values in the HAPO study population. The associations remain
according to standard criteria (Table 6) at their initial prena-
strong after adjustments for multiple potentially confounding
tal visit receive a diagnosis of overt, not gestational, diabetes
factors. Strong associations were also found with infant adipos-
(21). The IADPSG recommendations are not identical to the
ity (177), with some secondary outcomes (including risks of
criteria for nonpregnant individuals, in that an OGTT result
shoulder dystocia and/or birth injury), and with preeclampsia
with an FPG value <7.0 mmol/L (126 mg/dL) and 2-h value
(176). On the strength of these results, an expert consensus panel
>11.1 mmol/L (200 mg/dL) is not called overt diabetes. As
appointed by the IADPSG recommended outcome based
the prevalence of obesity and type 2 diabetes has increased, the
criteria for the classification of glucose concentrations in preg-
number of women with undiagnosed diabetes has risen (173).
nancy (178). All pregnant women not previously known to have
Therefore, the ADA now recommends that women with risk
diabetes should be evaluated by a 75-g OGTT for GDM at 2428
factors for type 2 diabetes be screened for diabetes according
weeks of gestation (178). Diagnostic cutpoints for fasting, 1-h,
to standard diagnostic criteria (Table 6) at the first prenatal
and 2-h plasma glucose concentrations have been established
visit (93). Women with diabetes diagnosed with this approach
(Table 8). These recommendations were adopted by the ADA in
should receive a diagnosis of overt diabetes.
2011 (93) and are currently under consideration by the American
Two randomized clinical trials have now demonstrated a
College of Obstetrics and Gynecology in the US and by corre-
benefit from the treatment of mild GDM. Both studies found
sponding groups in other countries. Using the new criteria sub-
that treatment of GDM can reduce both serious adverse outcomes
stantially increases the incidence of GDM, mainly because only
and the frequency of large babies (macrosomia) (174, 175).
1 increased glucose value is required to diagnose GDM (prior
recommendations required 2 increased glucose concentrations).
Table 8. Screening for and diagnosis of GDM. Treatment will require additional resources, and outcome studies
Glucose concentration Percentage will be necessary to ascertain whether therapy is beneficial for
Glucose GDM diagnosed with the new criteria; however, the 2 trials that
threshold, mmol/L .threshold
measure
(mg/dL)a (cumulative)b focused on the treatment of mild GDM (identified with the
FPG 5.1 (92) 8.3% old criteria) achieved an improvement in outcomes, with only
1-h PG 10.0 (180) 14.0% 10%20% of the patients requiring pharmacologic treatment in
2-h PG 8.5 (153) 16.1%c
addition to medical nutritional therapy (174, 175).
a
 ne or more of these values from a 75-g OGTT must be equaled
O
or exceeded for the diagnosis of GDM.
b
Cumulative proportion of HAPO cohort equaling or exceeding 3. ANALYTICAL CONSIDERATIONS
those thresholds.
c
In addition, 1.7% of the participants in the initial cohort were un-
blinded because of an FPG value >5.8 mmol/L (105 mg/dL) or a 2-h These considerations have been addressed earlier in the Glu-
OGTT values >11.1 mmol/L (200 mg/dL), bringing the total to 17.8%. cose sections. Given the strict cutoffs, it is very important that

19
20 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

close attention be paid to stringent sample-handling procedures They were a 1-step approach, in which an OGTT was per-
to minimize glycolysis after phlebotomy. formed initially, or a 2-step approach, in which an adminis-
tered 50-g oral glucose load (regardless of whether the patient
was fasting) was followed by a plasma glucose measurement
4. INTERPRETATION at 1 h. A plasma glucose value 7.8 mmol/L (140 mg/dL)
indicates the need for definitive testing with an OGTT; how-
ever, a consensus was lacking as to whether a 100-g or 75-g
Recommendation
OGTT should be performed and what cutoff values should be
Gdm should be diagnosed by a 75-g ogtt according to the used for diagnosis.
IADPSG criteria derived from the HAPO study. Some GDM cases may represent preexisting, but undiag-
A (moderate) nosed, type 2 diabetes. Therefore, women with GDM should be
screened for diabetes 612 weeks post-partum according to the
OGTT criteria for nonpregnant women (Table 7) (93). In addi-
The ADA previously recommended that a risk assessment tion, because women with GDM are at a considerably increased
(based on age, weight, past history, and so on) be performed risk of developing diabetes later (179), lifelong screening for
and that patients at average or high risk receive a glucose- diabetes should be performed at least every 3 years according
challenge test. Several diagnostic strategies could be used. to standard criteria for nonpregnant women (Table 6) (93).
Chapter 7
Urinary Glucose

1. USE dL)]. This fact alone limits its usefulness for monitoring dia-
betes under modern care recommendations. Semiquantitative
urine glucose tests also cannot distinguish between euglyce-
Recommendation
mia and hypoglycemia. Furthermore, the extent to which the
Semiquantitative urine glucose testing is not recommended kidney concentrates the urine will affect urine glucose con-
for routine care of patients with diabetes mellitus. centrations, and only mean glucose values between voidings
B (low) are reflected. These facts further minimize the value of urine
glucose measurements.
Semiquantitative urine glucose testing, once the hallmark of
diabetes care in the home setting, has now been replaced by
SMBG (see above). Semiquantitative urine glucose monitoring 3. ANALYTICAL CONSIDERATIONS
should be considered only for patients who are unable or refuse
to perform SMBG, because the urine glucose concentration Semiquantitative test-strip methods that use reactions specific
does not accurately reflect the plasma glucose concentration for glucose are recommended. Commercially available strips
(147, 180). Notwithstanding these limitations, urine glucose use the glucose oxidase reaction (181). Test methods that
monitoring is supported by the IDF in those situations in which detect reducing substances are not recommended because they
blood glucose monitoring is not accessible or affordable, par- are subject to numerous interferences, including numerous
ticularly in resource-poor settings (23). drugs and nonglucose sugars. When used, single voided urine
samples are recommended (147).

2. RATIONALE
4. INTERPRETATION
Although urine glucose is detectable in patients with grossly
increased blood glucose concentrations, it provides no infor- Because of the limited use of urine glucose measurements,
mation about blood glucose concentrations below the variable semiquantitative specific reactionbased test strip methods are
renal glucose threshold [approximately 10 mmol/L (180 mg/ adequate.

21
22
Chapter 8
Ketone Testing

1. USE dosis (184186). Thus, assay methods for ketones that do not
include HBA measurement may provide misleading clinical
information by underestimating total ke-tone body concentra-
tion (187).
Recommendation
Ketones measured in urine or blood in the home setting by
patients with diabetes and in the clinic/hospital setting should
be considered only an adjunct to the diagnosis of DKA. 3. ANALYTICAL CONSIDERATIONS
GPP
A. Urine ketone. Preanalytical. The concentrations of ketones in
the urine of healthy individuals are below the detection limits of
The ketone bodies acetoacetate (AcAc), acetone, and commercially available testing materials. False-positive results
-hydroxybutyric acid (HBA) are catabolic products of have been reported with highly colored urine and in the pres-
free fatty acids. Measurements of ketones in urine and ence of several sulfhydryl-containing drugs, including angio-
blood are widely used in the management of patients with tensin-converting enzyme inhibitors (188). Urine test reagents
diabetes as adjuncts for both diagnosis and ongoing moni- deteriorate with exposure to air, giving false-negative readings;
toring of DKA. Measurements of ketone bodies are rou- therefore, testing material should be stored in tightly sealed con-
tinely performed, both in an office/ hospital setting and by tainers and discarded after the expiration date on the manufactur-
patients at home. The ADA recommends that ketosis-prone ers label (189). False-negative readings have also been reported
patients with diabetes check urine or blood ketones in situ- with highly acidic urine samples, such as after large intakes of
ations characterized by deterioration in glycemic control ascorbic acid. Loss of ketones from urine attributable to micro-
in order to detect and preempt the development of DKA bial action can also cause false-negative readings. Because ace-
(21, 182). tone is a highly volatile substance, samples should be kept in a
closed container. For point-of-care analyses in medical facilities
and for patients in the home setting, control materials (that give
2. RATIONALE both negative and positive readings) are not commercially avail-
able but would be desirable to ensure accuracy of test results.
Ketone bodies are usually present in urine and blood, but
in very low concentrations (e.g., total serum ketones, <0.5 Analytical. Several assay principles have been described. Most
mmol/L). Increased ketone concentrations detected in patients commonly used is the colorimetric reaction that occurs between
with known diabetes or in previously undiagnosed patients AcAc and nitroprusside (sodium nitroferricyanide) to produce a
presenting with hyperglycemia suggest impending or estab- purple color (181). This method is widely available in the form
lished DKA, a medical emergency. The 2 major mechanisms of dipsticks and tablets and is used to measure ketones in both the
for high ketone concentrations in patients with diabetes are urine and blood (either serum or plasma). Several manufacturers
increased production from triglycerides and decreased utili- offer dipsticks for measuring glucose and ketones. A combina-
zation in the liver both of which are due to an absolute or tion dipstick is necessary only if the patient monitors urine glu-
relative insulin deficiency and increased counter-regulatory cose instead of or in addition to blood glucose. The nitroprusside
hormones, including cortisol, epinephrine, glucagon, and method measures only AcAc unless the reagent contains glycine,
growth hormone (183). in which case acetone is also measured. The nitroprusside-con-
The principal ketone bodies HBA and AcAc are typi- taining reagent is much more sensitive to AcAc than acetone
cally present in approximately equimolar amounts. Acetone, with respect to color generation. Importantly, this reagent cannot
usually present in only small quantities, is derived from spon- be used to measure HBA (181).
taneous decarboxylation of AcAc. The equilibrium between B. Blood ketones. Preanalytical. Serum/plasma ketones can
AcAc and HBA is shifted towards HBA formation in any be measured with the tablets or dipsticks routinely used for
condition that alters the redox state of hepatic mitochondria urine ketone measurements. Although samples can be diluted
to increase NADH concentrations, such as hypoxia, fasting, with saline to titer the ketone concentration (results are
metabolic disorders (including DKA), and alcoholic ketoaci- typically reported as positive at a 1/x dilution), HBA, the

23
24 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

predominant ketone body in DKA, is not detected, as with not previously diagnosed with diabetes but who presents with
urine ketone testing. typical symptoms of diabetes and hyperglycemia suggests the
For specific HBA measurements, sample requirements possibility of impending or established DKA. Although DKA
differ among methods, as is described below. In general, blood is most commonly associated with type 1 diabetes, it may occur
samples can be collected into tubes containing heparin, EDTA, rarely in type 2 patients (193). Patients with alcoholic ketoaci-
fluoride, citrate, or oxalate. Ascorbic acid interferes with some dosis will have positive urine ketone readings, but hypergly-
assay methods. AcAc interferes with some assay methods unless cemia is not usually present. Positive urine ke-tone readings
the samples are highly dilute. Sample stability differs among are found in up to 30% of first morning urine samples from
methods, but whole-blood samples are generally stable at 4 C pregnant women (with or without diabetes), during starvation,
for up to 24 h. Serum/plasma samples are stable for up to 1 week and after hypoglycemia (187).
at 4 C and for at least several weeks at 220 C (long-term sta-
bility data are not available for most assay methods). Recommendation
Analytical. Although several different assay methods (e.g., Blood ketone determinations that rely on the nitroprusside
colorimetric, gas chromatography, capillary electrophoresis, reaction should be used only as an adjunct to diagnose dka
and enzymatic) have been described for blood ketones, includ- and should not be used to monitor dka treatment. Specific
ing specific measurement of HBA, enzymatic methods appear measurement of hba in blood can be used for diagnosis
to be the most widely used for the quantification of HBA for and monitoring of dka.
routine clinical management (190192). The principle of the B (moderate)
enzymatic methods is that -hydroxybutyrate dehydrogenase
in the presence of NAD+ converts HBA to AcAc and NADH. B. Blood ketone measurements. Blood ketone measurements that
Under alkaline conditions (pH 8.59.5), the reaction favors the rely on the nitroprusside reaction should be used with caution
formation of AcAc from HBA. The NADH produced can be for DKA diagnosis, because the results do not quantify HBA,
quantified spectrophotometrically (usually kinetically) with the predominant ketone in DKA. The test should not be used to
the use of a peroxidase reagent. Most methods permit the use monitor the course of therapy, because AcAc and acetone may
of whole blood, plasma, or serum samples (required volumes increase as HBA decreases during successful therapy (147,
are generally 200 L). Some methods permit the analysis of 183187). Blood ketone measurements that measure HBA
multiple analytes; these methods are designed for point-of-care specifically are useful for both the diagnosis and ongoing moni-
testing. Several methods are available as handheld meters, which toring of DKA (194196). Reference intervals for HBA differ
have been FDA cleared for both laboratory use and home use by among assay methods, but concentrations in healthy individuals
patients. These methods use dry-chemistry test strips to which who have fasted overnight are generally <0.5 mmol/L. Patients
a drop of whole blood, serum, or plasma is added. Results are with well-documented DKA [serum CO2 <17 mmol/L, arterial
displayed on the instruments within approximately 2 min. pH <7.3, plasma glucose >14.9 mmol/L (250 mg/dL)] generally
have HBA concentrations >2 mmol/L.

4. INTERPRETATION
5. EMERGING CONSIDERATIONS
Recommendation Further studies are needed to determine whether blood ketone
Urine ketone measurements should not be used to diagnose measurements by patients with diabetes are preferable (e.g.,
or monitor the course of DKA. better accepted by patients, more prompt diagnosis of DKA) to
GPP urine ketone measurements. Studies are necessary to evaluate
whether the test offers any clinical advantage over more tradi-
A. Urine ketone measurements. The presence of positive urine tional management approaches (e.g., measurements of serum
ketone readings in a patient with known diabetes or a patient CO2, anion gap, or pH).
Chapter 9
Hb A1c

1. USE The ADA and other organizations that have addressed this
issue recommend Hb A1c measurement in both type 1 and type
2 diabetes patients to document the degree of glycemic con-
trol and to assess response to therapy (21, 93, 201). The ADA
Recommendation
has recommended specific treatment goals for Hb A1c on the
Hb A1c should be measured routinely in all patients with dia- basis of results from prospective randomized clinical trials,
betes mellitus to document their degree of glycemic control. most notably the DCCT for type 1 diabetes (44, 197) and the
A (moderate) UKPDS for type 2 diabetes (46). These trials have documented
the relationship between glycemic control (as quantified by
longitudinal Hb A1c measurements) and the risks for the devel-
Measurement of glycated proteins, primarily Hb A1c, is widely opment and progression of chronic complications of diabetes.
used for routine monitoring of long-term glycemic status in Because different GHb assays can produce different GHb val-
patients with diabetes.11 Hb A1c is used both as an index of mean ues, the ADA recommends that laboratories use only assay
glycemia and as a measure of risk for the development of dia- methods that have been certified as traceable to the DCCT
betes complications (147, 197). Hb A1c testing and maintenance GHb reference (21, 187); these results are reported as Hb A1c.
of specified concentrations during pregnancy in patients with The ADA recommends that in general an Hb A1c target of <7%
preexisting type 1 or type 2 diabetes are important for maximiz- (53 mmol/mol) is desirable for nonpregnant adults, with higher
ing the health of the newborn and decreasing perinatal risks for values recommended for children and adolescents (21). Hb A1c
the mother. Specifically, stringent control of Hb A1c values dur- goals should be individualized according to the potential for
ing pregnancy decreases the risk of congenital malformations, benefit with regard to long-term complications and be balanced
large-for-date infants, and the complications of pregnancy and against the increased risk for the hypoglycemia that attends
delivery that can otherwise occur when glycemic control is not intensive therapy. For selected individual patients, more-strin-
carefully managed (198). A recent consensus statement (198) gent targets could be suggested, provided that this goal can be
recommends an Hb A1c value of <6% (42 mmol/mol) in these achieved without substantial hypoglycemia or other adverse
patients if it can be achieved without excessive hypoglycemia. effects of treatment. Such patients might include those with a
Hb A1c is also being used increasingly by quality-assurance pro- short duration of diabetes, a long life expectancy, and no sig-
grams to assess the quality of diabetes care (e.g., requiring that nificant cardiovascular disease (93). Conversely, higher Hb A1c
healthcare providers document the frequency of Hb A1c testing in goals should be chosen for patients with a history of severe
patients with diabetes and the proportion of patients with Hb A1c hypoglycemia, a limited life expectancy, advanced microvas-
values below a specified value) (199, 200). cular or macrovascular complications, or extensive comorbid
conditions. Other clinical organizations recommend similar Hb
11
 he terms glycated hemoglobin, glycohemoglobin, glycosylated
T
A1c targets, which range from 6.5% to 7% (48 to 53 mmol/mol)
(which should not be used), glucosylated hemoglobin, Hb A1, and
Hb A1c have all been used to refer to hemoglobin that has been (53, 202).
modified by the nonenzymatic addition of glucose. These terms
are not interchangeable, however. The current acceptable term for
glycation of hemoglobin in general is glycated hemoglobin (GHb).
Hb A1c is the specific glycated species that is modified by glucose
on the N terminus of the hemoglobin chain. Hb A1c is also the
2. RATIONALE
internationally accepted term for reporting all GHb results. Assay
methods that measure total GHbs (e.g., boronate affinity methods) Glycated proteins are formed posttranslationally from the slow,
should be calibrated to report an equivalent Hb A1c and be reported nonenzymatic reaction between glucose and free amino groups
as Hb A1c for purposes of harmonization of results. Hb A1 is com-
on proteins (203). For Hb, the rate of GHb synthesis is prin-
posed of Hb A1a, Hb A1b, and Hb A1c and should not be measured or
reported. The term A1C test is used by the ADA in place of Hb A1c cipally a function of the glucose concentration to which the
to facilitate communication with patients. As described in the text, erythrocytes are exposed, integrated over the time of expo-
most of the clinical-outcome data that are available for the effects sure. GHb is a clinically useful index of mean glycemia during
of metabolic control on complications (at least for the DCCT and
the preceding 120 days, the average life span of erythrocytes
UKPDS) involved the use of assay methods that quantified Hb A1c.
In this report, we use the abbreviation GHb to include all forms of (147, 203206). Several studies have demonstrated a close
glycated hemoglobin. mathematical relationship between Hb A1c con centration and

25
26 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

mean glycemia, which should allow the expression of Hb A1c show that any method type or analyte is clinically superior to
as an estimated average glucose (eAG) concentration (205, any other. The GHb results reported for the same blood sample
207209). Analogous to Hb (in erythrocytes), serum proteins could differ considerably among methods, however, unless
become glycated. Commercial assays are available that mea- they have been standardized to a common reference [e.g., with-
sure total glycated protein (termed fructosamine) or glycated out standardization, the same blood sample could be read as
albumin in the serum. The concentrations of these glycated 7% (42 mmol/ mol) in one laboratory and 9% (75 mmol/mol)
proteins also reflect mean glycemia, but over a much shorter in another] (53, 147, 204, 212215).
time (1530 days) than GHb (60120 days) (147, 203206, In 1996, the NGSP was initiated to standardize GHb
210, 211). The clinical utility of glycated proteins other than test results among laboratories to DCCT-equivalent values
Hb has not been clearly established, however, and there is no (215). The rationale for standardizing GHb test results to
convincing evidence that relates their concentrations to the DCCT values was that the DCCT had determined the rela-
chronic complications of diabetes (147, 187). tionship between the results obtained for a specific GHb test
(Hb A1c) and long-term complications in patients with type
1 diabetes (44, 147, 187). The NGSP was developed under
the auspices of the AACC and is endorsed by the ADA,
3. ANALYTICAL CONSIDERATIONS which recommends that laboratories use only GHb meth-
ods that have passed certification testing by the NGSP (21,
147). In addition, the ADA recommends that all laboratories
Recommendation
performing GHb testing participate in the CAP proficiency-
Laboratories should use only Hb A1c assay methods that are testing survey for Hb A1c, which uses fresh whole-blood
certified by the National Glycohemoglobin Standardization samples (216).
Program (NGSP) as traceable to the dcct reference. The The NGSP Laboratory Network includes a variety of cer-
manufacturers of Hb A1c assays should also show traceability tified assay methods, each calibrated to the DCCT reference.
to the IFCC reference method. The DCCT reference is an HPLC cation-exchange method that
GPP quantifies Hb A1c; this method is a CLSI-designated compari-
son method (217). The assay method has been used since 1978
and has demonstrated good long-term precision (between-run
CVs are consistently <3%) (216). Secondary reference labo-
Recommendation ratories in the Network interact with manufacturers of GHb
methods to assist them, first in calibrating their methods and
Laboratories that measure Hb A1c should participate in a
then in providing comparison data for certification of traceabil-
proficiency-testing program, such as the College of Ameri-
ity to the DCCT. Certification is valid for 1 year. An impor-
can Pathologists (CAP) Hb A1c survey, that uses fresh blood
tant adjunct to the program is the Hb A1c proficiency-testing
samples with targets set by the NGSP Laboratory Network.
survey administered by CAP. Since 1996 (starting with a pilot
GPP
project including 500 laboratories and expanded to all labora-
tories in 1998), the survey has used fresh whole-blood sam-
Approximately 100 different GHb assay methods are in cur- ples with NGSP-assigned target values. Since initiation of the
rent use. They range from low-throughput research laboratory NGSP in 1996, the survey has documented a steady improve-
component systems and manual minicolumn methods to high- ment in comparability of GHb values among laboratories, both
throughput automated systems dedicated to Hb A1c measure- within and between methods (216, 218). In 2007, CAP initi-
ments. Most methods can be classified into one of 2 groups ated accuracy-based grading with the value of each sample
according to assay principle (147, 181, 204). The first group assigned by the NGSP Network. The objective is to reduce bias
includes methods that quantify GHb on the basis of charge and imprecision among assays. The NGSP Web site (http://
differences between glycated and nonglycated components. www.ngsp.org) provides detailed information on the certifica-
Examples include cation-exchange chromatography and agar- tion process and maintains a listing of certified assay methods
gel electrophoresis. The second group includes methods that (updated monthly) and factors that are known to interfere with
separate components on the basis of structural differences specific methods.
between glycated and nonglycated components. Examples In 1997, the IFCC formed a committee to develop a
include boronate affinity chromatography and immunoassay. higher-order reference method and reference materials for Hb
Most charge-based and immunoassay methods quantify Hb A1c analysis; the method was approved in 2001 (219, 220).
A1c, which is defined as Hb A with glucose attached to the The analysis is performed by cleaving Hb with endoprotein-
N-terminal valine of one or both chains. Other methods ase Glu-C and separating the resulting glycated and nongly-
quantify total glycated hemoglobin, which includes both cated N-terminal -chain hexapeptides by HPLC (220). The
Hb A1c and other Hbglucose adducts (e.g., glucoselysine hexapeptides are quantified with electrospray ionization mass
adducts and glucosea-chain N-terminal valine adducts). Gen- spectrometry or capillary electrophoresis. The 2 methods use
erally, the results of methods that use different assay principles the same primary reference materials, and the results are essen-
show excellent correlation, and there are no convincing data to tially identical. Hb A1c is measured as the ratio of the glycated
Hb A1c 27

N-terminal peptide to the nonglycated N-terminal peptide and but statistically significant, progressive increase in normal
is reported in millimoles of deoxyfructosyl Hb per mole of Hb. Hb A1c levels with aging remain to be determined (226).
Of note, preparing and measuring samples with this method is The effects of race on Hb A1c values are controversial.
laborious, very expensive, and time-consuming. The method Several studies have suggested a relatively higher Hb A1c in
was never envisioned as a practical means of assaying clinical African American and Hispanic populations than in Caucasian
samples. It will only be used by manufacturers to standardize populations at the same level of glycemia (225, 227, 228). The
the assays. Like the NGSP, the IFCC has established a network accumulated evidence suggests that there are differences in Hb
of laboratories (221) (11 at the time of writing). The IFCC A1c among racial groups; however, the measurement of chronic
offers manufacturers calibrators and controls as well as a mon- glucose concentrations in these studies has not been suffi-
itoring program (221). Unlike the NGSP, the IFCC network ciently frequent to capture adequately the actual mean glyce-
does not have a certification program. mia. Moreover, it is not clear that the differences in Hb A1c have
A comparison of Hb A1c results obtained with pooled clinical significance. A recent analysis of 11 092 adults showed
blood samples in the IFCC and NGSP (DCCT-aligned) net- that blacks had mean Hb A1c val ues 0.4% higher than whites
works has revealed a linear relationship (termed the mas- (229); however, race did not modify the association between
ter equation): NGSP% = (0.915 x IFCC%) + 2.15 (220). Hb A1c concentration and adverse cardiovascular outcomes or
Although the clinical values obtained with assays stan- death (229). The ADAG study, which included frequent glu-
dardized with the new IFCC method correlate tightly with cose measurements, did not show a significantly different rela-
NGSP values, the absolute Hb A1c values reported differ by tionship between the calculated mean glucose concentration
1.5%2.0% Hb A1c. Concern regarding the clinical impact of during 3 months and the Hb A1c value at the end of the 3 months
changing patients Hb A1c values led in 2007 to an agreement for Africans/African Americans and Caucasians. The relatively
between the IFCC and the major diabetes organizations to small size of the African/African American population, how-
report IFCC Hb A1c results (in millimoles per mole) as the ever, limits the interpretation of this finding (209).
equivalent NGSP DCCT-aligned result (a percentage based Any condition that shortens erythrocyte survival or
on the master equation) and as a calculated eAG based on decreases mean erythrocyte age (e.g., recovery from acute
the A1c-Derived Average Glucose (ADAG) study (209, 222). blood loss, hemolytic anemia) falsely lowers Hb A1c test
In the revised agreement, published in 2010 (223), both results, regardless of the assay method (147). Vitamins C and
NGSP and IFCC units were recommended, but the decision E are reported to falsely lower test results, possibly by inhibit-
to report eAG was left to the discretion of individual coun- ing Hb glycation (230, 231). Iron deficiency anemia increases
tries. Notwithstanding the agreement, it appears unlikely test results (232). Food intake has no significant effect on test
that universal reporting of Hb A1c will be adopted; however, results. Hypertriglyceridemia, hyperbilirubinemia, uremia,
the master equation allows conversion between IFCC and chronic alcoholism, chronic ingestion of salicylates, and opiate
NGSP numbers. addiction reportedly interfere with some assay methods, falsely
increasing results (204, 233).
A. Preanalytical. Several Hb variants (e.g., Hbs S, C, D, and E) and chemi-
cally modified Hb derivatives interfere with some assay meth-
Recommendation ods [independently of any effects due to shortened erythrocyte
survival (234236); for a review, see (233)]. Depending on the
Laboratories should be aware of potential interferences, particular hemoglobinopathy and assay method, results can be
including hemoglobinopathies, that may affect Hb A1c test either falsely increased or falsely decreased. Some methods may
results, depending on the method used. In selecting assay give a value in the reference interval for a nondiabetic individual
methods, laboratories should consider the potential for inter- with an Hb variant, but that is no assurance that no interference
ferences in their particular patient population. In addition, is present. The interference may be subtle in the reference inter-
disorders that affect erythrocyte turnover may cause spurious val but may increase steadily with increasing Hb A1c. Boronate
results, regardless of the method used. affinity chromatography assay methods are generally considered
GPP to be less affected by Hb variants than other methods. In some
instances, such as with most cation-exchange HPLC methods,
manual inspection of chromatograms or an automated report by
Patient variables. Hb A1c results are not affected significantly the device can alert the laboratory to the presence of either a
by acute fluctuations in blood glucose concentrations, such as variant or a possible interference. If an appropriate method is
those occurring with illness or after meals; however, age and used, Hb A1c can be measured accurately in the vast majority of
race reportedly influence Hb A1c. Published data show age- individuals heterozygous for Hb variants (for a summary of pub-
related increases in Hb A1c values of approximately 0.1% per lished studies, see http:// www.ngsp.org). If altered erythrocyte
decade after age 30 years (224, 225). Careful phenotyping of turnover interferes with the relationship between mean blood
individuals with OGTT supports an increase in Hb A1c with age, glucose and Hb A1c values, or if a suitable assay method is not
even after removing from the study population patients with available for interfering Hb variants, alternative nonHb-based
otherwise undiagnosed diabetes and persons with impaired methods for assessing long-term glycemic control (such as fruc-
glucose tolerance (224). The clinical implications of the small, tosamine assay) may be useful (233).
28 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

Given that interferences are method specific, product ing on the assay method, may show matrix effects when new
instructions from the manufacturer should be reviewed before reagents or columns are introduced. We recommend that a lab-
the Hb A1c assay method is used. A list of interfering factors oratory consider using both commercial and in-house controls
for specific assays is maintained on the NGSP Web site (http:// to optimize performance monitoring.
www.ngsp.org). In selecting an assay method, a laboratory
Reference intervals. A laboratory should determine its own ref-
should consider characteristics of the patient population served
erence interval according to CLSI guidelines (CLSI Document
(e.g., a high prevalence of Hb variants).
C28A), even if the manufacturer has provided one. Nondiabetic
Sample collection, handling, and storage. Blood can be test individuals should be nonobese, have an FPG concentration
obtained by venipuncture or by finger-stick capillary sam- <5.6 mmol/L (100 mg/dL), and, ideally, have a 2-h post-OGTT
pling (237, 238). Blood tubes should contain the anticoagu- plasma glucose value of <11.1 mmol/L (200 mg/dL). For NGSP-
lant specified by the manufacturer of the Hb A1c assay method certified assay methods, reference intervals should not deviate
(EDTA can be used unless the manufacturer specifies other- substantially (e.g., >0.5%) from 4%6% (2042 mmol/mol).
wise). Sample stability is assay method specific (239, 240). Note that treatment target values recommended by the ADA and
In general, whole-blood samples are stable for up to 1 week at other clinical organizations, not reference intervals, are used to
4 C (240). For most methods, whole-blood samples stored at evaluate metabolic control in patients.
270 C or colder are stable over the long term (at least 1 year), but
samples are not as stable at 220 C. Improper handling of sam-
Recommendation
ples, such as storage at high temperatures, can introduce large arti-
facts that may not be detectable, depending on the assay method. Samples with Hb A1c results below the lower limit of the ref-
Manufacturers have introduced a number of convenient erence interval or >15% Hb A1c should be verified by repeat
blood-collection systems, including filter paper and small testing.
vials containing stabilizing/lysing reagent (241243). These B (low)
systems are designed for field collection of samples and
routine mailing to the laboratory and are generally matched
with specific assay methods. They should be used only if
Recommendation
studies have been performed to establish the comparabil-
ity of test results for these collection systems with standard Hb A1c values that are inconsistent with the clinical presenta-
sample-collection and handling methods for the specific tion should be investigated further.
assay method used. GPP
B. Analytical. Performance goals and quality control. Several
expert groups have presented recommendations for assay per- Out-of-range samples. A laboratory should repeat testing for
formance. Early reports recommended that the inter-assay CV all sample results below the lower limit of the reference inter-
be <5% at normal and diabetic GHb concentrations (244). Sub- val, and if these results are confirmed, the physician should
sequent reports have suggested lower CVs [e.g., intralaboratory be informed to determine whether the patient has a variant
CVs <3% (245) or <2% (246), and interlaboratory CVs <5% Hb or shows evidence of erythrocyte destruction. If possible,
(245)]. Intraindividual CVs for healthy persons are very small the repeat Hb A1c measurement should be performed with a
(<2%), and many current assay methods can achieve intralabo- method based on an analytical principle that is different from
ratory and interlaboratory CVs of <2% and <3%, respectively the initial assay. In addition, samples with results >15% Hb
(247). A recent statistical analysis calculated appropriate goals A1c (140 mmol/mol) should be assayed a second time; if the
for Hb A1c assay performance (218). If the reference change results are confirmed, the possibility of an Hb variant should
value (also termed critical difference) is used, an analytical CV be considered (233). Any result that does not correlate with
2% will produce a 95% probability that a difference of 0.5% the clinical impression should also be investigated.
Hb A1c between successive patient samples is due to a significant
change in glycemic control [when Hb A1c is 7% (53 mmol/mol)]. Removal of labile GHb. The formation of Hb A1c involves an
In addition, if a method has no bias, a CV of 3.5% is necessary intermediate Schiff base, which is called pre-A1c or labile
to have 95% confidence that the Hb A1c result for a patient with A1c (248). This Schiff base is formed rapidly with hyperglyce-
a true Hb A1c of 7% (53 mmol/mol) will be between 6.5% and mia and can interfere with some Hb A1c assay methods if it is
7.5% (between 48 and 58 mmol/mol) (218). We recommend an not completely removed or separated. Most currently available
intralaboratory CV <2% and an interlaboratory CV <3.5%. For automated assays either remove the labile preHb A1c dur ing
a single method, the goal should be an interlaboratory CV <3%. the assay process or do not measure the labile product.
A laboratory should include 2 control materials with dif-
ferent mean values (high and low) at both the beginning and
the end of each days run. Frozen whole-blood controls stored 4. INTERPRETATION
in single-use aliquots at 270 C or colder are ideal and are sta-
ble for months or even years, depending on the assay method. A. Laboratoryphysician interactions. A laboratory should
Lyophilized controls are commercially available but, depend- work closely with physicians who order Hb A1c testing. Proper
Hb A1c 29

interpretation of test results requires an understanding of the interventions that achieved Hb A1c values 0.8% to 1.1% higher
assay method, including its known interferences. For example, (5052). The ACCORD (Action to Control Cardiovascular Risk
if the assay method is affected by hemoglobinopathies (inde- in Diabetes) study demonstrated increased mortality with very
pendently of any shortened erythrocyte survival) or uremia, the intensive diabetes therapy [Hb A1c, 6.4% vs 7.5% (46 vs 58 mmol/
physician should be made aware of this interference. mol)]. These Hb A1c values apply only to assay methods that have
An important advantage of using an NGSP-certified been certified as traceable to the DCCT reference, with a reference
method is that the laboratory can provide specific infor- interval of approximately 4%6% Hb A1c (2042 mmol/mol). In
mation relating Hb A1c test results to both mean glycemia the DCCT, each 10% reduction in Hb A1c (e.g., 12% vs 10.8% or
and outcome risks as defined in the DCCT and UKPDS 8% vs 7.2%) was associated with an approximately 45% lower
(44, 147, 187). This information is available on the NGSP risk for the progression of diabetic retinopathy (42). Comparable
Web site. For example, each 1% (approximately 11 mmol/ risk reductions were found in the UKPDS (197). Also of note is
mol) change in Hb A1c is related to a change in the mean that decreases in Hb A1c were associated in the DCCT and UKPDS
plasma glucose concentration of approximately 1.6 mmol/L with an increased risk for severe hypoglycemia.
(29 mg/dL). Reporting Hb A1c results with a calculated eAG
will eliminate the need for healthcare providers or patients
to perform these calculations themselves. The equation gen- Recommendation
erated by the ADAG study is the most reliable to date (209). Hb A1c testing should be performed at least biannually in all
Some evidence suggests that immediate feedback of patients and quarterly for patients whose therapy has changed
Hb A 1c test results to patients at the time of the clinic visit or who are not meeting treatment goals.
leads to an improvement in their long-term glycemic con- B (low)
trol (249, 250). Not all publications have supported this
observation (251), however, and additional studies are
needed to confirm these findings before this strategy can Testing frequency. There is no consensus on the optimal fre-
be generally recommended. It is possible to achieve the quency of Hb A1c testing. The ADA recommends (21), For
goal of having Hb A1c test results available at the time of any individual patient, the frequency of A1C testing should
the clinic visit by either having the patient send in a blood be dependent on the clinical situation, the treatment regimen
sample shortly before the scheduled clinic visit or having a used, and the judgment of the clinician. In the absence of well-
rapid-assay system convenient to the clinic. controlled studies that suggest a definite testing protocol, expert
B. Clinical application. opinion recommends Hb A1c testing at least two times a year in
patients who are meeting treatment goals (and who have stable
glycemic control) . . . and quarterly in patients whose therapy
Recommendation
has changed or who are not meeting glycemic goals (21).
Treatment goals should be based on ADA recommendations, These testing recommendations are for nonpregnant patients
which include generally maintaining Hb A1c concentrations at with either type 1 or type 2 diabetes. In addition, all patients
<7% and more-stringent goals in selected individual patients with diabetes who are admitted to a hospital should have Hb A1c
if they can be achieved without significant hypoglycemia or measured if the results of testing in the previous 23 months are
other adverse treatment effects. Somewhat higher intervals not available (21). Diabetes quality-assurance programs [e.g.,
are recommended for children and adolescents and may be Provider Recognition Program and HEDIS (Healthcare Effec-
appropriate for patients with a limited life expectancy, exten- tiveness Data and Information Set) (199, 200)] have generally
sive comorbid illnesses, a history of severe hypoglycemia, or required documentation of the percentage of diabetes patients
advanced complications (note that these values are applicable who have had at least one Hb A1c measurement during the pre-
only if the NGSP has certified the assay method as traceable ceding year. Studies have established that serial Hb A1c mea-
to the DCCT reference). surements (quarterly for 1 year) produce large improvements in
A (high) Hb A1c values in patients with type 1 diabetes (252).
Interpretation. Hb A1c values in patients with diabetes consti-
Treatment goals. Hb A1c measurements are now a routine com- tute a continuum. They range from within the reference interval
ponent of the clinical management of patients with diabetes. in a small percentage of patients whose mean plasma glucose
Principally on the basis of the DCCT results, the ADA has recom- concentrations are close to those of nondiabetic individuals, to
mended that a primary goal of therapy be an Hb A1c value <7% (53 markedly increased values (e.g., two- to threefold increases in
mmol/mol) (21). Lower targets may be considered for individual some patients) that reflect an extreme degree of hyperglycemia.
patients, e.g., in diet-treated type 2 diabetes. Other major clinical A proper interpretation of Hb A1c test results requires that phy-
organizations have recommended similar targets (53); however, sicians understand the relationship between Hb A1c values and
recent studies that used multiple medications to treat type 2 dia- mean plasma glucose, the kinetics of Hb A1c, and specific assay
betes and aimed for Hb A1c concentrations <6.5% (48 mmol/mol) limitations/interferences (147). Small changes in Hb A1c (e.g.,
have not demonstrated consistent benefits and failed to observe 0.3% Hb A1c) over time may reflect assay imprecision rather
any benefit with regard to macrovascular disease, compared with than a true change in glycemic status (218).
30 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

5. EMERGING CONSIDERATIONS few point-of- care devices that measure Hb A1c met acceptable
analytical performance criteria (254). Absent objectiveand
ongoingdocumentation of performance with accuracy-based
Recommendation proficiency testing that uses whole blood (or other suitable
material that is free from matrix effects), point-of-care Hb A1c
Hb A1c may be used for the diagnosis of diabetes, with values devices should not be used for diabetes diagnosis or screening.
>6.5% being diagnostic. An ngsp-certified method should be The ADA has endorsed the use of Hb A1c for the diagnosis of
performed in an accredited laboratory. Analogous to its use in the diabetes (Table 6) (21), as have the Endocrine Society (255) and
management of diabetes, factors that interfere with or adversely the WHO. The American Association of Clinical Endocrinolo-
affect the Hb A1c assay will preclude its use in diagnosis. gists supports it in a more limited fashion. Other international
A (moderate) organizations, including the IDF, are considering Hb A1c testing
for diabetes diagnosis and screening. Note that glucose-based
testing for diagnosis remains valid. Analogous to the concept of
Recommendation
impaired fasting glucose and impaired glucose tolerance, indi-
Point-of-care Hb A1c assays are not sufficiently accurate to viduals with Hb A1c values between 5.7% and 6.4% (39 and 46
use for the diagnosis of diabetes. mmol/mol) should be considered at high risk for future diabetes
B (moderate) and should be counseled about effective measures to reduce their
risk (93).
A. Use of Hb A1c for diabetes screening/diagnosis. The role of B. Use of other glycated proteins, including advanced glyca-
Hb A1c in the diagnosis of diabetes has been considered for sev- tion end products, for routine management of diabetes. Fur-
eral years (19, 24, 37, 253). In the past, the lack of standard- ther studies are needed to determine whether other glycated
ization has been a major barrier. With improved standardization proteins, such as fructosamine or glycated serum albumin, are
through the NGSP and the IFCC, and new data demonstrating clinically useful for routine monitoring of patients glycemic
the association between Hb A1c concentrations and the risk for status. Further studies are also needed to determine if measure-
retinopathy, the International Expert Committee recommended ments of advanced glycation end products are clinically useful
the use of Hb A1c in the diagnosis of diabetes (20). In making its as predictors of risk for chronic diabetes complications (256).
recommendation, the Committee also considered several techni- Only 1 study of a subset of DCCT patients evaluated advanced
cal advantages of Hb A1c testing compared with glucose testing, glycation end products in dermal collagen obtained with skin
such as its preanalytical stability and decreased biological varia- biopsies. Interestingly, the concentration of advanced glycation
tion. Finally, the clinical convenience of the Hb A1c assay, which end products in dermal collagen correlated more strongly with
requires no patient fasting or tolerance tests, compared with glu- the presence of complications than the mean Hb A1c values
cose-based diagnosis, convinced the Committee to recommend (257). The clinical role of such measurements remains unde-
Hb A1c testing for diagnosis. A value 6.5% (48 mmol/mol) was fined. Similarly, the role of noninvasive methods that use light
considered diagnostic on the basis of the observed relationship to measure glycation transdermally is undefined.
with retinopathy. For diagnosis, a positive test result [6.5%
(48 mmol/mol)] should be confirmed with a repeat assay. The C. Global harmonization of Hb A1c testing and uniform report-
ADA indicates that although either an Hb A1c assay or a glu- ing of results. As noted above, the NGSP has largely succeeded
cose assay (FPG or OGTT) can be used as the confirmatory test, in standardizing the GHb assay across methods and laborato-
repeating the same test is preferred (93). The frequency of Hb ries. Furthermore, the IFCC standardization, which provides
A1c testing for diagnosis has not been established, but guidelines a chemically discrete standard, is being implemented world-
similar to those for glucose-based testing seem appropriate. Only wide. The reporting recommendations (223) need to be imple-
NGSP-certified Hb A1c methods should be used to diagnose mented with the education of healthcare providers and patients.
(or screen for) diabetes. The ADA cautions that point-of-care Some believe that reporting eAG should complement the cur-
devices for measuring Hb A1c should not be used for diagnosis rent reporting in NGSP/DCCT-aligned units (percentages) and
(93). Although several point-of-care Hb A1c assays are NGSP the new IFCC results (millimoles per mole), because the eAG
certified, the test is waived in the US, and proficiency testing is results will be in the same units (millimoles per liter or mil-
not necessary. Therefore, no objective information is available ligrams per deciliter) as patients self-monitoring. Educational
concerning their performance in the hands of those who mea- campaigns will be necessary, however, to ensure clear under-
sure Hb A1c in patient samples. A recent evaluation revealed that standing of this assay, which is central to diabetes management.
Chapter 10
Genetic Markers

1. USE indicate a modified risk of type 1 diabetes in persons positive

for islet cell autoantibodies, because protective alleles do not
prevent the appearance of islet cell autoantibodies (most often
as single autoantibodies) but may delay the onset of clinical
Recommendation
diabetes. Typing of the class II major histocompatibility anti-
Routine measurement of genetic markers is not of value at gens or HLA-DRB1, -DQA1, and -DQB1 is not diagnostic for
this time for the diagnosis or management of patients with type 1 diabetes. Some haplotypes induce susceptibility, how-
type 1 diabetes. For selected diabetic syndromes, including ever, whereas others provide significant delay or even protec-
neonatal diabetes, valuable information can be obtained with tion. Thus, HLA-DR/DQ typing can be used only to increase
definition of diabetes-associated mutations. or decrease the probability of type 1 diabetes presentation and
A (moderate) cannot be recommended for routine clinical diagnosis or clas-
sification (264).
The precision in the genetic characterization of type 1 dia-
A. Diagnosis/screening. Type 1 diabetes. Genetic markers are
betes may be extended by typing for polymorphisms in several
currently of limited clinical value in evaluating and managing
genetic factors identified in genome-wide association stud-
patients with diabetes; however, mutational analysis is rapidly
ies (265). Non-HLA genetic factors include the INS (insulin),
emerging for classifying diabetes in the neonate (258260) and in
PTPN22 [protein tyrosine phosphatase, non-receptortype 22
young patients with a dominant family history of diabetes, often
(lymphoid)], and CTLA4 (cytotoxic T-lymphocyte-associated
referred to as maturity-onset diabetes of the young (MODY)
protein 4) genes and several others (263, 265). These additional
(261). Type 1 or autoimmune diabetes is strongly associated
genetic factors may assist in assigning a probability for a diag-
with HLADR12 (major histocompatibility complex, class II, DR)
nosis of type 1 diabetes of uncertain etiology (266).
and HLA-DQ (major histocompatibility complex, class II, DQ)
It is possible to screen newborn children to identify those
genes. HLA-DQA1 and HLA-DQB1 genotyping can be useful to
at increased risk for developing type 1 diabetes (267269).
indicate the absolute risk of diabetes. The HLA DQA1*0301
This strategy cannot be recommended until a proven interven-
DQB1*0302 and DQA1*0501DQB1*0201 haplotypes, alone
tion is available to delay or prevent the disease (270). There
or in combination, may account for up to 90% of children and
is some evidence that early diagnosis may prevent hospital-
young adults with type 1 diabetes (262). These 2 haplotypes may
ization for ketoacidosis and preserve residual beta cells (271).
be present in 30%40% of a Caucasian population, and HLA
The rationale for the approach is thus discussed below under
is therefore necessary but not sufficient for disease. The HLA-
Emerging Considerations.
DQ and HLA-DR genetic factors are by far the most important
determinants of type 1 diabetes risk (263). HLA typing may be
used in combination with islet autoantibody analyses to exclude Recommendation
type 1 diabetes in assisting in the diagnosis of genetic forms of
There is no role for routine genetic testing in patients with
diabetes.
type 2 diabetes. These studies should be confined to the
As indicated below, HLA-DR/DQ typing can be useful to
research setting and evaluation of specific syndromes.
12
 uman genes: HLA-DR, major histocompatibility complex, class II,
H A (moderate)
DR; HLA-DQ, major histocompatibility complex, class II, DQ; INS,
insulin; PTPN22, protein tyrosine phosphatase, non-receptor type
22 (lymphoid); CTLA4, cytotoxic T-lymphocyte-associated protein 4; Type 2 diabetes. Fewer than 5% of patients with type 2 diabetes
KCNJ11, potassium inwardly-rectifying channel, subfamily J, mem- have been resolved on a molecular genetic basis, and, not sur-
ber 11; HLA-A, major histocompatibility complex, class I, A; HLA-B,
major histocompatibility complex, class I, B; HLA-C, major histo-
prisingly, most of these patients have an autosomal dominant
compatibility complex, class I, C; HNF4A, hepatocyte nuclear factor form of the disease or very high degrees of insulin resistance.
4, alpha; HNF1A, HNF1 homeobox A; HNF1B, HNF1 homeobox Type 2 diabetes is a heterogeneous polygenic disease with
B; PDX1, pancreatic and duodenal homeobox 1 (formerly known both resistance to the action of insulin and defective insulin
as IPF1); NEUROD1, neurogenic differentiation 1 (also known as
NeuroD and BETA2); KLF1, Kruppel-like factor 1 (erythroid); GCK,
secretion (3, 4). Multiple genetic factors interact with exog-
glucokinase (hexokinase 4); CEL, carboxyl ester lipase (bile salt- enous influences (e.g., environmental factors such as obesity)
stimulated lipase). to produce the phenotype. Identification of the affected genes

31
32 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

is therefore highly complex. Recent genome-wide association PraderWilli syndrome, which maps to chromosome 15q, or
studies have identified >30 genetic factors that increase the risk to the absence of adipose tissue inherent to the recessive Seip
for type 2 diabetes (272, 273). The risk alleles in these loci all Berardinelli syndrome of generalized lipodystrophy, which
have rela-tively small effects (odds ratios of 1.1 to 1.3), how- maps to chromo-some 9q34 (1, 276). More than 60 distinct
ever, and do not significantly enhance our ability to predict the genetic disorders are associated with glucose intolerance or
risk of type 2 diabetes (274). frank diabetes. Many forms of type 2 diabetes (which are usu-
ally strongly familial) will probably be understood in defined
MODY. Detecting mutations in MODY patients and their rela-
genetic terms. The complexity of the genetic factors that con-
tives is technically feasible. The reduced costs of sequencing
tribute to type 2 diabetes risk is substantial (272, 273). Several
and emerging new technologies make it possible to identify
genetic factors for MODY have been identified, and there are
mutations and toproperly classify MODY patients on the basis
large numbers of individual mutants. Persons at risk within
of specific mutations. As direct automated sequencing of genes
MODY pedigrees can be identified through genetic means.
becomes standard, it is likely that the detection of specific dia-
Depending on the specific MODY mutation, the disease can
betes mutations will become routine.
be mild (e.g., glucokinase mutation) and not usually associ-
B. Monitoring/prognosis. Although genetic screening may ated with long-term complications of diabetes, or it can be as
provide information about prognosis and could be useful for severe as typical type 1 diabetes [e.g., hepatocyte nuclear fac-
genetic counseling, genotype may not correlate with the pheno- tor (HNF) mutations] (277).
type. In addition to environmental factors, interactions among Eight different MODYs have been identified. MODY-1, -3,
multiple loci for the expression of quantitative traits may be -4, -5, -6, and -7 are all caused by mutations in the genes encod-
involved. Genetic identification of a defined MODY will have ing transcription factors that regulate the expression of genes
value for anticipating the prognosis. Infants with neonatal dia- in pancreatic beta cells. These genes are HNF4A (hepatocyte
betes due to a mutation in the KCNJ11 (potassium inwardly- nuclear factor 4, alpha) in MODY-1, HNF1A (HNF1 homeo-
rectifying channel, subfamily J, member 11; also known as box A) in MODY-3, HNF1B (HNF1 homeobox B) in MODY-5,
KIR6.2) gene may be treated with sulfonylurea rather than with PDX1 (pancreatic and duodenal homeobox 1; formerly known
insulin (258, 259). as IPF1) in MODY-4, NEUROD1 (neurogenic differentiation
1; also known as NeuroD and BETA2) in MODY-6, and KLF1
[Kruppel-like factor 1 (erythroid)] in MODY-7. Homozygous
mutations of the PDX1 gene have been shown to lead to pan-
2. RATIONALE creatic agenesis, and heterozygous PDX1 mutations have been
shown to cause MODY-4 (276). The modes of action of the HNF
The HLA system, which has a fundamental role in the adaptive lesions in MODY are still not clear. It is likely that mutations
immune response, exhibits considerable genetic complexity. in HNF1A, HNF1B, and HNF4A cause diabetes because they
The HLA complex on chromosome 6 contains class I and class impair insulin secretion. MODY-2 is caused by mutations in the
II genes that code for several polypeptide chains (275). The GCK [glucokinase (hexokinase 4)] gene. The product of the gene
major (classic) class I genes are HLA-A (major histocompat- is an essential enzyme in the glucose-sensing mechanism of beta
ibility complex, class I, A), HLA-B (major histocompatibility cells, and mutations in this gene lead to partial deficiencies of
complex, class I, B), and HLA-C (major histocompatibility insulin secretion. MODY-8 is due to mutations in the CEL [car-
complex, class I, C). The loci of class II genes are designated boxyl ester lipase (bile salt-stimulated lipase)] gene.
by 3 letters: the first (D) indicates the class, the second (M, O,
P, Q, or R) indicates the family, and the third (A or B) indi-
cates the chain. Both classes of the encoded molecules are
3. ANALYTICAL CONSIDERATIONS
heterodimers. Class I molecules consist of an a chain and
2-microglobulin, and class II molecules have a and chains.
A detailed review of analytical issues will not be attempted
The function of the HLA molecules is to present short pep-
here, because genetic testing for diabetes outside of a research
tides derived from pathogens or autoantigens to T cells to initi-
setting is currently not recommended for clinical care. Sero-
ate the adaptive immune response (275). Genetic studies have
logic HLA typing should be replaced by molecular methods,
revealed an association between certain HLA alleles and auto-
because antibodies with a mixture of specificities and cross-
immune diseases. These diseases include, but are not confined
reactivities have been estimated to give inaccurate results in
to, ankylosing spondylitis, celiac disease, Addison disease, and
approximately 15% of typings.
type 1 diabetes (275). Not only the disease but also autoanti-
bodies, which are markers of the diseases pathogenesis, are A. Preanalytical. Mutations are detected by using genomic
often associated with HLADRB1, HLA-DQA1, and HLA- DNA extracted from peripheral blood leukocytes. Blood sam-
DQB1, indicating that self-peptides may also be presented to ples should be drawn into test tubes containing EDTA, and the
T cells (262). DNA should be extracted within 3 days; longer periods both
Genetic testing for syndromic forms of diabetes is the lower the yield and degrade the quality of the DNA obtained.
same as that for the underlying syndrome itself (1). Such forms Genomic DNA can be isolated from fresh or frozen whole
of diabetes may be secondary to the obesity associated with blood by lysis, digestion with proteinase K, extraction with
Genetic Markers 33

phenol, and then dialysis. The average yield is 100200 g types of DRB1*04 are susceptible, whereas the 0403 and 0406
DNA from 10 mL of whole blood. DNA samples are best kept subtypes are negatively associated with the disease, even when
at 280 C in Tris-EDTA solution. These conditions maintain found in HLA genotypes with the susceptible DQA1*0301
DNA sample integrity virtually indefinitely. DQB1*0302. DR molecules are heterodimers also; however,
the DRa chain is invariant in all persons. Additional DR
B. Analytical. Methods for the detection of mutations vary with chains (B3, B4, and B5) are not important.
the type of mutation. MODY mutations have substitution, dele- Class II MHC molecules are involved in antigen presenta-
tion, or insertion of nucleotides in the coding regions of the tion to CD4 helper cells, and the associations outlined above
genes. These mutations are detected by the PCR. Detailed pro- are likely to be explained by defective affinities to islet cell
tocols for detecting specific mutations are beyond the scope of antigenic peptides, leading to persistence of T-helper cells
this review. that escape thymic ablation. Class I HLA molecules are also
implicated in type 1 diabetes. Multiple non-HLA loci also con-
tribute to susceptibility to type 1 diabetes (279). For example,
4. INTERPRETATION the variable nucleotide tandem repeat (VNTR) upstream from
the INS gene on chromosome 11q is useful for predicting the
For screening for the propensity for type 1 diabetes in general development of type 1 diabetes, with alleles with the longest
populations, HLA-D genes are the most important, contributing VNTR having protective effects. Typing newborn infants for
as much as 50% of familial susceptibility (278). HLA-DQ genes both HLA-DR and HLADQand to a lesser degree the INS
appear to be central to the HLA-associated risk of type 1 diabe- geneallows prediction of type 1 diabetes to better than 1 in
tes, albeit HLA-DR genes may be independently involved [for 10 in the general population. The risk of type 1 diabetes in
reviews, see (279, 280)]. The heterodimeric proteins that are HLA-identical siblings of a proband with type 1 diabetes is 1
expressed on antigen-presenting cells, B lymphocytes, plate- in 4, whereas siblings who have HLA haplotype identity have
lets, and activated T cellsbut not other somatic cellsare a 1 in 12 risk and those with no shared haplotype have a 1
composed of cis- and transcomplementated a- and -chain het- in 100 risk (280). Genome-wide association studies have con-
erodimers. Thus, in any individual, 4 possible DQ dimers are firmed that the following non-HLA genetic factors increase the
encoded. Persons at the highest genetic risk for type 1 diabetes risk for type 1 diabetes, both in first-degree relatives of type 1
are those in whom all 4 DQ combinations meet this criterion. diabetes patients and in the general population: INS, VNTR,
Thus, persons heterozygous for HLA DRB1*04DQA1*0301 CTLA4, PTPN22, and others (263, 265, 282, 283).
DQB1*0302 and DRB1*03 DQA1*0501DQB1*0201 are
the most susceptible, with an absolute lifetime risk of type 1
diabetes in the general population of about 1 in 12. Persons 5. EMERGING CONSIDERATIONS
who are protected from developing type 1 diabetes at a young
age are those with HLA DRB1*15DQA1*0201 DQB1*0602 The sequencing of the human genome and the formation of
haplotypes in particular (281). Individuals with DRB1*11 or consortia have produced advances in the identification of the
04 who also have DQB1*0301 are not likely to develop type 1 genetic bases for both type 1 and type 2 diabetes. This progress
diabetes at a young age. HLA-DR is also involved in suscep- should ultimately lead to family counseling, prognostic infor-
tibility to type 1 diabetes, in that the B1*0401 and 0405 sub- mation, and the selection of optimal treatments (276, 284).
34
Chapter 11
Autoimmune Markers

1. USE of hyper-glycemia and subsequent symptoms of diabetes. After

years of type 1 diabetes, some antibodies fall below detec-
tion limits, but GAD65A usually remains increased. Patients
Recommendation
with type 1A diabetes have a significantly increased risk of
Islet cell autoantibodies are recommended for screening non- other autoimmune disorders, including celiac disease, Graves
diabetic family members who wish to donate part of their disease, thyroiditis, Addison disease, and pernicious anemia
pancreas for transplantation into a relative with end-stage (128). As many as 1 in 4 females with type 1 diabetes have
type 1 diabetes. autoimmune thyroid disease, whereas 1 in 280 patients develop
B (low) adrenal autoantibodies and adrenal insufficiency. A minority
of patients with type 1 diabetes (type 1B, idiopathic) have no
known etiology and no evidence of autoimmunity. Many of
these patients are of African or Asian origin.
Recommendation
Islet cell autoantibodies are not recommended for routine Recommendation
diagnosis of diabetes, but standardized islet cell autoantibody
tests may be used for classification of diabetes in adults and Screening patients with type 2 diabetes for islet cell autoan-
in prospective studies of children at genetic risk for type 1 tibodies is not recommended at present. Standardized islet
diabetes after HLA typing at birth. cell autoantibodies are tested in prospective clinical studies
B (low) of type 2 diabetes patients to identify possible mechanisms of
secondary failures of treatment of type 2 diabetes.
B (low)
No therapeutic intervention that will prevent diabetes has been
identified (279, 280). Therefore, although several islet cell
autoantibodies have been detected in individuals with type 1
Recommendation
diabetes, their measurement has limited use outside of clini-
cal studies. Currently, islet cell autoantibodies are not used in Screening for islet cell autoantibodies in relatives of patients
routine management of patients with diabetes. This section with type 1 diabetes or in persons from the general population
focuses on the pragmatic aspects of clinical laboratory testing is not recommended at present. Standardized islet cell auto-
for islet cell autoantibodies. antibodies are tested in prospective clinical studies.
B (low)
A. Diagnosis/screening. Diagnosis. In type 1 diabetes, the
pancreatic islet beta cells are destroyed and lost. In the vast
majority of these patients, the destruction is mediated by an Screening. Only about 15% of patients with newly diag-
autoimmune attack (285). This disease is termed type 1A or nosed type 1 diabetes have a first-degree relative with the
immune-mediated diabetes (Table 5). Islet cell auto-antibod- disease (294). The risk of developing type 1 diabetes in rela-
ies comprise autoantibodies to islet cell cytoplasm (ICA), to tives of patients with the disease is approximately 5%, which
native insulin [referred to as insulin autoantibodies (IAA) is 15-fold higher than the risk in the general population (1 in
(286)], to the 65-kDa isoform of glutamic acid decarboxyl- 250300 lifetime risk). Screening relatives of type 1 diabetes
ase (GAD65A) (287289), to 2 insulinoma antigen 2 proteins patients for islet cell autoantibodies can identify those at high
[IA-2A (290) and IA-2A (also known as phogrin) (291)], and risk for the disease; however, as many as 1%2% of healthy
to 3 variants of zinc transporter 8 (ZnT8A) (292, 293). Auto- individuals have a single autoantibody against insulin, IA-2,
antibody markers of immune destruction are usually present in GAD65, or ZnT8 and are at low risk of developing type 1
85% to 90% of individuals with type 1 diabetes when fasting diabetes (295). Because of the low prevalence of type 1 diabe-
hyperglycemia is initially detected (1). Autoimmune destruc- tes (approximately 0.3% in the general population), the posi-
tion of beta cells has multiple genetic predispositions and is tive predictive value of a single islet cell autoantibody will be
modulated by undefined environmental influences. The auto- low (280). The presence of multiple islet cell autoantibodies
immunity may be present for months or years before the onset (IAA, GAD65A, IA-2A/IA-2A, or ZnT8A) is associated with

35
36 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

a >90% risk of type 1 diabetes (292, 295, 296); however, until to screen nondiabetic family members who wish to donate
cost-effective screening strategies can be developed for young a kidney or part of their pancreas for transplantation; (c) to
children and until effective intervention therapy to prevent or screen women with GDM to identify those at high risk of
delay the onset of the disease becomes available, such testing progression to type 1 diabetes; and (d) to distinguish type 1
cannot be recommended outside of a research setting. from type 2 diabetes in children to institute insulin therapy at
Children with certain HLA-DR and/or HLADQB1 chains the time of diagnosis (304, 305). For example, some pediatric
(*0602/*0603/*0301) are mostly protected from type 1 dia- diabetologists now treat children thought to have type 2 dia-
betes, but not from developing islet cell autoantibodies (297). betes with oral medications but treat autoantibody-positive
Because islet cell auto-antibodies in these individuals have children immediately with insulin. It is possible, however, to
substantially reduced predictive significance, they are often follow patients who are islet cell autoantibody positive to the
excluded from prevention trials. point of metabolic decompensation and then institute insulin
Approximately 5%10% of adult Caucasian patients therapy. The Diabetes Prevention Trial of Type 1 Diabetes
who present with a type 2 diabetes phenotype also have (DPT-1) study failed to show a protective effect of parenteral
islet cell autoantibodies (298), particularly GAD65A, which insulin (306).
predict insulin dependency. This condition has been termed
latent autoimmune diabetes of adulthood (LADA) (299),
type 1.5 diabetes (300), or slowly progressive IDDM
(301). Although GAD65A-positive diabetic patients prog-
2. RATIONALE
ress faster to absolute insulinopenia than do antibody-neg-
The presence of islet cell autoantibodies suggests that
ative patients, many antibody-negative (type 2) diabetic
insulin therapy is the most appropriate therapeutic option,
adults also progress (albeit more slowly) to insulin depen-
especially in a young person. Conversely, in children or
dency with time. Some of these patients may show T-cell
young people without islet cell autoantibodies, consider-
reactivity to islet cell components (300). Islet cell autoan-
ation may be given to a trial of oral agents and lifestyle
tibody testing in patients with type 2 diabetes has limited
changes. There is no unanimity of opinion, but the presence
utility, because the institution of insulin therapy is based on
of islet cell autoantibodies may alter therapy for subsets
glucose control.
of patients, including Hispanic and African American chil-
dren with a potential diagnosis of nonautoimmune diabetes,
Recommendation adults with islet cell autoantibodies but clinically classified
as type 2 diabetes, and children with transient hyperglyce-
There is currently no role for measurement of islet cell auto-
mia. The majority of nondiabetic individuals who have only
antibodies in the monitoring of patients in clinical practice.
1 autoantibody may never develop diabetes. Although the
Islet cell autoantibodies are measured in research protocols
production of multiple islet cell autoantibodies is associ-
and in some clinical trials as surrogate end points.
ated with considerably increased diabetes risk (295, 296),
B (low)
approximately 20% of individuals presenting with new-
onset diabetes produce only a single autoantibody. Prospec-
B. Monitoring/prognosis. No acceptable therapy has been dem- tive studies of children reveal that islet cell autoantibodies
onstrated to prolong the survival of islet cells once diabetes has may be transient, indicating that an islet autoantibody may
been diagnosed or to prevent the clinical onset of diabetes in have disappeared prior to the onset of hyperglycemia or
islet cell autoantibody positive individuals (279). Thus, the diabetes symptoms (307).
use of repeated testing for islet cell autoantibodies to monitor
islet cell autoimmunity is not clinically useful at present. In islet
cell or pancreas transplantation, the presence or absence of islet 3. ANALYTICAL CONSIDERATIONS
cell autoantibodies may clarify whether subsequent failure of
the transplanted islets is due to recurrent autoimmune disease
or to rejection (302). When a partial pancreas has been trans- Recommendation
planted from an identical twin or other HLA-identical sibling, It is important that islet cell autoantibodies be measured only
the appearance of islet cell autoantibodies may raise consider- in an accredited laboratory with an established quality-control
ation regarding the use of immunosuppressive agents to try to program and participation in a proficiency-testing program.
halt the recurrence of diabetes. Notwithstanding these theoreti- GPP
cal advantages, the value of this therapeutic strategy has not been
established.
Some experts have proposed that testing for islet cell For IAAs, a radioisotopic method that calculates the dis-
autoantibodies may be useful in the following situations: (a) placeable insulin radioligand binding after the addition
to identify a subset of adults initially thought to have type 2 of excess nonradiolabeled insulin (308) is recommended.
diabetes but who have islet cell autoantibody markers of type Results are reported as positive when specific antibody
1 diabetes and who progress to insulin dependency (303); (b) binding exceeds the 99th percentile or possibly exceeds the
Autoimmune Markers 37

mean plus 2 (or 3) SDs for healthy persons. Insulin auto- varies with sex and age. GAD65A is associated with HLA
antibody binding has been noted not to be normally dis- DR3DQA1*0501DQB1*0201 in both patients and healthy
tributed. Each laboratory needs to assay at least 100200 individuals. IA-2As may be present in 40%50% of patients
healthy individuals to determine the distribution of binding. with newly diagnosed type 1 diabetes, but the frequency is
An important caveat concerning IAA measurement is that highest in the young. The frequency decreases with increas-
insulin antibodies develop after insulin therapy, even in per- ing age. IA-2As are associated with HLA DR4 DQA1*0301
sons who use human insulin. Data from the Diabetes Auto- DQB1*0302. IAA positivity occurs in >70%80% of children
antibody Standardization Program (DASP) demonstrate that who develop type 1 diabetes before 5 years of age but occurs
the interlaboratory imprecision for IAA is inappropriately in <40% of individuals who develop diabetes after the age of
large (309). 12 years. IAAs are associated with HLA DR4DQA1*0301
GAD65A and IA-2A are measured with standardized radio- DQB1*0302 and with INS VNTR (262). ICA is found in about
binding assays, which are performed with 35S-labeled recom- 75%85% of new-onset patients.
binant human GAD65 or IA-2 generated by coupled in vitro The ICA assay is labor-intensive and difficult to standard-
transcription translation with [35S]methionine or other 35S- or ize, and marked interlaboratory variation in sensitivity and
3
H-labeled amino acids (310). Commercially available methods specificity has been demonstrated in workshops (284, 314).
for GAD65A and IA-2A are available as a radioimmunoassay Few clinical laboratories are likely to implement this test.
with 125I-labeled GAD65 (truncated at the N-terminal end to The immunoassays are more reproducible and are amenable
promote solubility) and IA-2, respectively. In addition, immuno- to standardization (309). Measurement of T-cell reactivity in
assays without radio-label are commercially available for both peripheral blood is theoretically appealing, but the imprecision
GAD65A and IA-2A. Major efforts have been made to stan- of such as-says precludes their use from a clinical setting (315,
dardize GAD65A and IA-2A measurements (309, 311). A WHO 316). Autoantibody positivity (by definition) occurs in healthy
standard for both GAD65A and IA-2A has been established, individuals despite an absence of a family history of autoim-
and GAD65A and IA-2A amounts are expressed in international mune diseases. Islet cell autoantibodies are no exception. If one
units (312). The binding of labeled autoantigen to autoantibod- autoantibody is found, the others should be assayed, because
ies is normally distributed. Cutoff values should be determined the risk of type 1 diabetes increases if an individual tests posi-
from 100200 serum samples obtained from healthy individuals. tive for 2 or more autoantibodies (306).
GAD65A and IA-2A results should be reported as positive when The following suggestions (279) have been proposed as
the signal exceeds the 99th percentile. Comparison of multiple a rational approach to the use of autoantibodies in diabetes:
laboratories worldwide is carried out in the DASP, a proficiency- (a) Antibody assays should have a specificity >99%; (b) profi-
testing program organized by the CDC under the auspices of ciency testing should be documented; (c) multiple autoantibod-
the Immunology of Diabetes Society. That commercially avail- ies should be assayed; and (d) sequential measurement should
able GAD65A and IA-2A methods are also participating in the be performed. These strategies will reduce false-positive and
DASP program demonstrates that it should be possible not only false-negative results.
to harmonize participating laboratories but also eventually to
standardize GAD65A and IA-2A (311).
ICAs are measured by indirect immunofluorescence of
5. EMERGING CONSIDERATIONS
frozen sections of human pancreas (313). ICA assays mea-
sure the degree of immunoglobulin binding to islets, and
Immunoassays for IAA, GAD65A, IA-2A/IA-2A, and ZnT8A
results are compared with a WHO standard serum avail-
are now available, and a panel of these autoantibodies is cur-
able from the National Institute of Biological Standards and
rently used in screening studies (317). Because ICA assays are
Control (312). The results are reported in Juvenile Diabe-
difficult to standardize, their use has declined substantially.
tes Foundation (JDF) units. Positive results depend on the
It is likely that other islet cell antigens will be discovered, and
study or context in which they are used, but many laborato-
such discoveries could lead to additional diagnostic and predic-
ries use 10 JDF units measured on 2 separate occasions or
tive tests for type 1 diabetes. Autoantibody screening of dried
a single result 20 JDF units as titers that may indicate a
spots obtained from finger-stick blood samples appears quite
significantly increased risk of type 1 diabetes. The method
feasible in the future. For individuals who are positive for islet
is cumbersome and has proved difficult to standardize. The
cell autoantibodies, HLA-DR/HLA-DQ genotyping will help
number of laboratories that still carry out the ICA assay has
define the absolute risk of type 1 diabetes.
decreased markedly, and the test is no longer included in the
Several clinical trials to prevent or intervene with type
DASP program.
1 diabetes are being actively pursued (317). Such trials can
now be done with relatives of patients with type 1 diabetes
or in the general population on the basis of the islet cell auto-
4. INTERPRETATION antibody and HLA-DR/HLA-DQ genotype status. Risk can
be assessed by islet cell auto-antibodies alone, without the
GAD65A may be present in approximately 60%80% of need for evaluating endogenous insulin reserves, as was done
patients with newly diagnosed type 1 diabetes, but the frequency for the US DPT-1 trial (306). Rates of islet cell autoantibody
38 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

positivity are distinctly lower in the general population than increased risk. Phase II clinical trials with alum-formulated
in relatives of individuals with type 1 diabetes; consequently, GAD65 have reported no adverse events and some preserva-
trials with the latter group are more economical. Potential tion of endogenous insulin production in GAD65A-positive
interventional therapies (for type 1 diabetes) undergoing diabetes patients (319, 320). Additional trials of other anti-
clinical trials include oral insulin (317) or nasal insulin (318) gen-based immunotherapies, adjuvants, cytokines, and T-cell
given to nondiabetic (but islet cell autoantibody positive) accessory molecule blocking agents are likely in the future
relatives of individuals with type 1 diabetes or to children (270). Decreased islet cell autoimmunity will be one impor-
with islet cell auto-antibodies and HLA genotypes conferring tant outcome measure of these therapies.
Chapter 12
Albuminuria (formerly microalbuminuria)

Albuminuria (formerly microalbuminuria) are a well-estab- A. Diagnosis/screening. Diabetes is associated with a very high
lished cardiovascular risk marker, in which increases over time rate of cardiovascular events and is the leading cause of end-
to macroalbuminuria (>300 mg/day) are associated with kid- stage renal disease in the Western world (321). Early detection
ney disease and an increased risk for progression to end-stage of risk markers, such as albumin in the urine (formerly termed
renal disease. Annual testing for albuminuria is recommended microalbuminuria), relies on tests for urinary excretion of
by all major guidelines for patients with diabetes and/or kidney albumin. Conventional qualitative tests (chemical strips or
disease. To be useful, semiquantitative or qualitative screening dipsticks) for albuminuria do not detect the small increases
tests must be shown to be positive in >95% of patients with of urinary albumin excretion. For this purpose, tests to detect
albuminuria. Positive results of such tests must be confirmed albumin concentrations are used (Table 9) (322324). Low
by quantitative testing in an accredited laboratory. levels of albuminuria have been defined by the Joint National
Committee (JNC) 7 and the ADA and have more recently been
redefined by the Kidney Disease: Improving Global Outcomes
Committee (21, 325327) as excretion of 30300 mg of albu-
1. USE
min/24 h, 20200 g/min, or 30300 g/mg creatinine (Table
10) on 2 of 3 urine collections. Recent data, however, suggest
that risk extends below the lower limit of 20 g/min (328330),
Recommendation reinforcing the notion that this factor is a continuous variable
Annual testing for albuminuria in patients without clinical for cardiovascular risk (331333).
proteinuria should begin in pubertal or postpubertal individu- The JNC 7, the National Kidney Foundation (NKF), and
als 5 years after diagnosis of type 1 diabetes and at the time of the ADA all recommend the use of morning spot albumin/
diagnosis of type 2 diabetes, regardless of treatment. creatinine measurement for annual quantitative testing for
B (moderate) urine albumin in adults with diabetes (21, 326, 327). Indi-
viduals should be fasting. The optimal time for spot urine
collection is the early morning, but for minimizing varia-
Recommendation tion, all collections should be at the same time of day; the
individual preferably should not have ingested food for at
Urine albumin at concentrations 30 mg/g creatinine least 2 h (334).
should be considered a continuous risk marker for cardio- Positive test results represent albuminuria in these
vascular events. guidelines, corresponding to protein excretion of >300
B (moderate) mg/24 h, >200 g/min, or >300 mg/g creatinine (Table 10).
In these patients, quantitative measurement of urine albumin
excretion is used in assessing the severity of albuminuria and
Table 9. Review of assays to assess albuminuria. its progression, in planning treatment, and in determining
Detection the impact of therapy. To properly assess the stage of kidney
Method Interassay CV disease, the estimated glomerular filtration rate (eGFR) can
limit
Immunonephelometry 4.2% at 12.1 mg/L 2 mg/L be calculated from the serum creatinine value, age, sex, and
(Beckman Coulter Array race of the patient (335). An eGFR of <60 mL/min, regardless
analyzer) of the presence of low levels of albuminuria, is an indepen-
5.3% at 45 mg/L dent cardiovascular risk factor (325, 327). A urine albumin
Immunoturbidimetry 4.1% at 10.6 mg/L 6 mg/L value of <30 mg/g creatinine, although considered normal,
(Dade Behring should be reassessed annually, because values as low as 10
turbimeter)
mg/g creatinine have been associated in some studies with an
2.2% at 77.9 mg/L
Hemocue (point of care) 2.2% at 77.9 mg/L 5 mg/L increased cardiovascular risk. If the value is 30 mg/g cre-
4.3% at 82 mg/L atinine, changes should be reassessed after 6 to 12 months if
Radioimmunoassay 9.2% at 12.2 mg/dL 16 g/L antihypertensve therapy is required or annually in those who
4.8% at 33 mg/L are normotensive (326). For children with type 1 diabetes,

39
40 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

Table 10. Definitions of albuminuria.a

Test for microalbuminuria.
Unit of measure

mg/mg No
mg/24 h mg/min  for albumin?
Creatinine
Normal ,30 ,20 ,30 Yes
High albuminuria 30300 20200 30300
Condition that may invalidate
(formerly
urine albumin excretion?
microalbuminuria)
Very high albuminuriab .300 .200 .300 Yes No
a
From the ADA (21).
b
Also called overt nephropathy. No Treat and/or wait until
resolved. Repeat test.
 for protein?
testing for low levels of albuminuria is recommended to begin
after puberty and after a diabetes duration of 5 years. Of note Yes
is that most longitudinal cohort studies have reported signifi- Repeat microalbuminuria
cant increases in the prevalence of low levels of albuminuria test twice within 36 months.
only after diabetes has been present for 5 years (326, 336).
In the algorithms of both the NKF and the ADA for urine No
Rescreen
protein testing (321), the diagnosis of low levels of albuminuria 2 of 3 tests positive?
in 1 year.
requires both the demonstration of increased albumin excretion Yes
(as defined above) on 2 of 3 tests repeated at intervals of 3 to
6 months and the exclusion of conditions that invalidate the Microalbuminuria; begin treatment.
test (Fig. 2).
B. Prognosis. Albuminuria values >30 mg/g creatinine Fig. 2. Algorithm for urine protein testing.
[and lower values if the eGFR is <60 mL/min (Table
10)], has prognostic significance. Multiple epidemio-logic
studies have shown it to be an independent risk marker for dietary protein restriction, and therapy with blockers of the
cardiovascular death (325, 337, 338). In 80% of patients renin angiotensin system (321). Several factors are known
with type 1 diabetes and low levels of albuminuria, urinary to slow the rate of urinary albumin excretion or to prevent
albumin excretion can increase by as much as 10%20%/ its development. They include reducing blood pressure
year, with the development of clinical proteinuria (>300 mg (with a blocker of the renin angiotensin system as part of
albumin/day) in 1015 years in more than half the patients. the regimen), glycemic control, and lipid-lowering therapy
After clinical-grade proteinuria occurs, .90% of patients (45, 343345).
develop a decreased GFR and, ultimately, end-stage renal
disease. In type 2 diabetes, 20%40% of patients with stage
A2 albuminuria (Table 10) progress to overt nephropathy,
2. RATIONALE
but by 20 years after overt nephropathy, approximately 20%
develop end-stage renal disease. In addition, patients with
Early detection of albuminuria allows early intervention with
diabetes (type 1 or type 2) and stage A2 albuminuria are at
the goal of reducing cardiovascular risk and delaying the onset
increased risk for cardiovascular disease. Of note is that low
of overt diabetic nephropathy. Thus, it is an indicator of the
levels of albuminuria alone indicate neither an increased
need for more intensive efforts to reduce cardiovascular risk
risk for progression to end-stage kidney disease nor kidney
factors.
disease per se; hypertension needs to be present for the risk
Albuminuria (stage A2) rarely occurs with a short duration
of progression (339, 340). Moreover, about 20% of people
of type 1 diabetes or before puberty. Thus, testing is less urgent
progress to end-stage kidney disease without an increase in
in these situations. Nevertheless, the difficulty in precisely dat-
low levels of albuminuria (341). Another factor that indi-
ing the onset of type 2 diabetes warrants initiation of annual
cates progression is an increase in albuminuria from stage
testing at the time of diagnosis of diabetes. Although older
A2 to A3 over time despite achievement of blood pressure
patients (age >75 years or a life expectancy <20 years) may
goals (342).
not be at risk for clinically significant nephropathy because of
C. Monitoring. The roles of routine urinalysis and albu- a short projected life span, they will be at higher cardiovas-
min measurements are less clear in patients with stage A2 cular risk. In such patients, the role of treating albuminuria is
albuminuria. Some experts have advocated urine protein far from clear. Published studies have demonstrated that it is
testing to monitor treatment, which may include improved cost-effective to screen all patients with diabetes and/or kidney
glycemic control, more assiduous control of hypertension, disease for albumin-uria (346, 347).
Albuminuria (formerly Microalbuminuria) 41

3. ANALYTICAL CONSIDERATIONS Qualitative (or semiquantitative) assays have been pro-

posed as screening tests for low levels of albuminuria. To be
useful, screening tests must have high detection rates, i.e., a
Recommendation high clinical sensitivity. Although many studies have assessed
the ability of reagent strips (dipstick methods) to detect
The analytical cv of methods to measure low levels of albu-
increased albumin concentrations in urine, the important ques-
minuria should be <15%.
tion is whether the method can detect low levels of albuminuria,
B (moderate)
that is, an increased albumin excretion rate or its surrogate, an
increased albumin creatinine ratio. We can find no documen-
A. Analytical. Analytical goals can be related to the degree tation of any test in which the sensitivity for detection of an
of biological variation, with less precision required for ana- increased albumin excretion rate consistently reached 95% in
lytes that vary widely. Detection limits and imprecision data >1 study. For example, in a large study (352), the sensitivity for
are summarized in Table 9. Commercially available quantita- detection of an albumin excretion rate >30 mg/24 h was 91%
tive methods for low levels of albuminuria have documented when the test was performed by a single laboratory technician,
detection limits of approximately 20 g/L or less. Within- 86% when performed by nurses, and 66% when performed by
run imprecision and day-to-day (total) imprecision are well general practitioners. In 2 subsequent studies (353, 354), the
within the analytical goal of approximately 15% and are often sensitivities were 67% 86%. False-positive results also appear
considerably less. Most, but not all, methods agree well and to be common, with rates as high as 15% (352). Thus, it appears
support a reference interval of 220 g albumin/mg creati- that at least some of the tests, especially as used in practice,
nine (348). have the wrong characteristics for screening because of low
The within-person variation in albumin excretion is large sensitivity (high false-negative rates), and positive results must
in people without diabetes and is even higher in patients with be confirmed by a laboratory method. Of the available meth-
diabetes. Howey et al. (349) studied dayto-day variation, over ods, the immunoturbidimetric assay is the most reliable and
34 weeks, in the 24-h albumin excretion, the concentration should be considered the standard for comparison, because it
of albumin, and the albumincreatinine ratio. The last 2 vari- has >95% sensitivity and specificity to detect very low levels
ables were measured in the 24-h urine sample, the first morn- of albuminuria. Semiquantitative or qualitative screening tests
ing void, and random untimed urine collections. In healthy should be positive in >95% of patients for the detection of
volunteers, the lowest within-person CVs were obtained for albuminuria to be useful for assessment of cardiovascular risk
the concentration of albumin in the first morning void (36%) and progression of kidney disease. Positive results obtained
and for the albumincreatinine ratio in that sample (31%) with such methodologies must be confirmed by an immunotur-
(349). Multiple studies have evaluated the best procedure bidimetric assay in an accredited laboratory (355).
to assess albuminuria. Most studies have found that the spot
urine albumin creatinine concentration in the first morning Recommendation
void, rather than the 24-h urinary excretion of albumin or the
timed collection, is the most practical and reliable technique Currently available dipstick tests do not have adequate ana-
(346, 350, 351). lytical sensitivity to detect low levels of albuminuria.
To keep the analytical CV less than half the biological B (moderate)
CV, an analytical goal of an 18% CV has been proposed (349).
Alternatively, if the albumincreatinine ratio is to be used, one
may calculate the need for a somewhat lower imprecision (that Chemical-strip methods are not sensitive when the albumin
is, a better precision) to accommodate the lower biological CV concentration in the urine is in the interval of 20 50 mg/L. Thus,
for the ratio and the imprecision contributed by the creatinine no recommendation can be made for the use of any specific
measurement. Assuming a CV of 5% for creatinine measure- screening test. Dipstick tests for low levels of albuminuria can-
ment, we calculate a goal of 14.7% for the analytical CV for not be recommended as a replacement for the quantitative tests.
albumin when it is used to estimate the albumincreatinine The available dipstick methods to detect low levels of
ratio. A goal of 15% appears reasonable to accommodate use albuminuria do not appear to lend themselves to viable screen-
of the measured albumin concentration for calculating either ing strategies, either in the physicians office or for home
the timed excretion rate or the albumincreatinine ratio. testing. Usual screening tests (e.g., for phenylketonuria) have
low false-negative rates, and thus only positive results require
confirmation by a quantitative method. If a screening test has
Recommendation low sensitivity, negative results also must be confirmed, a com-
Semiquantitative or qualitative screening tests should be pos- pletely untenable approach. With semiquantitative tests, it may
itive in >95% of patients with low levels of albuminuria to be be possible (or indeed necessary) to use a cutoff <20 mg/L to
useful for screening. Positive results must be confirmed by ensure the detection of samples with albumin values >20 mg/L
analysis in an accredited laboratory. as measured by laboratory methods.
GPP Recent studies have compared selected dipstick methods to
laboratory assays. One dipstick was found to have >95% sen-
42 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

sitivity (322, 324). One such study evaluated an office-screen- analyses of clinical trials, however, have found that the albu-
ing test that uses a monoclonal antibody against human serum min creatinine ratio is a reasonable method to assess change
albumin (Immuno-Dip; Genzyme Diagnostics) (322). Screening over time (357). For screening, an untimed sample for albu-
182 patient samples with this method with an albumin cre- min measurement (without creatinine) may be considered if a
atinine ratio of 30 g/mg as positive yielded a sensitivity of one uses a concentration cutoff that allows high sensitivity for
96%, a specificity of 80%, a positive predictive value of 66%, detecting an increased albumin-excretion rate.
and a negative predictive value of 98%. In a separate study, 165 Albumin is stable in untreated urine stored at 4 C or 20 C
patients had the HemoCue point-of-care system for albumin for at least a week (358). Neither centrifugation nor filtration
compared with the Clinitek Microalbumin (Siemens) and Chem- appears necessary before storage at -20 C or -80 C (359).
strip Micral (Roche Diagnostics) tests, as well as with an HPLC Whether a urine sample is centrifuged, filtered, or not treated,
assay, for spot albumincreatinine ratio measurement (324). the albumin concentration decreases by 0.27%/day at -20 C but
Further studies are needed before the dipstick tests for low lev- shows no decreases over 160 days at -80 C (359). The urinary
els of albuminuria can be recommended as replacements for the albumin excretion rate does not show marked diurnal variation
quantitative tests. The use of qualitative tests at the point of care in diabetes but does so in essential hypertension (360).
is reasonable only when it can be shown that this approach elimi-
nates quantitative testing in a sizeable proportion of patients and
detects those patients who have early renal disease.
4. INTERPRETATION
Recommendation A. Nonanalytical sources of variation. Transient increases in uri-
nary albumin excretion have been reported with short-term hyper-
Acceptable samples to test for increased urinary albumin
glycemia, exercise, uri-nary tract infections, marked hypertension,
excretion are timed collections (e.g., 12 or 24 h) for mea-
heart failure, acute febrile illness, and hyperlipidemia (321).
surement of the albumin concentration and timed or untimed
samples for measurement of the albumincreatinine ratio.
B (moderate). Recommendation
Low urine albumin concentrations (i.e., <30 mg/g creatinine)
are not associated with high cardiovascular risk if the eGFR
Recommendation
is >60 ml min21 (1.73 m2)21 and the patient is normo-
The optimal time for spot urine collection is the early morn- tensive. If the eGFR is <60 ml min21 (1.73 m2)21 and/
ing. All collections should be at the same time of day to mini- or the level of albuminuria is >30 mg/g creatinine on a spot
mize variation. The patient should not have ingested food urine sample, a repeat measurement should be taken within
within the preceding 2 h but should be well hydrated (i.e., the year to assess change among people with hypertension.
Not volume depleted). A (moderate)
GPP
B. Frequency of measurement. The NKF, the ADA, and JNC 7
B. Preanalytical. Collection of 24-h samples has disadvantages, recommend annual measurement in diabetic patients with albu-
specifically because many samples are collected inadequately mincreatinine ratios <30 g/mg. After the documentation of
and because total creatinine is not routinely checked to evalu- stage A2 albuminuria (i.e., with results as defined above on 2 of 3
ate the adequacy of collection. The albumin creatinine ratio tests performed within 3 to 6 months), repeated testing is reason-
is the superior method to predict renal events in patients with able to determine whether a chosen therapy is effective. It may
type 2 diabetes (356). The ratio has a within-person biologi- also be useful in determining the rate of disease progression and
cal variation similar to that of the excretion rate and correlates thus may support planning for care of end-stage renal disease.
well with both timed excretion and the albumin concentra- Although the ADA recommendations suggest that such testing
tion in a first morning void of urine (349). For the ratio, a first is not generally needed before puberty, testing may be consid-
morning void sample is preferable because this sample has a ered on an individual basis if it appears appropriate because of
lower within-person variation than the ratio for a random urine an early onset of diabetes, poor control, or a family history of
sample taken during the day (349). Although the ratio appears diabetic nephropathy. The duration of diabetes prior to puberty
entirely acceptable for screening, limited data are available on is reportedly an important risk factor in this age group and thus
its use in monitoring the response to therapy. Recent post hoc can be used to support such testing in individual patients (361).
Chapter 13
Miscellaneous Potentially Important Analytes

MISCELLANEOUS POTENTIALLY IMPOR hyperinsulinemia would also appear to be a logical risk pre-
dictor for incident type 2 diabetes.
TANT ANALYTES. I. INSULIN AND Earlier studies may not have controlled well for glyce-
PRECURSORS mic status and other confounders. More-recent analyses sug-
gest that insulin values do not add significantly to diabetes risk
1. USE prediction carried out with more traditional clinical and labo-
ratory measurements (366) and that measures of insulin resis-
Recommendation tance (that include insulin measurements) predict the risk of
diabetes or coronary artery disease only moderately well, with
There is no role for routine testing for insulin, C-peptide, no threshold effects (367). Consequently, it seems of greater
or proinsulin in most patients with diabetes. Differentiation clinical importance to quantify the consequences of the insu-
between type 1 and type 2 diabetes may be made in most lin resistance and hyperinsulinemia (or hyperproinsulinemia)
cases on the basis of the clinical presentation and the subse- rather than the hormone values themselves, i.e., by measuring
quent course. These assays are useful primarily for research blood pressure, the degree of glucose tolerance, and plasma
purposes. Occasionally, C-peptide measurements may help lipid/lipoprotein concentrations. It is these variables that are
distinguish type 1 from type 2 diabetes in ambiguous cases, the focus of clinical interventions, not plasma insulin or proin-
such as patients who have a type 2 phenotype but present in sulin concentrations (366, 367).
ketoacidosis. The clinical utility of measuring insulin, C-peptide, or pro-
B (moderate) insulin concentrations to help select the best antihyperglycemic
agent for initial therapy in patients with type 2 diabetes is a
question that arises from consideration of the pathophysiology
Recommendation of type 2 diabetes. In theory, the lower the pretreatment insu-
There is no role for measurement of insulin concentration in lin concentration, the more appropriate might be insulin, or an
the assessment of cardiometabolic risk, because knowledge insulin secretagogue, as the drug of choice to initiate treatment.
of this value does not alter the management of these patients. Although this line of reasoning may have some intellectual
B (moderate) appeal, there is no evidence that measurement of plasma insu-
lin or proinsulin concentrations will lead to more efficacious
treatment of patients with type 2 diabetes.
A. Diagnosis. In the last several years, interest has increased In contrast to the above considerations, measurement of
in the possibility that measurements of the concentrations of plasma insulin and proinsulin concentrations is necessary to
plasma insulin and its precursors might be of clinical benefit. establish the pathogenesis of fasting hypoglycemia (368). The
In particular, published evidence reveals that increased con- diagnosis of an islet cell tumor is based on the persistence of
centrations of insulin and/or proinsulin in nondiabetic indi- inappropriately increased plasma insulin concentrations in the
viduals predict the development of coronary artery disease face of a low glucose concentration. In addition, an increase in
(362). Although this possibility may be scientifically valid, its the ratio of fasting pro-insulin to insulin in patients with hypo-
clinical value is questionable. An increased insulin concentra- glycemia strongly suggests the presence of an islet cell tumor.
tion is a surrogate marker that can be used to estimate resis- The absence of these associated changes in glucose, insulin,
tance to insulin-mediated glucose disposal, and it can identify and proinsulin concentrations in an individual with fasting
individuals at risk for developing syndrome X, also known as hypoglycemia makes the diagnosis of an islet cell tumor most
the insulin resistance syndrome or the metabolic syndrome unlikely, and alternative explanations should be sought for the
(363). Accurate measurement of insulin sensitivity requires inability to maintain fasting euglycemia.
the use of complex methods, such as the hyperinsulinemic Measurement of the C-peptide response to intravenous
euglycemic clamp technique, which are generally confined to glucagon can aid in instances in which it is difficult to differ-
research laboratories (364, 365). Because of the critical role entiate between the diagnosis of type 1 and type 2 diabetes (5).
of insulin resistance in the pathogenesis of type 2 diabetes, Even in this clinical situation, however, the response to drug

43
44 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

therapy will provide useful information, and measurement of lish its own reference interval. Although it has been suggested
C-peptide may not be clinically necessary. Measurementof by some, insulin measurement should not be used in an OGTT
C-peptide is essential in the investigation of possible factitious to diagnose diabetes. In the case of C-peptide, there is a discrep-
hypoglycemiadue to surreptitious insulin administration (369). ancy in reliability because of variable specificity among antisera,
In the past, some advocated insulin assays in the evalua- lack of standardization of C-peptide calibration, and variable
tion and management of patients with the polycystic ovary syn- cross-reactivity with proinsulin. Of note is the requirement of
drome. Women with this syndrome manifest insulin resistance the US Centers for Medicare and Medicaid Services that Medi-
by androgen excess, as well as by abnormalities of carbohydrate care patients have C-peptide measured in order to be eligible for
metabolism; both abnormalities may respond to treatment with coverage of insulin pumps. Initially, the requirement was that
metformin or thiazolidinediones. Although clinical trials have the C-peptide concentration be 0.5 ng/mL; however, because
generally evaluated insulin resistance by using the hyperinsu- of the noncomparability of results from different assays, which
linemic euglycemic clamp, ratios of fasting glucose to insulin, led to denial of payment for some patients with values >0.5 ng/
and other modalities, the optimal laboratory evaluation of these mL, the requirement now states that the C-peptide concentration
patients in routine clinical care has not been clearly defined. It is should be 110% of the lower limit of the reference interval of
unclear whether assessing insulin resistance through insulin mea- the laboratorys measurement method (376).
surement has any advantage over assessment of physical signs of
insulin resistance (body mass index, presence of acanthosis nigri-
cans), and routine measurements of insulin are not recommended
by the American College of Obstetrics and Gynecology (370). Miscellaneous Potentially Impor-
tant Analytes. II. Insulin Antibodies

2. ANALYTICAL CONSIDERATIONS Recommendation

There is no published evidence to support the use of insulin
antibody testing for routine care of patients with diabetes.
Recommendation
C (very low)
Because current measures of insulin are poorly harmonized, a
standardized insulin assay should be developed to encourage
Given sufficiently sensitive techniques, insulin antibodies can
the development of measures of insulin sensitivity that will
be detected in any patient being treated with exogenous insulin
be practical for clinical care.
(371). In the vast majority of patients, the titer of insulin antibod-
GPP
ies is low, and their presence is of no clinical significance. Very
low values are seen in patients treated exclusively with human
Although it has been assayed for >40 years, there is no standard- recombinant insulin (377). On occasion, however, the titer of
ized method available to measure serum insulin (371). Attempts insulin antibodies in the circulation can be quite high and associ-
to harmonize insulin assays with commercial insulin reagent ated with a dramatic resistance to the ability of exogenous insulin
sets have produced greatly discordant results (372). Recently, to lower plasma glucose concentrations. This clinical situation is
an insulin standardization workgroup of the ADA, in conjunc- quite rare, it usually occurs in insulin-treated patients with type
tion with the National Institute of Diabetes and Digestive and 2 diabetes, and the cause-and-effect relationships between the
Kidney Diseases, the CDC, and European Association for the magnitude of the increase in insulin antibodies and the degree
Study of Diabetes, called for harmonization of insulin assay of insulin resistance are unclear. There are several therapeutic
results through traceability to an isotope-dilution liquid chro- approaches for treating these patients, and a quantitative esti-
matographytandom mass spec-trometry reference (373). The mate of the concentration of circulating insulin antibodies does
Insulin Standardization Workgroup called for harmonization of not appear to be of significant benefit.
the insulin assay to encourage the development of measures of The prior version of these guidelines (14) contained short
insulin sensitivity and secretion that will be practical for clini- sections on amylin and leptin, both of which were the focus of
cal care (374). Analogous to insulin, considerable imprecision active clinical studies. The evidence that has accumulated in
among laboratories has also been observed for measurement of the last 7 to 8 years has failed to identify any clinical value in
C-peptide. A comparison of 15 laboratories that used 9 different measuring these analytes in patients with diabetes. Similarly,
routine C-peptide assay methods, found within- and between- although cardiovascular disease is the major cause of mortality
run CVs as high as >10% and 18%, respectively (375). A com- for persons with diabetes, no evidence supports the measure-
mittee has been established under the auspices of the CDC to ment of nontraditional cardiovascular risk factors for routine
harmonize C-peptide analysis. assessment of risk in patients with diabetes. These sections
Measurement of proinsulin and C-peptide are accomplished have, therefore, been removed.
by immunometric methods. Proinsulin reference intervals are This Guideline is being simultaneously published in Clini-
dependent on methodology, and each laboratory should estab- cal Chemistry, Diabetes Care, and by the NACB.
REFERENCES

1. ADA. Report of the Expert Committee on the Diagnosis and Clas- cated haemoglobin and fasting and two hour plasma glucose concen-
sification of Diabetes Mellitus. Diabetes Care 1997;20:118397. trations as diagnostic methods for diabetes. BMJ 1994;308:13238.
2. Castano L, Eisenbarth GS. Type-I diabetes: a chronic autoim- 19. WHO. Definition and diagnosis of diabetes mellitus and inter-
mune disease of human, mouse, and rat. Annu Rev Immunol mediate hyperglycemia: report of a WHO/IDF consultation.
1990;8:64779. Geneva: WHO; 2006.
3. Reaven GM. Banting Lecture 1988. Role of insulin resistance in 20. The International Expert Committee. International Expert Com-
human disease. Diabetes 1988;37:1595607. mittee report on the role of the A1c assay in the diagnosis of dia-
4. Sacks DB, McDonald JM. The pathogenesis of type II diabetes betes. Diabetes Care 2009;32:132734.
mellitus. A polygenic disease. Am J Clin Pathol 1996;105:149 21. ADA. Standards of medical care in diabetes 2010. Diabetes
56. Care 2010;33(Suppl 1):S1161.
5. Balasubramanyam A, Garza G, Rodriguez L, Hampe CS, Gaur 22. ADA. Tests of glycemia in diabetes. Diabetes Care 2001;24(Suppl
L, Lernmark A, Maldonado MR. Accuracy and predictive value 1):S802.
of classification schemes for ketosis-prone diabetes. Diabetes 23. IDF. Task Force. Global guideline for type 2 diabetes. Brussels:
Care 2006;29:25759. IDF; 2005. p 111.
6. IDF. Diabetes atlas. 3rd ed. Brussels: IDF; 2008. 24. Saudek CD, Herman WH, Sacks DB, Bergenstal RM, Edelman
7. Cowie CC, Rust KF, Byrd-Holt DD, Eberhardt MS, Flegal D, Davidson MB. A new look at screening and diagnosing dia-
KM, Engelgau MM, et al. Prevalence of diabetes and impaired betes mellitus. J Clin Endocrinol Metab 2008;93:244753.
fasting glucose in adults in the U.S. population: National Health 25. ADA. Type 2 diabetes in children and adolescents. Diabetes
and Nutrition Examination Survey 19992002. Diabetes Care Care 2000;23:3819.
2006;29:12638. 26. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF,
8. Cowie CC, Rust KF, Ford ES, Eberhardt MS, Byrd-Holt DD, Lachin JM, Walker EA, Nathan DM. Reduction in the incidence
Li C, et al. Full accounting of diabetes and pre-diabetes in the of type 2 diabetes with lifestyle intervention or metformin.
U.S. population in 19881994 and 20052006. Diabetes Care N Engl J Med 2002;346:393403.
2009;32:28794. 9. Chan JC, Malik V, Jia W, Kadowaki T, 27. Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen
Yajnik CS, Yoon KH, Hu FB. Diabetes in Asia: epidemiology, H, Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus
risk factors, and pathophysiology. JAMA 2009; 301:212940. by changes in lifestyle among subjects with impaired glucose
10. Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J, et al. Prevalence tolerance. N Engl J Med 2001;344:134350.
of diabetes among men and women in China. N Engl J Med 28. Icks A, Rathmann W, Haastert B, John J, Lwel H, Holle R,
2010;362:1090101. Giani G. Cost-effectiveness of type 2 diabetes screening: results
11. ADA. Economic costs of diabetes in the U.S. in 2007. Diabetes from recently published studies. Gesundheitswesen 2005;
Care 2008;31:596615. 67(Suppl 1):S16771.
12. Nathan DM. Long-term complications of diabetes mellitus. N 29. Perry RC, Shankar RR, Fineberg N, McGill J, Baron AD. HbA1c
Engl J Med 1993;328:167685. measurement improves the detection of type 2 diabetes in high-
13. Roglic G, Unwin N, Bennett PH, Mathers C, Tuomilehto J, Nag risk individuals with nondiagnostic levels of fasting plasma glu-
S, et al. The burden of mortality attributable to diabetes: realistic cose: the Early Diabetes Intervention Program (EDIP). Diabetes
estimates for the year 2000. Diabetes Care 2005;28: 21305. Care 2001;24:46571.
14. Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald 30. Jesudason DR, Dunstan K, Leong D, Wittert GA. Macrovascular
JM, Parrott M. Guidelines and recommendations for laboratory risk and diagnostic criteria for type 2 diabetes: implications for
analysis in the diagnosis and management of diabetes mellitus. the use of FPG and HbA1c for cost-effective screening. Diabetes
Clin Chem 2002;48:43672. Care 2003;26:48590.
15. National Diabetes Data Group. Classification and diagnosis of 31. Dallo FJ, Weller SC. Effectiveness of diabetes mellitus screening
diabetes mellitus and other categories of glucose intolerance. recommendations. Proc Natl Acad Sci U S A 2003;100:105749.
Diabetes 1979;28:103957. 32. Diabetes Cost-Effectiveness Study Group, Centers for Disease
16. WHO. WHO Expert Committee on Diabetes Mellitus: second Control and Prevention. The cost-effectiveness of screening for
report. World Health Organ Tech Rep Ser 1980;646:180. type 2 diabetes. JAMA 1998;280:175763.
17. Engelgau MM, Thompson TJ, Herman WH, Boyle JP, Aubert 33. Hoerger TJ, Harris R, Hicks KA, Donahue K, Sorensen S,
RE, Kenny SJ, et al. Comparison of fasting and 2-hour glucose Engelgau M. Screening for type 2 diabetes mellitus: a cost-
and HbA1c levels for diagnosing diabetes. Diagnostic criteria effectiveness analysis. Ann Intern Med 2004;140:68999.
and performance revisited. Diabetes Care 1997;20:78591.18. 34. Glumer C, Yuyun M, Griffin S, Farewell D, Spiegelhalter D,
McCance DR, Hanson RL, Charles MA, Jacob-sson LT, Kinmonth AL, Wareham NJ. What determines the cost-effec-
Pettitt DJ, Bennett PH, Knowler WC. Comparison of tests for gly- tiveness of diabetes screening? Diabetologia 2006;49:153644.

45
46 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

35. Kahn R, Alperin P, Eddy D, Borch-Johnsen K, Buse J, Feigel- 52. Duckworth W, Abraira C, Moritz T, Reda D, Emanuele N,
man J, et al. Age at initiation and frequency of screening to Reaven PD, et al. Glucose control and vascular complications in
detect type 2 diabetes: a cost-effectiveness analysis. Lancet veterans with type 2 diabetes. N Engl J Med 2009;360:129 39.
2010;375:136574. 53. Berg AH, Sacks DB. Haemoglobin A1c analysis in the manage-
36. Greenberg RA, Sacks DB. Screening for diabetes: is it war- ment of patients with diabetes: from chaos to harmony. J Clin
ranted? Clin Chim Acta 2002;315:619. Pathol 2008;61: 9837.
37. Genuth S, Alberti KG, Bennett P, Buse J, Defronzo R, Kahn R, 54. Howe-Davies S, Simpson RW, Turner RC. Control of maturity-
et al. Follow-up report on the diagnosis of diabetes mellitus. onset diabetes by monitoring fasting blood glucose and body
Diabetes Care 2003;26:31607. weight. Diabetes Care 1980;3:60710.
38. Forouhi NG, Balkau B, Borch-Johnsen K, Dekker J, Glumer C, 55. Muir A, Howe-Davies SA, Turner RC. General practice care of
Qiao Q, et al. The threshold for diagnosing impaired fasting glu- non-insulin-dependent diabetes with fasting blood glucose mea-
cose: a position statement by the European Diabetes Epidemiol- surements. Am J Med 1982;73:63740.
ogy Group. Diabetologia 2006;49:8227. 56. Harris MI. Undiagnosed NIDDM: clinical and public health
39. Tai ES, Goh SY, Lee JJ, Wong MS, Heng D, Hughes K, et al. issues. Diabetes Care 1993;16: 64252.
Lowering the criterion for im- paired fasting glucose: impact on 57. Troisi RJ, Cowie CC, Harris MI. Diurnal variation in fast-
disease prevalence and associated risk of diabetes and ischemic ing plasma glucose: implications for diagnosis of diabetes in
heart disease. Diabetes Care 2004;27: 172834. patients examined in the afternoon. JAMA 2000;284:31579.
40. Gabir MM, Hanson RL, Dabelea D, Imperatore G, Roumain J, 58. Bruns DE, Knowler WC. Stabilization of glucose in blood sam-
Bennett PH, Knowler WC. The 1997 American Diabetes Asso- ples: why it matters. Clin Chem 2009;55:8502.
ciation and 1999 World Health Organization criteria for hyper- 59. Chan AY, Swaminathan R, Cockram CS. Effectiveness of
glycemia in the diagnosis and prediction of diabetes. Diabetes sodium fluoride as a preservative of glucose in blood. Clin Chem
Care 2000;23:110812. 1989;35:3157.
41. Tirosh A, Shai I, Tekes-Manova D, Israeli E, Pereg D, Shochat T, 60. Ladenson JH. Nonanalytical sources of variation in clinical
et al. Normal fasting plasma glucose levels and type 2 diabetes chemistry results. In: Sonnenwirth A, Jarett L, eds. Clinical
in young men. N Engl J Med 2005;353:145462. laboratory methods and diagnosis. St. Louis: C.V. Mosby Co.;
42. DCCT. The relationship of glycemic exposure (HbA1c) to the 1980. p 14992.
risk of development and progression of retinopathy in the diabe- 61. Sacks DB. Carbohydrates. In: Burtis CA, Ash-wood ER, Bruns
tes control and complications trial. Diabetes 1995;44:96883. DE, eds. Tietz textbook of clinical chemistry and molecular
43. Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull diagnostics. 4th ed. St. Louis: Elsevier Saunders; 2006. p 837
CA, et al. Association of glycaemia with macrovascular and 902.
microvascular complications of type 2 diabetes (UKPDS 35): 62. Gambino R, Piscitelli J, Ackattupathil TA, Theriault JL, Andrin
prospective observational study. BMJ 2000;321: 40512. RD, Sanfilippo ML, Etienne M. Acidification of blood is supe-
44. DCCT. The effect of intensive treatment of diabetes on the devel- rior to sodium fluoride alone as an inhibitor of glycolysis. Clin
opment and progression of long-term complications in insulin- Chem 2009;55:101921.
dependent diabetes mellitus. N Engl J Med 1993;329:97786. 63. Ladenson JH, Tsai LM, Michael JM, Kessler G, Joist JH. Serum
45. Nathan DM, Cleary PA, Backlund JY, Genuth SM, Lachin JM, versus heparinized plasma for eighteen common chemistry tests:
Orchard TJ, et al. Intensive diabetes treatment and cardiovas- Is serum the appropriate specimen? Am J Clin Pathol 1974;
cular disease in patients with type 1 diabetes. N Engl J Med 62:54552.
2005;353:264353. 64. Stahl M, Jorgensen LG, Hyltoft Petersen P, Brandslund I, de
46. UKPDS Group. Intensive blood-glucose control with sulpho- Fine Olivarius N, Borch-Johnsen K. Optimization of preana-
nylureas or insulin compared with conventional treatment and lytical conditions and analysis of plasma glucose. 1. Impact
risk of complications in patients with type 2 diabetes (UKPDS of the new WHO and ADA recommendations on diagnosis of
33). Lancet 1998;352:83753. diabetes mellitus. Scand J Clin Lab Invest 2001;61:16979.
47. Selvin E, Marinopoulos S, Berkenblit G, Rami T, Brancati FL, 65. Carstensen B, Lindstrom J, Sundvall J, Borch-Johnsen K,
Powe NR, Golden SH. Meta-analysis: glycosylated hemoglobin Tuomilehto J. Measurement of blood glucose: comparison
and cardiovascular disease in diabetes mellitus. Ann Intern Med between different types of specimens. Ann Clin Biochem
2004;141:42131. 2008;45:1408.
48. Ray KK, Seshasai SR, Wijesuriya S, Sivakumaran R, Nethercott 66. Boyanton BL Jr, Blick KE. Stability studies of twenty-four ana-
S, Preiss D, et al. Effect of intensive control of glucose on car- lytes in human plasma and serum. Clin Chem 2002;48:22427.
diovascular outcomes and death in patients with diabetes mel- 67. Miles RR, Roberts RF, Putnam AR, Roberts WL. Comparison
litus: a meta-analysis of randomised controlled trials. Lancet of serum and heparinized plasma samples for measurement of
2009;373:176572. chemistry analytes. Clin Chem 2004;50:17045.
49. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 68. Larsson-Cohn U. Differences between capillary and venous
10-year follow-up of intensive glucose control in type 2 diabe- blood glucose during oral glucose tolerance tests. Scand J Clin
tes. N Engl J Med 2008; 359:157789. Lab Invest 1976; 36:8058.
50. Gerstein HC, Miller ME, Byington RP, Goff DC Jr, Bigger JT, 69. Lind T, de Groot HA, Brown G, Cheyne GA. Observa-
Buse JB, et al. Effects of intensive glucose lowering in type 2 tions on blood glucose and insulin determinations. Br Med J
diabetes. N Engl J Med 2008;358:254559. 1972;3:3203.
51. Patel A, MacMahon S, Chalmers J, Neal B, Billot L, Wood- 70 Sacks DB. Carbohydrates. In: Burtis CA, Ash-wood ER, Bruns
ward M, et al. Intensive blood glucose control and vascular DE, eds. Tietz textbook of clinical chemistry and molecular
outcomes in patients with type 2 diabetes. N Engl J Med diagnostics. 5th ed. St. Louis: Elsevier Saunders. Accepted;
2008;358:256072. forthcoming 2012.
References 47

71. Tchobroutsky G. Blood glucose levels in diabetic and non-dia- 89. ADA. Standards of medical care in diabetes 2009. Diabetes
betic subjects. Diabetologia 1991; 34:6773. Care 2009;32(Suppl 1):S1361.
72. Blunt BA, Barrett-Connor E, Wingard DL. Evaluation of fasting 90. Harris MI, Cowie CC, Howie LJ. Self-monitoring of blood glu-
plasma glucose as screening test for NIDDM in older adults. cose by adults with diabetes in the United States population.
Rancho Bernardo Study. Diabetes Care 1991;14:98993. Diabetes Care 1993; 16:111623.
73. DECODE. Consequences of the new diagnostic criteria for 91. CDC. Self-monitoring of blood glucose among adults with dia-
diabetes in older men and women. DECODE Study (Diabetes betesUnited States, 1997 2006. MMWR Morb Mortal Wkly
Epidemiology: Collaborative Analysis of Diagnostic Criteria in Rep 2007;56: 11337.
Europe). Diabetes Care 1999;22:166771. 92. ADA. Self-monitoring of blood glucose. Diabetes Care
74. Ferrannini E, Vichi S, Beck-Nielsen H, Laakso M, Paolisso G, 1996;19(Suppl 1):S626.
Smith U. Insulin action and age. European Group for the Study 93. ADA. Standards of medical care in diabetes 2011. Diabetes
of Insulin Resistance (EGIR). Diabetes 1996;45:94753. Care 2011;34(Suppl 1):S1161.
75. Imbeault P, Prins JB, Stolic M, Russell AW, OMoore-Sullivan 94. Ipp E, Aquino RL, Christenson P. Point: Self-monitoring of
T, Despres JP, et al. Aging per se does not influence glucose blood glucose in type 2 diabetic patients not receiving insulin:
homeostasis: in vivo and in vitro evidence. Diabetes Care 2003; the sanguine approach. Diabetes Care 2005;28:152830.
26:4804. 95. Davidson MB. Counterpoint: Self-monitoring of blood glu-
76. Miller WG, Myers GL, Ashwood ER, Killeen AA, Wang E, cose in type 2 diabetic patients not receiving insulin: a waste of
Ehlers GW, et al. State of the art in trueness and interlabora- money. Diabetes Care 2005;28:15313.
tory harmonization for 10 analytes in general clinical chemistry. 96. Faas A, Schellevis FG, Van Eijk JT. The efficacy of self-mon-
Arch Pathol Lab Med 2008;132:83846. itoring of blood glucose in NIDDM subjects. A criteria-based
77. Fraser CG, Petersen PH. Analytical performance characteristics literature review. Diabetes Care 1997;20:14826.
should be judged against objective quality specifications. Clin 97. Coster S, Gulliford MC, Seed PT, Powrie JK, Swaminathan R.
Chem 1999;45: 3213. Self-monitoring in type 2 diabetes mellitus: a meta-analysis.
78. Stockl D, Baadenhuijsen H, Fraser CG, Libeer JC, Petersen Diabet Med 2000;17:75561.
PH, Ricos C. Desirable routine analytical goals for quantities 98. Harris MI. Frequency of blood glucose monitoring in relation to
assayed in serum. Discussion paper from the members of the glycemic control in patients with type 2 diabetes. Diabetes Care
External Quality Assessment (EQA) Working Group A on ana- 2001;24: 97982.
lytical goals in laboratory medicine. Eur J Clin Chem Clin Bio- 99. Davis WA, Bruce DG, Davis TM. Is self-monitoring of blood
chem 1995;33:15769. glucose appropriate for all type 2 diabetic patients? The Freman-
79. Fraser CG. The necessity of achieving good laboratory perfor- tle Diabetes Study. Diabetes Care 2006;29:176470.
mance. Diabet Med 1990;7: 4903. 100. Guerci B, Drouin P, Grange V, Bougneres P, Fontaine P, Kerlan
80. Olefsky JM, Reaven GM. Insulin and glucose responses to iden- V, et al. Self-monitoring of blood glucose significantly improves
tical oral glucose tolerance tests performed forty-eight hours metabolic control in patients with type 2 diabetes mellitus: the
apart. Diabetes 1974;23:44953. Auto-Surveillance Intervention Active (ASIA) study. Diabetes
81. Widjaja A, Morris RJ, Levy JC, Frayn KN, Manley SE, Turner Metab 2003;29:58794.
RC. Within- and between-subject variation in commonly mea- 101. Davidson MB, Castellanos M, Kain D, Duran P. The effect of
sured anthropometric and biochemical variables. Clin Chem self monitoring of blood glucose concentrations on glycated
1999; 45:5616. hemoglobin levels in diabetic patients not taking insulin: a
82. Sebastian-Gambaro MA, Liron-Hernandez FJ, Fuentes-Arderiu blinded, randomized trial. Am J Med 2005;118:4225.
X. Intra- and inter-individual biological variability data bank. 102. Franciosi M, Pellegrini F, De Berardis G, Belfiglio M, Di Nardo
Eur J Clin Chem Clin Biochem 1997;35:84552. B, Greenfield S, et al. Self-monitoring of blood glucose in non-
83. Mooy JM, Grootenhuis PA, de Vries H, Kostense PJ, Popp- insulin-treated diabetic patients: a longitudinal evaluation of its
Snijders C, Bouter LM, Heine RJ. Intra-individual variation of impact on metabolic control. Diabet Med 2005;22:9006.
glucose, specific insulin and proinsulin concentrations measured 103. Karter AJ, Parker MM, Moffet HH, Spence MM, Chan J, Ettner
by two oral glucose tolerance tests in a general Caucasian popu- SL, Selby JV. Longitudinal study of new and prevalent use of self-
lation: the Hoorn Study. Diabetologia 1996;39:298305. monitoring of blood glucose. Diabetes Care 2006;29:175763.
84. Selvin E, Crainiceanu CM, Brancati FL, Coresh J. Short-term 104. Martin S, Schneider B, Heinemann L, Lodwig V, Kurth HJ,
variability in measures of glycemia and implications for the Kolb H, Scherbaum WA. Self-monitoring of blood glucose in
classification of diabetes. Arch Intern Med 2007;167:154551. type 2 diabetes and long-term outcome: an epidemiological
85. Lacher DA, Hughes JP, Carroll MD. Estimate of biological varia- cohort study. Diabetologia 2006;49:2718.
tion of laboratory analytes based on the Third National Health and 105. Welschen LM, Bloemendal E, Nijpels G, Dekker JM, Heine RJ,
Nutrition Ex- amination Survey. Clin Chem 2005;51:4502. Stalman WA, Bouter LM. Self-monitoring of blood glucose in
86. Westgard QC. Desirable specifications for total error, impreci- patients with type 2 diabetes who are not using insulin. Cochrane
sion, and bias, derived from biologic variation. http://www.west- Database Syst Rev 2005:CD005060.
gard.com/biodatabase1. htm (Accessed July 2009). 106. Welschen LM, Bloemendal E, Nijpels G, Dekker JM, Heine RJ,
87. Howanitz PJ, Cembrowski GS, Steindel SJ, Long TA. Physi- Stalman WA, Bouter LM. Self-monitoring of blood glucose in
cian goals and laboratory test turnaround times. A College of patients with type 2 diabetes who are not using insulin: a system-
American Pathologists Q-Probes study of 2763 clinicians and atic review. Diabetes Care 2005;28: 15107.
722 institutions. Arch Pathol Lab Med 1993;117: 228. 107. Farmer A, Wade A, Goyder E, Yudkin P, French D, Craven A,
88. van den Berghe G, Wouters P, Weekers F, Verwaest C, et al. Impact of self monitoring of blood glucose in the manage-
Bruyninckx F, Schetz M, et al. Intensive insulin therapy in the ment of patients with non-insulin treated diabetes: open parallel
critically ill patients. N Engl J Med 2001;345:135967. group randomised trial. BMJ 2007;335:132.
48 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

108. Simon J, Gray A, Clarke P, Wade A, Neil A, Farmer A. Cost betes mellitus. Geneva: ISO; 2003. ISO 15197:2003; 1st ed.
effectiveness of self monitoring of blood glucose in patients with 2003-05-01.
non-insulin treated type 2 diabetes: economic evaluation of data 128. Clarke WL, Cox D, Gonder-Frederick LA, Carter W, Pohl SL.
from the DiGEM trial. BMJ 2008;336: 117780. Evaluating clinical accuracy of systems for self-monitoring of
109. OKane MJ, Bunting B, Copeland M, Coates VE. Efficacy of blood glucose. Diabetes Care 1987;10:6228.
self monitoring of blood glucose in patients with newly diag- 129. Skeie S, Thue G, Sandberg S. Patient-derived quality specifica-
nosed type 2 diabetes (ESMON study): randomised controlled tions for instruments used in self-monitoring of blood glucose.
trial. BMJ 2008;336:11747. Clin Chem 2001;47:6773.
110. Allemann S, Houriet C, Diem P, Stettler C. Self-monitoring of 130. Boyd JC, Bruns DE. Quality specifications for glucose meters:
blood glucose in non-insulin treated patients with type 2 diabe- assessment by simulation modeling of errors in insulin dose.
tes: a systematic review and meta-analysis. Curr Med Res Opin Clin Chem 2001; 47:20914.
2009;25:290313. 131. Novis DA, Jones BA. Interinstitutional comparison of bed-
111. Poolsup N, Suksomboon N, Rattanasookchit S. Meta-analysis side blood glucose monitoring program characteristics,
of the benefits of self-monitoring of blood glucose on glycemic accuracy performance, and quality control documentation: a
control in type 2 diabetes patients: an update. Diabetes Technol College of American Pathologists Q-Probes study of bedside
Ther 2009;11:77584. blood glucose monitoring performed in 226 small hospitals.
112. Montori VM, Fernndez-Balsells M. Glycemic control in type Arch Pathol Lab Med 1998;122: 495502.
2 diabetes: time for an evidence-based about-face? Ann Intern 132. Brunner GA, Ellmerer M, Sendlhofer G, Wutte A, Trajanoski
Med 2009;150: 8038. Z, Schaupp L, et al. Validation of home blood glucose meters
113. Gerich JE, Mokan M, Veneman T, Korytkowski M, Mitrakou A. with respect to clinical and analytical approaches. Diabetes Care
Hypoglycemia unawareness. Endocr Rev 1991;12:35671. 1998;21:58590.
114. Tang Z, Lee JH, Louie RF, Kost GJ. Effects of different 133. Weinzimer SA, Beck RW, Chase HP, Fox LA, Buckingham BA,
hematocrit levels on glucose measurements with hand- Tamborlane WV, et al. Accuracy of newer-generation home
held meters for point-of-care testing. Arch Pathol Lab Med blood glucose meters in a Diabetes Research in Children Net-
2000;124:113540. work (DirecNet) inpatient exercise study. Diabetes Technol Ther
115. ADA. Consensus statement on self-monitoring of blood glu- 2005;7:67580; discussion 6813.
cose. Diabetes Care 1994;17:816. 134. Mahoney J, Ellison J. Assessing the quality of glucose moni-
116. Tate PF, Clements CA, Walters JE. Accuracy of home blood glu- tor studies: a critical evaluation of published reports. Clin Chem
cose monitors. Diabetes Care 1992;15:5368. 2007;53:11228.
117. Chan JC, Wong RY, Cheung CK, Lam P, Chow CC, Yeung VT, et 135. Kristensen GB, Nerhus K, Thue G, Sandberg S. Standardized
al. Accuracy, precision and user-acceptability of self blood glucose evaluation of instruments for self-monitoring of blood glucose
monitoring machines. Diabetes Res Clin Pract 1997;36: 91104. by patients and a technologist. Clin Chem 2004;50:106871.
118. Kabadi UM, OConnell KM, Johnson J, Kabadi M. The effect 136. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight
of recurrent practice at home on the acceptability of capillary glucose control in critically ill adults: a meta-analysis. JAMA
blood glucose readings. Accuracy of self blood glucose testing. 2008;300:93344.
Diabetes Care 1994;17:111023. 137. Scott MG, Bruns DE, Boyd JC, Sacks DB. Tight glucose control
119. Burnett RW, DOrazio P, Fogh-Andersen N, Kuwa K, Kulp- in the intensive care unit: Are glucose meters up to the task? Clin
mann WR, Larsson L, et al. IFCC recommendation on reporting Chem 2009; 55:1820.
results for blood glucose. Clin Chim Acta 2001;307:2059. 138. Dungan K, Chapman J, Braithwaite SS, Buse J. Glucose mea-
120. DOrazio P, Burnett RW, Fogh-Andersen N, Jacobs E, Kuwa surement: confounding issues in setting targets for inpatient
K, Kulpmann WR, et al. Approved IFCC recommendation on management. Diabetes Care 2007;30:4039.
reporting results for blood glucose (abbreviated). Clin Chem 139. Boyd JC, Bruns DE. Monte Carlo simulation in establishing ana-
2005; 51:15736. lytical quality requirements for clinical laboratory tests meeting
121. Steffes MW, Sacks DB. Measurement of circulating glucose clinical needs. Methods Enzymol 2009;467:41133.
concentrations: The time is now for consistency among methods 140. Finkielman JD, Oyen LJ, Afessa B. Agreement between bed-
and types of samples. Clin Chem 2005;51:156970. side blood and plasma glucose measurement in the ICU setting.
122. Weitgasser R, Gappmayer B, Pichler M. Newer portable glucose Chest 2005; 127:174951.
metersanalytical improvement compared with previous gen- 141. Hoedemaekers CW, Klein Gunnewiek JM, Prinsen MA, Wil-
eration devices? Clin Chem 1999;45:18215. lems JL, Van der Hoeven JG. Accuracy of bedside glucose mea-
123. Bohme P, Floriot M, Sirveaux MA, Durain D, Ziegler O, Drouin surement from three glucometers in critically ill patients. Crit
P, Guerci B. Evolution of analytical performance in portable glu- Care Med 2008;36:30626.
cose meters in the last decade. Diabetes Care 2003;26: 11705. 142. Meynaar IA, van Spreuwel M, Tangkau PL, Dawson L, Sleeswijk
124. Skeie S, Thue G, Nerhus K, Sandberg S. Instruments for self-mon- Visser S, et al. Accuracy of AccuChek glucose measurement in
itoring of blood glucose: comparisons of testing quality achieved intensive care patients. Crit Care Med 2009;37:26916.
by patients and a technician. Clin Chem 2002;48: 9941003. 143. Kanji S, Buffie J, Hutton B, Bunting PS, Singh A, McDonald K,
125. ADA. Consensus statement on self-monitoring of blood et al. Reliability of point-of-care testing for glucose measure-
glucose. Diabetes Care 1987;10:939. ment in critically ill adults. Crit Care Med 2005;33:277885.
126. NCCLS. Ancillary (bedside) blood glucose testing in acute and 144. Schiffrin A, Belmonte M. Multiple daily self-glucose moni-
chronic care facilities; approved guideline C30-A. Villanova toring: its essential role in long-term glucose control in
(PA): NCCLS; 1994; 14:114. insulin-dependent diabetic patients treated with pump and
127. ISO. In vitro diagnostic test systems requirements for blood- multiple subcutaneous injections. Diabetes Care 1982;5:
glucose monitoring systems for self-testing in managing dia- 47984.
References 49

145 Nathan D. The importance of intensive supervision in deter- 164. Heise HM, Lampen P. Transcutaneous glucose measurements
mining the efficacy of insulin pump therapy. Diabetes Care using near-infrared spectroscopy: validation of statistical cali-
1983;6:2957. bration models. Diabetes Care 2000;23:120810.
146. de Veciana M, Major CA, Morgan MA, Asrat T, Toohey JS, 165. Gutman S, Bernhardt P, Pinkos A, Moxey-Mims M, Knott T,
Lien JM, Evans AT. Postprandial versus preprandial blood Cooper J. Regulatory aspects of noninvasive glucose measure-
glucose monitoring in women with gestational diabetes ments. Diabetes Technol Ther 2002;4:77981.
mellitus requiring insulin therapy. N Engl J Med 1995;333: 166. Rhiel MH, Amrhein MI, Marison IW, von Stockar U. The influ-
123741. ence of correlated calibration samples on the prediction perfor-
147. Goldstein DE, Little RR, Lorenz RA, Malone JI, Nathan D, mance of multivariate models based on mid-infrared spectra of
Peterson CM, Sacks DB. Tests of glycemia in diabetes. Diabetes animal cell cultures. Anal Chem 2002;74:522736.
Care 2004;27: 176173. 167. Arnold MA, Liu L, Olesberg JT. Selectivity assessment of non-
148. Garg S, Zisser H, Schwartz S, Bailey T, Kaplan R, Ellis S, Jova- invasive glucose measurements based on analysis of multivari-
novic L. Improvement in glycemic excursions with a transcu- ate calibration vectors. J Diabetes Sci Technol 2007;1:45462.
taneous, real-time continuous glucose sensor: a randomized 168. Small GW. Chemometrics and near-infrared spectroscopy:
controlled trial. Diabetes Care 2006;29:4450. avoiding the pitfalls. Trends Anal Chem 2006;25:105766.
149. Chetty VT, Almulla A, Odueyungbo A, Thabane L. The effect 169. Gabrielsson J, Trygg J. Recent developments in multivariate
of continuous subcutaneous glucose monitoring (CGMS) calibration. Crit Rev Anal Chem 2006;36:24355.
versus intermittent whole blood finger-stick glucose moni- 170. Arnold MA, Small GW, Xiang D, Qui J, Murhammer DW. Pure
toring (SBGM) on hemoglobin A1c (HBA1c) levels in type I component selectivity analysis of multivariate calibration mod-
diabetic patients: a systematic review. Diabetes Res Clin Pract els from near-infrared spectra. Anal Chem 2004;76:258390.
2008;81:7987. 171. Arnold MA, Olesberg JT, Small GW. Near-infrared spectroscopy
150. Tamborlane WV, Beck RW, Bode BW, Bucking-ham B, for noninvasive glucose sensing. In: Cunningham D, Stenken
Chase HP, Clemons R, et al. Continuous glucose monitor- JA, eds. Analytical chemistry of in vivo glucose measurements.
ing and intensive treatment of type 1 diabetes. N Engl J Med Hoboken, NJ: John Wiley & Sons; 2009. p 35790.
2008;359:146476. 172. Shih W-C, Bechtel KB, Feld MS. Noninvasive glucose sensing
151. Haupt K, Mosbach K. Plastic antibodies: developments and with Raman spectroscopy. In: Cunningham D, Stenken JA, eds.
applications. Trends Biotechnol 1998;16:46875. Analytical chemistry of in vivo glucose measurements. Hobo-
152. Chen G, Guan Z, Chen CT, Fu L, Sundaresan V, Arnold FH. A ken, NJ: John Wiley & Sons; 2009. p 391 419.
glucose-sensing polymer. Nat Biotechnol 1997;15:3547. 173. Lawrence JM, Contreras R, Chen W, Sacks DA. Trends in the
153. James TC, Sananayake DRAS, Shinkai S. A glucose-selec- prevalence of preexisting diabetes and gestational diabetes mel-
tive molecular fluorescence sensor. Angew Chem Int Ed Engl litus among a racially/ethnically diverse population of pregnant
1994;33:22079. women, 19992005. Diabetes Care 2008;31: 899904.
154. Birch DJS, Imhof RE. Time-domain fluorescence spectroscopy 174. Crowther CA, Hiller JE, Moss JR, McPhee AJ, Jeffries WS,
using time-correlated single-photon counting. Top Fluoresc Robinson JS. Effect of treatment of gestational diabetes mellitus
Spectrosc 1994; 1:195. on pregnancy outcomes. N Engl J Med 2005;352:247786.
155. Tolosa L, Szmacinski H, Rao G, Lakowicz JR. Lifetime-based 175. Landon MB, Spong CY, Thom E, Carpenter MW, Ramin SM,
sensing of glucose using energy transfer with a long lifetime Casey B, et al. A multicenter, randomized trial of treatment for
donor. Anal Biochem 1997;250:1028. mild gestational diabetes. N Engl J Med 2009;361:133948.
156. Rolinski OJ, Birch DJS, McCartney LJ, Pickup JC. Near-infra- 176. Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U,
red assay for glucose determination. Proc Soc Photo Opt Instrum Coustan DR, et al. Hyperglycemia and adverse pregnancy out-
Eng 1999;3602:614. comes. N Engl J Med 2008;358:19912002.
157. Marvin JS, Hellinga HW. Engineering biosensors by introducing 177. HAPO Study Cooperative Research Group. Hyperglycemia and
fluorescent allosteric signal transducers: construction of a novel Adverse Pregnancy Outcome (HAPO) Study: associations with
glucose sensor. J Am Chem Soc 1998;120:711. neonatal anthropometrics. Diabetes 2009;58:4539.
158. Wentholt IM, Vollebregt MA, Hart AA, Hoekstra JB, DeVries 178 International Association of Diabetes and Pregnancy Study
JH. Comparison of a needle-type and a microdialysis continu- Groups. International Association of Diabetes and Pregnancy
ous glucose monitor in type 1 diabetic patients. Diabetes Care Study Groups recommendations on the diagnosis and clas-
2005;28: 28716. sification of hyperglycemia in pregnancy. Diabetes Care
159. Khalil OS. Non-invasive glucose measurement technologies: an 2010;33:67682.
update from 1999 to the dawn of the new millennium. Diabetes 179. Kim C, Newton KM, Knopp RH. Gestational diabetes and the
Technol Ther 2004;6:66097. incidence of type 2 diabetes: a systematic review. Diabetes Care
160. Arnold MA, Small GW. Noninvasive glucose sensing. Anal 2002;25: 18628.
Chem 2005;77:542939. 180. ADA. Tests of glycemia in diabetes. Diabetes Care 1999;22:S779.
161. Tura A, Maran A, Pacini G. Non-invasive glucose monitoring: 181. Sacks DB. Diabetes mellitus. In: Burtis CA, Ash-wood ER,
assessment of technologies and devices according to quantita- Bruns DE, eds. Tietz textbook of clinical chemistry and molecu-
tive criteria. Diabetes Res Clin Pract 2007;77:1640. lar diagnostics. 5th ed. St. Louis: Elsevier Saunders. Accepted,
162. Sieg A, Guy RH, Delgado-Charro MB. Noninvasive and mini- forthcoming 2012.
mally invasive methods for transdermal glucose monitoring. 182. Kitabchi AE, Umpierrez GE, Murphy MB, Barrett EJ, Kreisberg
Diabetes Technol Ther 2005;7:17497. RA, Malone JI, Wall BM. Hyperglycemic crises in diabetes.
163. Kohl M, Essenpreis M, Cope M. The influence of glucose con- Diabetes Care 2004; 27(Suppl 1):S94102.
centration upon the transport of light in tissue-simulating phan- 183. Kreisberg RA. Diabetic ketoacidosis: new concepts and trends
toms. Phys Med Biol 1995;40:126787. in pathogenesis and treatment. Ann Intern Med 1978;88:68195.
50 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

184. Owen OE, Trapp VE, Skutches CL, Mozzoli MA, Hoeldtke RD, 204. Goldstein DE, Little RR, Wiedmeyer HM, England JD, McK-
Boden G, Reichard GA Jr. Acetone metabolism during diabetic enzie EM. Glycated hemoglobin: methodologies and clinical
ketoacidosis. Diabetes 1982;31:2428. applications. Clin Chem 1986;32:B6470.
185 Stephens JM, Sulway MJ, Watkins PJ. Relationship of blood 205. Svendsen PA, Lauritzen T, Soegaard U, Nerup J. Glycosyl-
acetoacetate and 3-hydroxybutyrate in diabetes. Diabetes ated haemoglobin and steady-state mean blood glucose con-
1971;20:4859. centration in type 1 (insulin-dependent) diabetes. Diabetologia
186. Porter WH, Yao HH, Karounos DG. Laboratory and clinical 1982;23:4035.
evaluation of assays for beta-hydroxybutyrate. Am J Clin Pathol 206. Cefalu WT, Wang ZQ, Bell-Farrow A, Kiger FD, Izlar C. Gly-
1997;107: 3538. cohemoglobin measured by automated affinity HPLC correlates
187. ADA. Tests of glycemia in diabetes [Position statement]. Diabe- with both short-term and long-term antecedent glycemia. Clin
tes Care 2000;23 Suppl 1:S802. Chem 1994;40:131721.
188. Csako G. Causes, consequences, and recognition of false-posi- 207. Murata GH, Hoffman RM, Duckworth WC, Wendel CS, Shah
tive reactions for ketones. Clin Chem 1990;36:13889. JH. Contributions of weekly mean blood glucose values to
189. Rosenbloom AL, Malone JI. Recognition of impending hemoglobin A1c in insulin-treated type 2 diabetes: the Dia-
ketoacidosis delayed by ketone reagent strip failure. JAMA betes Outcomes in Veterans Study (DOVES). Am J Med Sci
1978;240:24624. 2004;327:31923.
190. McMurray CH, Blanchflower WJ, Rice DA. Automated kinetic 208. Nathan DM, Turgeon H, Regan S. Relationship between gly-
method for D-3-hydroxybutyrate in plasma or serum. Clin Chem cated haemoglobin levels and mean glucose levels over time.
1984;30:4215. Diabetologia 2007;50: 223944.
191. Koch DD, Feldbruegge DH. Optimized kinetic method for 209. Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine
automated determination of beta-hydroxybutyrate. Clin Chem RJ. Translating the hemoglobin A1c assay into estimated average
1987;33:17616. glucose values. Diabetes Care 2008;31:14738.
192 Darrigo T. Beyond blood glucose. DiabetesForecast 210. Baker JR, Johnson RN, Scott DJ. Serum fructosamine concen-
1999;52:378. trations in patients with type II (non-insulin-dependent) diabetes
193. Westphal SA. The occurrence of diabetic ketoacidosis in non- mellitus during changes in management. Br Med J (Clin Res Ed)
insulin-dependent diabetes and newly diagnosed diabetic adults. 1984;288:14846.
Am J Med 1996;101:1924. 211. Tahara Y, Shima K. Kinetics of HbA1c, glycated albumin, and
194. Wiggam MI, OKane MJ, Harper R, Atkinson AB, Hadden DR, fructosamine and analysis of their weight functions against pre-
Trimble ER, Bell PM. Treatment of diabetic ketoacidosis using ceding plasma glucose level. Diabetes Care 1995;18:4407.
normalization of blood 3-hydroxybutyrate concentration as the 212. Little RR, Wiedmeyer HM, England JD, Wilke AL, Rohlfing
endpoint of emergency management. A randomized controlled CL, Wians FH Jr, et al. Interlaboratory standardization of mea-
study. Diabetes Care 1997; 20:134752. surements of glycohemoglobins. Clin Chem 1992;38:24728.
195. Umpierrez GE, Watts NB, Phillips LS. Clinical utility of beta- 213. Bodor GS, Little RR, Garrett N, Brown W, Goldstein DE, Nahm
hydroxybutyrate determined by reflectance meter in the manage- MH. Standardization of glyco-hemoglobin determinations in
ment of diabetic ketoacidosis. Diabetes Care 1995;18: 1378. the clinical laboratory: three years of experience. Clin Chem
196. Noyes KJ, Crofton P, Bath LE, Holmes A, Stark L, Oxley CD, 1992;38:24148.
Kelnar CJ. Hydroxybutyrate near-patient testing to evaluate 214. Weykamp CW, Penders TJ, Muskiet FA, van der Slik W. Effect
a new end-point for intravenous insulin therapy in the treat- of calibration on dispersion of glycohemoglobin values deter-
ment of diabetic ketoacidosis in children. Pediatr Diabetes mined by 111 laboratories using 21 methods. Clin Chem 1994;
2007;8:1506. 40:13844.
197. ADA. Implications of the Diabetes Control and Complications 215. Goldstein DE, Little RR. Bringing order to chaos: the National
Trial [Position statement]. Diabetes Care 2000;23 Suppl 1:S246. Glycohemoglobin Standardization Program. Contemp Int Med
198. Kitzmiller JL, Block JM, Brown FM, Catalano PM, Conway 1997;9:2732.
DL, Coustan DR, et al. Managing preexisting diabetes for preg- 216. Little RR, Rohlfing CL, Wiedmeyer HM, Myers GL, Sacks
nancy: summary of evidence and consensus recommendations DB, Goldstein DE. The National Glycohemoglobin Stan-
for care. Diabetes Care 2008;31:106079. dardization Program: a five-year progress report. Clin Chem
199. Davidson MB. Diabetes research and diabetes care. Where do 2001;47:198592.
we stand? Diabetes Care 1998; 21:215260. 217. DCCT. Feasibility of centralized measurements of glycated
200. ADA. Provider Notes: the newsletter of the ADA/ NCQA Pro- hemoglobin in the Diabetes Control and Complications Trial: a
vider Recognition Program. Provid Notes 2000;1:14. multicenter study. Clin Chem 1987;33:226771.
201. Nathan DM, Buse JB, Davidson MB, Heine RJ, Holman RR, 218. Little RR, Rohlfing CL, Sacks DB. Status of HbA1c measurement
Sherwin R, Zinman B. Management of hyperglycaemia in type and goals for improvement: from chaos to order for improving
2 diabetes: a consensus algorithm for the initiation and adjust- diabetes care. Clin Chem 2011;57:20514.
ment of therapy. A consensus statement from the American Dia- 219. Jeppsson JO, Kobold U, Barr J, Finke A, Hoelzel W, Hoshino
betes Association and the European Association for the Study of T, et al. Approved IFCC reference method for the mea-
Diabetes. Diabetologia 2006;49:171121. surement of HbA1c in human blood. Clin Chem Lab Med
202. Qaseem A, Vijan S, Snow V, Cross JT, Weiss KB, Owens DK. 2002;40:7889.
Glycemic control and type 2 diabetes mellitus: the optimal 220. Hoelzel W, Weykamp C, Jeppsson JO, Miedema K, Barr JR,
hemoglobin A1c targets. A guidance statement from the Ameri- Goodall I, et al. IFCC reference system for measurement of
can College of Physicians. Ann Intern Med 2007;147:41722. hemoglobin A1c in human blood and the national standardiza-
203. Bunn HF. Nonenzymatic glycosylation of protein: relevance to tion schemes in the United States, Japan, and Sweden: a method-
diabetes. Am J Med 1981; 70:32530. comparison study. Clin Chem 2004;50:16674.
References 51

221. Weykamp C, John WG, Mosca A, Hoshino T, Little R, Jeppsson as measured by ion-exchange chromatography, affinity chro-
J, et al. The IFCC reference measurement system for HbA1c: a matography, and colorimetry. Clin Chem 1983;29: 11135.
6-year progress report. Clin Chem 2008;54:2408. 240. Little RR, Rohlfing CL, Tennill AL, Connolly S, Hanson S.
222. Sacks DB. Translating hemoglobin A1c into average blood glucose: Effects of sample storage conditions on glycated hemoglo-
Implications for clinical chemistry. Clin Chem 2008;54:17568. bin measurement: evaluation of five different high perfor-
223. Hanas R, John G. 2010 consensus statement on the worldwide mance liquid chromatography methods. Diabetes Technol Ther
standardization of the hemoglobin A1c measurement. Clin Chem 2007;9:3642.
2010;56: 13624. 241. Baglin SK, Brown AS. Two capillary blood-collection tech-
224. Pani LN, Korenda L, Meigs JB, Driver C, Cha-many S, Fox CS, niques for estimating glycohemoglobin compared. Clin Chem
et al. Effect of aging on A1C levels in individuals without dia- 1995;41:3302.
betes: evidence from the Framingham Offspring Study and the 242. Voss EM, Cembrowski GS, Clasen BL, Spencer ML, Ainslie
National Health and Nutrition Examination Survey 20012004. MB, Haig B. Evaluation of capillary collection system for HbA1c
Diabetes Care 2008;31: 19916. specimens. Diabetes Care 1992;15:7001.
225. Ziemer DC, Kolm P, Weintraub WS, Vaccarino V, Rhee MK, 243. Little RR, Wiedmeyer HM, Huang DH, Goldstein DE, Parson
Twombly JG, et al. Glucose-independent, black-white differ- RG, Kowal R, et al. A simple blood collection device for anal-
ences in hemoglobin A1c levels: a cross-sectional analysis of 2 ysis of glycohemoglobin (GHB). Clin Chem 1998;44:A139.
studies. Ann Intern Med 2010;152:7707. 244. Baynes JW, Bunn HF, Goldstein D, Harris M, Martin DB, Peter-
226. Little RR, Sacks DB. HbA1c: How do we measure it and what does son C, Winterhalter K. National Diabetes Data Group: report
it mean? Curr Opin Endocrinol Diabetes Obes 2009;16:1138. of the expert committee on glucosylated hemoglobin. Diabetes
227. Herman WH, Ma Y, Uwaifo G, Haffner S, Kahn SE, Horton ES, Care 1984;7:6026.
et al. Differences in A1C by race and ethnicity among patients 245. Marshall SM, Barth JH. Standardization of HbA1c measure-
with impaired glucose tolerance in the Diabetes Prevention Pro- ments: a consensus statement. Ann Clin Biochem 2000;37:456.
gram. Diabetes Care 2007;30:24537. 246. Goodall I, Colman PG, Schneider HG, McLean M, Barker G.
228. Saaddine JB, Fagot-Campagna A, Rolka D, Narayan KM, Desirable performance standards for HbA1c analysis preci-
Geiss L, Eberhardt M, Flegal KM. Distribution of HbA1c lev- sion, accuracy and standardisation: consensus statement of the
els for children and young adults in the U.S.: Third National Australasian Association of Clinical Biochemists (AACB),
Health and Nutrition Examination Survey. Diabetes Care the Australian Diabetes Society (ADS), the Royal College
2002;25:132630. of Pathologists of Australasia (RCPA), Endocrine Society of
229. Selvin E, Steffes MW, Zhu H, Matsushita K, Wagenknecht L, Australia (ESA), and the Australian Diabetes Educators Asso-
Pankow J, et al. Glycated hemoglobin, diabetes, and cardiovascu- ciation (ADEA). Clin Chem Lab Med 2007;45: 108397.
lar risk in nondiabetic adults. N Engl J Med 2010;362: 80011. 247. Sacks DB. CAP surveys: participant summary for glycohemo-
230. Davie SJ, Gould BJ, Yudkin JS. Effects of vitamin C on glyco- globin survey 2010 set GH2-A. Northfield, IL: College of Amer-
sylation of proteins. Diabetes 1992; 41:16773. ican Pathologists; 2010.
231. Ceriello A, Giugliano D, Quatraro A, Donzella C, Dipalo G, 248. Goldstein DE, Peth SB, England JD, Hess RL, Da Costa J.
Lefebvre PJ. Vitamin E reduction of protein glycosylation in Effects of acute changes in blood glucose on HbA1c. Diabetes
diabetes. New prospect for prevention of diabetic complica- 1980;29:6238.
tions? Diabetes Care 1991;14:6872. 249. Cagliero E, Levina EV, Nathan DM. Immediate feedback of
232. Tarim O, Kucukerdogan A, Gunay U, Eralp O, Ercan I. Effects HbA1c levels improves glycemic control in type 1 and insulin-
of iron deficiency anemia on hemoglobin A1c in type 1 diabetes treated type 2 diabetic patients. Diabetes Care 1999;22: 17859.
mellitus. Pediatr Int 1999;41:35762. 250. Kennedy L, Herman WH, Strange P, Harris A. Impact of active
233. Bry L, Chen PC, Sacks DB. Effects of hemoglobin variants and versus usual algorithmic titration of basal insulin and point-of-
chemically modified derivatives on assays for glycohemoglobin care versus laboratory measurement of HbA1c on glycemic con-
[Review]. Clin Chem 2001;47:15363. trol in patients with type 2 diabetes: the Glycemic Optimization
234. Roberts WL, Chiasera JM, Ward-Cook KM. Glycohemoglobin with Algorithms and Labs at Point of Care (GOAL A1C) trial.
results in samples with hemoglobin C or S trait: a comparison of Diabetes Care 2006;29:18.
four test systems. Clin Chem 1999;45:9069. 251. Khunti K, Stone MA, Burden AC, Turner D, Raymond NT, Bur-
235. Weykamp CW, Penders TJ, Muskiet FA, van der Slik W. Influ- den M, Baker R. Randomised controlled trial of near-patient
ence of hemoglobin variants and derivatives on glycohemoglo- testing for glycated haemoglobin in people with type 2 diabetes
bin determinations, as investigated by 102 laboratories using 16 mellitus. Br J Gen Pract 2006;56:5117.
methods. Clin Chem 1993;39:171723. 252. Larsen ML, Horder M, Mogensen EF. Effect of long-term moni-
236. Schnedl WJ, Krause R, Halwachs-Baumann G, Trinker M, Lipp toring of glycosylated hemoglobin levels in insulin-dependent
RW, Krejs GJ. Evaluation of HbA1c determination methods in diabetes mellitus. N Engl J Med 1990;323:10215.
patients with hemoglobinopathies. Diabetes Care 2000;23: 33944. 253. Davidson MB, Peters AL, Schriger DL. An alternative approach
237. Rendell M, Brannan C, Nierenberg J, Rasbold K, Hestorff T. to the diagnosis of diabetes with a review of the literature. Dia-
Fingerstick glycosylated hemoglobin, plasma protein, and albu- betes Care 1995;18:106571.
min. Diabetes Care 1987;10:62932. 254. Lenters-Westra E, Slingerland RJ. Six of eight hemoglobin A1c
238. Ferrell RE, Hanis CL, Aguilar L, Tulloch B, Garcia C, Schull point-of-care instruments do not meet the general accepted ana-
WJ. Glycosylated hemoglobin determination from capillary lytical performance criteria. Clin Chem 2010;56:4452.
blood samples. Utility in an epidemiologic survey of diabetes. 255. American Association of Clinical Endocrinologists Board of
Am J Epidemiol 1984;119:15966. Directors/American College of Endocrinologists Board of
239. Little RR, England JD, Wiedmeyer HM, Goldstein DE. Effects Trustees. American Association of Clinical Endocrinologists/
of whole blood storage on results for glycosylated hemoglobin American College of Endocrinology statement on the use of
52 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

hemoglobin A1c for the diagnosis of diabetes. Endocr Pract 272. Saxena R, Voight BF, Lyssenko V, Burtt NP, de Bakker PI, Chen
2010;16:1556. H, et al. Genome-wide association analysis identifies loci for
256. Makita Z, Radoff S, Rayfield EJ, Yang Z, Skolnik E, Delaney type 2 diabetes and triglyceride levels. Science 2007;316:1331
V, et al. Advanced glycosylation end products in patients with 6.
diabetic nephropathy. N Engl J Med 1991;325:83642. 273. Scott LJ, Mohlke KL, Bonnycastle LL, Willer CJ, Li Y, Duren
257. Monnier VM, Bautista O, Kenny D, Sell DR, Fogarty J, Dahms WL, et al. A genome-wide association study of type 2 diabetes in
W, et al. Skin collagen glycation, glycoxidation, and cross- Finns detects multiple susceptibility variants. Science 2007;316:
linking are lower in subjects with long-term intensive versus 13415.
conventional therapy of type 1 diabetes: relevance of glycated 274. Meigs JB, Shrader P, Sullivan LM, McAteer JB, Fox CS, Dupuis
collagen products versus HbA1c as markers of diabetic compli- J, et al. Genotype score in addition to common risk factors for
cations. DCCT Skin Collagen Ancillary Study Group. Diabetes prediction of type 2 diabetes. N Engl J Med 2008;359: 220819.
Control and Complications Trial. Diabetes 1999;48: 87080. 275. Klein J, Sato A. The HLA system. First of two parts. N Engl J
258. Murphy R, Ellard S, Hattersley AT. Clinical implications of a Med 2000;343:7029.
molecular genetic classification of monogenic beta-cell diabetes. 276. Taylor SI, Arioglu E. Genetically defined forms of diabetes in
Nat Clin Pract Endocrinol Metab 2008;4:20013. children. J Clin Endocrinol Metab 1999;84:43906.
259. Edghill EL, Flanagan SE, Patch AM, Boustred C, Parrish A, 277. Fajans SS, Bell GI, Bowden DW, Halter JB, Polonsky KS.
Shields B, et al. Insulin mutation screening in 1,044 patients Maturity onset diabetes of the young (MODY). Diabet Med
with diabetes: mutations in the INS gene are a common cause 1996;13:S905.
of neonatal diabetes but a rare cause of diabetes diagnosed in 278. Todd JA. Genetics of type 1 diabetes. Pathol Biol (Paris)
childhood or adulthood. Diabetes 2008;57:103442. 1997;45:21927.
260. Stoy J, Greeley SA, Paz VP, Ye H, Pastore AN, Skowron KB, et 279. Atkinson MA, Eisenbarth GS. Type 1 diabetes: new perspectives
al. Diagnosis and treatment of neonatal diabetes: a United States on disease pathogenesis and treatment. Lancet 2001;358:2219.
experience. Pediatr Diabetes 2008;9:4509. 280. Harrison LC. Risk assessment, prediction and prevention of type
261. Fajans SS, Bell GI, Polonsky KS. Molecular mechanisms and 1 diabetes. Pediatr Diabetes 2001;2:7182.
clinical pathophysiology of maturity-onset diabetes of the 281. Redondo MJ, Kawasaki E, Mulgrew CL, Noble JA, Erlich HA,
young. N Engl J Med 2001;345:97180. Freed BM, et al. DR- and DQ-associated protection from type
262. Graham J, Hagopian WA, Kockum I, Li LS, Sanjeevi CB, Lowe 1A diabetes: comparison of DRB1*1401 and DQA1*0102-
RM, et al. Genetic effects on age-dependent onset and islet cell DQB1*0602*. J Clin Endocrinol Metab 2000;85: 37937.
autoantibody markers in type 1 diabetes. Diabetes 2002;51: 282. Wellcome Trust Case Control Consortium. Genome-wide asso-
134655. ciation study of 14,000 cases of seven common diseases and
263. Concannon P, Rich SS, Nepom GT. Genetics of type 1A diabe- 3,000 shared controls. Nature 2007;447:66178.
tes. N Engl J Med 2009;360: 164654. 283. Todd JA, Walker NM, Cooper JD, Smyth DJ, Downes K,
264. Nagi DK, Hendra TJ, Ryle AJ, Cooper TM, Temple RC, Clark Plagnol V, et al. Robust associations of four new chromosome
PM, et al. The relationships of concentrations of insulin, intact regions from genome-wide analyses of type 1 diabetes. Nat
proinsulin and 3233 split proinsulin with cardiovascular risk Genet 2007;39:85764.
factors in type 2 (non-insulin-dependent) diabetic subjects. Dia- 284. Maclaren NK, Kukreja A. Type 1 diabetes. In: Sly WS, ed. The
betologia 1990;33:5327. metabolic and molecular bases of inherited disease. 8th ed. St.
265. Barrett JC, Clayton DG, Concannon P, Akolkar B, Cooper JD, Louis: McGraw-Hill; 2001. p 147188.
Erlich HA, et al. Genome-wide association study and meta- 285. Alberti KG, Zimmet PZ. Definition, diagnosis and classification
analysis find that over 40 loci affect risk of type 1 diabetes. Nat of diabetes mellitus and its complications. Part 1: diagnosis and
Genet 2009;7037. classification of diabetes mellitus provisional report of a WHO
266. Nadler ST, Stoehr JP, Schueler KL, Tanimoto G, Yandell BS, consultation. Diabet Med 1998;15:53953.
Attie AD. The expression of adipogenic genes is decreased in 286. Palmer JP, Asplin CM, Clemons P, Lyen K, Tatpati O, Raghu
obesity and diabetes mellitus. Proc Natl Acad Sci U S A 2000;97: PK, Paquette TL. Insulin antibodies in insulin-dependent diabet-
113716. ics before insulin treatment. Science 1983;222:13379.
267. Ziegler AG, Bachmann W, Rabl W. Prophylactic insulin treat- 287. Baekkeskov S, Aanstoot HJ, Christgau S, Reetz A, Solimena
ment in relatives at high risk for type 1 diabetes. Diabetes Metab M, Cascalho M, et al. Identification of the 64K autoantigen in
Rev 1993;9: 28993. insulin-dependent diabetes as the GABA-synthesizing enzyme
268. Rewers M, Bugawan TL, Norris JM, Blair A, Beaty B, Hoffman glutamic acid decarboxylase. Nature 1990;347: 1516; erratum
M, et al. Newborn screening for HLA markers associated with 1990;347:728.
IDDM: Diabetes Autoimmunity Study in the Young (DAISY). 288. Kaufman DL, Erlander MG, Clare-Salzler M, Atkinson MA,
Diabetologia 1996;39:80712. Maclaren NK, Tobin AJ. Autoimmunity to two forms of glu-
269. Hagopian WA, Lernmark A, Rewers MJ, Simell OG, She JX, tamate decarboxylase in insulin-dependent diabetes mellitus. J
Ziegler AG, et al. TEDDYThe Environmental Determinants of Clin Invest 1992;89:28392.
Diabetes in the Young: an observational clinical trial. Ann N Y 289. Atkinson MA, Maclaren NK. Islet cell autoantigens in insulin-
Acad Sci 2006;1079:3206. dependent diabetes. J Clin Invest 1993;92:160816.
270. Kukreja A, Maclaren NK. Autoimmunity and diabetes. J Clin 290. Lan MS, Wasserfall C, Maclaren NK, Notkins AL. IA-2, a trans-
Endocrinol Metab 1999;84: 43718. membrane protein of the protein tyrosine phosphatase family, is
271. Barker JM, Goehrig SH, Barriga K, Hoffman M, Slover R, a major autoantigen in insulin-dependent diabetes mellitus. Proc
Eisenbarth GS, et al. Clinical characteristics of children diag- Natl Acad Sci U S A 1996;93:636770.
nosed with type 1 diabetes through intensive screening and fol- 291. Lu J, Li Q, Xie H, Chen ZJ, Borovitskaya AE, Maclaren NK,
low-up. Diabetes Care 2004;27:1399404. et al. Identification of a second transmembrane protein tyrosine
References 53

phosphatase, IA-2beta, as an autoantigen in insulin-dependent 306. Diabetes Prevention TrialType 1 Study Group. Effects of insu-
diabetes mellitus: precursor of the 37-kDa tryptic fragment. Proc lin in relatives of patients with type 1 diabetes mellitus. N Engl
Natl Acad Sci U S A 1996;93:230711. J Med 2002; 346:168591.
292. Wenzlau JM, Juhl K, Yu L, Moua O, Sarkar SA, Gottlieb P, et al. 307. Siljander H, Harkonen T, Hermann R, Simell S, Hekkala A,
The cation efflux transporter ZnT8 (Slc30A8) is a major autoan- Salonsaari RT, et al. Role of insulin autoantibody affinity as a
tigen in human type 1 diabetes. Proc Natl Acad Sci U S A 2007; predictive marker for type 1 diabetes in young children with
104:170405. HLA-conferred disease susceptibility. Diabetes Metab Res Rev
293. Wenzlau JM, Liu Y, Yu L, Moua O, Fowler KT, Rangasamy S, et 2009;25:61522.
al. A common nonsynonymous single nucleotide polymorphism 308. Williams AJ, Bingley PJ, Bonifacio E, Palmer JP, Gale EA. A
in the SLC30A8 gene determines ZnT8 autoantibody specificity novel micro-assay for insulin auto-antibodies. J Autoimmun
in type 1 diabetes. Diabetes 2008;57:26937. 1997;10:4738.
294. Patterson CC, Dahlquist GG, Gyurus E, Green A, Soltesz G. 309. Bingley PJ, Bonifacio E, Mueller PW. Diabetes Antibody Stan-
Incidence trends for childhood type 1 diabetes in Europe dur- dardization Program: first assay proficiency evaluation. Diabetes
ing 19892003 and predicted new cases 200520: a multicentre 2003;52:112836.
prospective registration study. Lancet 2009;373: 202733. 310. Grubin CE, Daniels T, Toivola B, Landin-Olsson M, Hagopian
295. Maclaren N, Lan M, Coutant R, Schatz D, Silverstein J, Muir A, WA, Li L, et al. A novel radioligand binding assay to determine
et al. Only multiple autoantibodies to islet cells (ICA), insulin, diagnostic accuracy of isoform-specific glutamic acid decarboxyl-
GAD65, IA-2 and IA-2beta predict immune-mediated (type 1) ase antibodies in childhood IDDM. Diabetologia 1994;37:34450.
diabetes in relatives. J Autoimmun 1999;12: 27987. 311. Torn C, Mueller PW, Schlosser M, Bonifacio E, Bingley PJ. Dia-
296. Verge CF, Gianani R, Kawasaki E, Yu L, Pietropaolo M, Jackson betes Antibody Standardization Program: evaluation of assays
RA, et al. Prediction of type I diabetes in first-degree relatives for autoantibodies to glutamic acid decarboxylase and islet anti-
using a combination of insulin, GAD, and ICA512bdc/ IA-2 gen-2. Diabetologia 2008;51:84652.
autoantibodies. Diabetes 1996;45:92633. 312. Mire-Sluis AR, Gaines Das R, Lernmark A. The World Health
297. Schott M, Schatz D, Atkinson M, Krischer J, Mehta H, Vold B, Organization International Collaborative Study for islet cell
Maclaren N. GAD65 autoantibodies increase the predictability antibodies. Diabetologia 2000;43:128292.
but not the sensitivity of islet cell and insulin autoantibodies for 313. Gleichmann H, Bottazzo GF. Progress toward standardization of
developing insulin dependent diabetes mellitus. J Autoimmun cytoplasmic islet cell-antibody assay. Diabetes 1987;36:57884.
1994;7:86572. 314. Verge CF, Stenger D, Bonifacio E, Colman PG, Pilcher C, Bing-
298. Turner R, Stratton I, Horton V, Manley S, Zimmet P, Mackay ley PJ, Eisenbarth GS. Combined use of autoantibodies (IA-2
IR, et al. UKPDS 25: autoantibodies to islet-cell cytoplasm and autoantibody, GAD autoantibody, insulin autoantibody, cyto-
glutamic acid decarboxylase for prediction of insulin require- plasmic islet cell antibodies) in type 1 diabetes: Combinatorial
ment in type 2 diabetes. UK Prospective Diabetes Study Group Islet Autoantibody Workshop. Diabetes 1998;47:185766.
[published erratum appears in Lancet 1998;351:376]. Lancet 315. Ellis TM, Schatz DA, Ottendorfer EW, Lan MS, Wasserfall C,
1997;350:128893. Salisbury PJ, et al. The relationship between humoral and cel-
299. Pozzilli P, Di Mario U. Autoimmune diabetes not requiring insu- lular immunity to IA-2 in IDDM. Diabetes 1998;47:5669.
lin at diagnosis (Latent Autoimmune Diabetes of the Adult): 316. Atkinson MA, Bowman MA, Campbell L, Darrow BL,
definition, characterization, and potential prevention. Diabetes Kaufman DL, Maclaren NK. Cellular immunity to a determi-
Care 2001;24:14607. nant common to glutamate decarboxylase and coxsackie virus
300. Palmer JP, Hampe CS, Chiu H, Goel A, Brooks-Worrell BM. Is in insulin-dependent diabetes [see comments]. J Clin Invest
latent autoimmune diabetes in adults distinct from type 1 dia- 1994;94:21259.
betes or just type 1 diabetes at an older age? Diabetes 2005; 317. Skyler JS, Krischer JP, Wolfsdorf J, Cowie C, Palmer JP, Green-
54(Suppl 2):S627. baum C, et al. Effects of oral insulin in relatives of patients with
301. Kobayashi T, Tanaka S, Harii N, Aida K, Shimura H, Ohm- type 1 diabetes: The Diabetes Prevention TrialType 1. Diabetes
ori M, et al. Immunopathological and genetic features in Care 2005;28:106876.
slowly progressive insulin-dependent diabetes mellitus and 318. Nanto-Salonen K, Kupila A, Simell S, Siljander H, Salonsaari
latent autoimmune diabetes in adults. Ann N Y Acad Sci T, Hekkala A, et al. Nasal insulin to prevent type 1 diabetes in
2006;1079:606. children with HLA genotypes and autoantibodies conferring
302. Braghi S, Bonifacio E, Secchi A, Di Carlo V, Pozza G, Bosi E. increased risk of disease: a double-blind, randomised controlled
Modulation of humoral islet autoimmunity by pancreas allo- trial. Lancet 2008;372: 174655.
transplantation influences allograft outcome in patients with 319. Agardh CD, Cilio CM, Lethagen A, Lynch K, Leslie RD, Palmer
type 1 diabetes. Diabetes 2000;49:21824. M, et al. Clinical evidence for the safety of GAD65 immuno-
303. Zimmet P, Turner R, McCarty D, Rowley M, Mackay I. Crucial modulation in adult-onset autoimmune diabetes. J Diabetes
points at diagnosis. Type 2 diabetes or slow type 1 diabetes. Dia- Complications 2005;19:23846.
betes Care 1999;22:B5964. 320. Lernmark A, Agardh CD. Immunomodulation with human
304. Petersen JS, Dyrberg T, Damm P, Kuhl C, Molsted-Pedersen L, recombinant autoantigens. Trends Immunol 2005;26:60812.
Buschard K. GAD65 auto-antibodies in women with gestational 321. ADA. Diabetes nephropathy. Diabetes Care 1999;22(Suppl
or insulin dependent diabetes mellitus diagnosed during preg- 1):S669.
nancy. Diabetologia 1996;39:132933. 322. Davidson MB, Bazargan M, Bakris G, Peters Harmel A,
305. Fuchtenbusch M, Ferber K, Standl E, Ziegler AG. Prediction of Campese V, Basta E. ImmunoDip: an improved screening
type 1 diabetes postpartum in patients with gestational diabetes method for microalbuminuria. Am J Nephrol 2004;24:2848.
mellitus by combined islet cell autoantibody screening: a pro- 323. Khosla N, Sarafidis PA, Bakris GL. Microalbuminuria. Clin Lab
spective multicenter study. Diabetes 1997; 46:145967. Med 2006;26:63553, vivii.
54 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

324. Sarafidis PA, Riehle J, Bogojevic Z, Basta E, Chugh A, Bakris 339. Weir MR, Bakris GL. Editorial perspective. Should microalbu-
GL. A comparative evaluation of various methods for microal- minuria ever be considered as a renal endpoint in any clinical
buminuria screening. Am J Nephrol 2008;28:3249. trial? Am J Nephrol 2010;31:46970.
325. Levey AS, de Jong PE, Coresh J, Nahas ME, Astor BC, Mat- 340. Glassock RJ. Debate: CON position. Should microalbuminuria
sushita K, et al. The definition, classification and prognosis of ever be considered as a renal endpoint in any clinical trial? Am J
chronic kidney disease: a KDIGO Controversies Conference Nephrol 2010;31:4625; discussion 4667.
report. Kidney Int [Epub ahead of print 2010 Dec 8]. 341. Steinke JM, Sinaiko AR, Kramer MS, Suissa S, Chavers BM,
326. KDOQI. KDOQI clinical practice guidelines and clinical prac- Mauer M. The early natural history of nephropathy in type 1
tice recommendations for diabetes and chronic kidney disease. diabetes: III. Predictors of 5-year urinary albumin excretion
Am J Kidney Dis 2007;49:S12154. rate patterns in initially normoalbuminuric patients. Diabetes
327. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, 2005;54:216471.
Izzo JL Jr, et al. Seventh report of the Joint National Commit- 342. Duka I, Bakris G. Influence of microalbuminuria in achiev-
tee on Prevention, Detection, Evaluation, and Treatment of High ing blood pressure goals. Curr Opin Nephrol Hypertens
Blood Pressure. Hypertension 2003;42:120652. 2008;17:45763.
328. Klausen KP, Scharling H, Jensen JS. Very low level of microal- 343. Schrier RW, Estacio RO, Mehler PS, Hiatt WR. Appropriate
buminuria is associated with increased risk of death in subjects blood pressure control in hypertensive and normotensive type 2
with cardiovascular or cerebrovascular diseases. J Intern Med diabetes mellitus: a summary of the ABCD trial. Nat Clin Pract
2006;260:2317. Nephrol 2007;3:42838.
329. Klausen KP, Scharling H, Jensen G, Jensen JS. New defini- 344. Ruggenenti P, Fassi A, Ilieva AP, Bruno S, Iliev IP, Brusegan V,
tion of microalbuminuria in hypertensive subjects: association et al. Preventing microalbuminuria in type 2 diabetes. N Engl J
with incident coronary heart disease and death. Hypertension Med 2004;351: 194151.
2005;46: 337. 345. Sinzinger H, Kritz H, Furberg CD. Atorvastatin reduces micro-
330. Ratto E, Leoncini G, Viazzi F, Vaccaro V, Parodi A, Falqui V, albuminuria in patients with familial hypercholesterolemia and
et al. Microalbuminuria and cardiovascular risk assessment in normal glucose tolerance. Med Sci Monit 2003;9:PI8892.
primary hypertension: should threshold levels be revised? Am J 346. Incerti J, Zelmanovitz T, Camargo JL, Gross JL, de Azevedo MJ.
Hypertens 2006;19:72834; discussion 7356. Evaluation of tests for microalbuminuria screening in patients
331. Wachtell K, Ibsen H, Olsen MH, Borch-Johnsen K, Lindholm with diabetes. Nephrol Dial Transplant 2005;20:24027.
LH, Mogensen CE, et al. Albumin-uria and cardiovascular risk 347. Lepore G, Maglio ML, Nosari I, Dodesini AR, Trevisan R. Cost-
in hypertensive patients with left ventricular hypertrophy: the effectiveness of two screening programs for microalbuminuria in
LIFE study. Ann Intern Med 2003;139:9016. type 2 diabetes. Diabetes Care 2002;25:21034; author reply 2104.
332. Rachmani R, Levi Z, Lidar M, Slavachevski I, Half-Onn E, 348. Roberts WL, Calcote CB, Cook CB, Gordon DL, Moore ML,
Ravid M. Considerations about the threshold value of micro- Moore S, et al. Comparison of four commercial urinary albumin
albuminuria in patients with diabetes mellitus: lessons from an (microalbumin) methods: implications for detecting diabetic
8-year follow-up study of 599 patients. Diabetes Res Clin Pract nephropathy using random urine specimens. Clin Chim Acta
2000;49:18794. 1998;273:2133.
333. Pambianco G, Costacou T, Orchard TJ. The prediction of major 349. Howey JE, Browning MC, Fraser CG. Biologic variation of uri-
outcomes of type 1 diabetes: a 12-year prospective evaluation of nary albumin: consequences for analysis, specimen collection,
three separate definitions of the metabolic syndrome and their interpretation of results, and screening programs. Am J Kidney
components and estimated glucose disposal rate: the Pittsburgh Dis 1989;13:357.
Epidemiology of Diabetes Complications Study experience. 350. Gansevoort RT, Verhave JC, Hillege HL, Burgerhof JG, Bakker
Diabetes Care 2007;30:124854. SJ, de Zeeuw D, de Jong PE. The validity of screening based on
334. Witte EC, Lambers Heerspink HJ, de Zeeuw D, Bakker SJ, de spot morning urine samples to detect subjects with microalbumin-
Jong PE, Gansevoort R. First morning voids are more reliable uria in the general population. Kidney Int Suppl 2005:S2835.
than spot urine samples to assess microalbuminuria. J Am Soc 351. Meinhardt U, Ammann RA, Fluck C, Diem P, Mullis PE. Micro-
Nephrol 2009;20:43643. albuminuria in diabetes mellitus: efficacy of a new screening
335. Levey AS, Coresh J, Greene T, Marsh J, Stevens LA, Kusek JW, method in comparison with timed overnight urine collection. J
Van Lente F. Expressing the Modification of Diet in Renal Disease Diabetes Complications 2003;17:2547.
Study equation for estimating glomerular filtration rate with stan- 352. Poulsen PL, Hansen B, Amby T, Terkelsen T, Mogensen CE.
dardized serum creatinine values. Clin Chem 2007;53:76672. Evaluation of a dipstick test for microalbuminuria in three dif-
336. Vassalotti JA, Stevens LA, Levey AS. Testing for chronic kidney ferent clinical settings, including the correlation with urinary
disease: a position statement from the National Kidney Founda- albumin excretion rate. Diabetes Metab 1992;18: 395400.
tion. Am J Kidney Dis 2007;50:16980. 353. Fernndez-Fernndez I, Pez Pinto JM, Hermosn Bono T,
337. Kistorp C, Raymond I, Pedersen F, Gustafsson F, Faber Vzquez Garijo P, Ortiz Camuez MA, Tarilonte Delgado MA.
J, Hildebrandt P. N-terminal pro-brain natriuretic peptide, Rapid screening test evaluation for microalbuminuria in diabe-
C-reactive protein, and urinary albumin levels as predictors tes mellitus. Acta Diabetol 1998;35: 199202.
of mortality and cardiovascular events in older adults. JAMA 354. Leong SO, Lui KF, Ng WY, Thai AC. The use of semi-quantitative
2005;293:160916. urine test-strip (Micral Test) for microalbuminuria screening in
338. Yuyun MF, Khaw KT, Luben R, Welch A, Bingham S, Day NE, patients with diabetes mellitus. Singapore Med J 1998;39: 1013.
Wareham NJ. Microalbuminuria independently predicts all-cause 355. Shaikh A, Seegmiller JC, Borland TM, Burns BE, Ladwig PM,
and cardiovascular mortality in a British population: The Euro- Singh RJ, et al. Comparison between immunoturbidimetry, size-
pean Prospective Investigation into Cancer in Norfolk (EPIC- exclusion chromatography, and LC-MS to quantify urinary albu-
Norfolk) population study. Int J Epidemiol 2004;33:18998. min. Clin Chem 2008;54:150410.
References 55

356. Lambers Heerspink HJ, Gansevoort RT, Brenner BM, Cooper middle-aged adults: the Framingham Offspring Study. Arch
ME, Parving HH, Shahinfar S, de Zeeuw D. Comparison of dif- Intern Med 2007;167:106874.
ferent measures of urinary protein excretion for prediction of 367. Rutter MK, Wilson PW, Sullivan LM, Fox CS, DAgostino
renal events. J Am Soc Nephrol 2010;21:135560. RB Sr, Meigs JB. Use of alternative thresholds defining insu-
357. Ibsen H, Olsen MH, Wachtell K, Borch-Johnsen K, Lindholm lin resistance to predict incident type 2 diabetes mellitus and
LH, Mogensen CE, et al. Reduction in albuminuria translates cardiovascular disease. Circulation 2008;117:10039.
to reduction in cardiovascular events in hypertensive patients: 368. Marks V. Recognition and differential diagnosis of spontane-
losartan intervention for endpoint reduction in hypertension ous hypoglycaemia. Clin Endocrinol (Oxf) 1992;37:30916.
study. Hypertension 2005;45:198 202. 369. Marks V, Teale JD. Hypoglycemia: factitious and felonious.
358. Collins AC, Sethi M, MacDonald FA, Brown D, Viberti GC. Endocrinol Metab Clin North Am 1999;28:579601.
Storage temperature and differing methods of sample prepa- 370. American College of Obstetrics and Gynecology. ACOG
ration in the measurement of urinary albumin. Diabetologia practice bulletin. Polycycstic ovary syndrome. Number 41,
1993;36: 9937. December 2002. Int J Gynaecol Obstet 2003;80:33548.
359. MacNeil ML, Mueller PW, Caudill SP, Steinberg KK. Consid- 371. Robbins DC, Andersen L, Bowsher R, Chance R, Dinesen B,
erations when measuring urinary albumin: precision, substances Frank B, et al. Report of the American Diabetes Associations
that may interfere, and conditions for sample storage. Clin Chem Task Force on Standardization of the Insulin Assay. Diabetes
1991;37:21203. 1996; 45:24256.
360. Hishiki S, Tochikubo O, Miyajima E, Ishii M. Circadian varia- 372. Marcovina S, Bowsher RR, Miller WG, Staten M, Myers G,
tion of urinary microalbumin excretion and ambulatory blood Caudill SP, et al. Standardization of insulin immunoassays:
pressure in patients with essential hypertension. J Hypertens report of the American Diabetes Association Workgroup. Clin
1998;16:21018. Chem 2007;53:7116.
361. Holl RW, Grabert M, Thon A, Heinze E. Urinary excretion of 373. Miller WG, Thienpont LM, Van Uytfanghe K, Clark PM,
albumin in adolescents with type 1 diabetes: persistent ver- Lindstedt P, Nilsson G, Steffes MW. Toward standardization
sus intermittent microalbuminuria and relationship to dura- of insulin immunoas-says. Clin Chem 2009;55:10118.
tion of diabetes, sex, and metabolic control. Diabetes Care 374. Staten MA, Stern MP, Miller WG, Steffes MW, Campbell SE.
1999;22:155560. Insulin assay standardization: leading to measures of insulin
362. Despres JP, Lamarche B, Mauriege P, Cantin B, Dagenais GR, sensitivity and secretion for practical clinical care. Diabetes
Moorjani S, Lupien PJ. Hyperinsulinemia as an independent Care 2010;33:2056.
risk factor for ischemic heart disease. N Engl J Med 1996;334: 375. Little RR, Rohlfing CL, Tennill AL, Madsen RW, Polonsky
9527. KS, Myers GL, et al. Standardization of C-peptide measure-
363. Reaven GM. Insulin resistance and its consequences; non-insu- ments. Clin Chem 2008;54: 10236.
lin dependent diabetes mellitus and coronary heart disease. 376. Centers for Medicare and Medicaid Services. Infusion
In: LeRoith D, Taylor SI, Olefsky JM, eds. Diabetes mellitus: pumps: C-peptide levels as a criterion for use. https://www.
a fundamental clinical text. Philadelphia: Lippincott-Raven; cms.gov/transmittals/ downloads/R513CP.pdf (Accessed
1996. p 50919. March 2008).
364. Chevenne D, Trivin F, Porquet D. Insulin assays and refer- 377. Burge MR, Schade DS. Insulins. Endocrinol Metab Clin
ence values. Diabetes Metab 1999;25: 45976. North Am 1997;26:57598.
365. Del Prato S. Measurement of insulin resistance in vivo. Drugs 378. ADA. Diagnosis and classification of diabetes mellitus. Dia-
1999;58:36; discussion 7582. betes Care 2010;33(Suppl 1): S629.
366. Wilson PW, Meigs JB, Sullivan L, Fox CS, Nathan DM,
DAgostino RB Sr. Prediction of incident diabetes mellitus in
56
Acknowledgment

We are grateful to the following individuals, who coauthored the original document on which these guidelines are based: David
Goldstein, Noel Maclaren, Jay McDonald, and Marian Parrott. We thank Robert Christenson, Christopher Price, Joseph Watine,
Jim Boyd, and Randie Little for insightful suggestions, and Rob Krikorian, Ibolya Bunda, and Betsy Garman for expert help in
compiling the document. We also thank all those who submitted comments on preliminary drafts of the document.

57
58
Appendix

The organizations and individuals listed below were invited to comment on the National Academy of Clinical Biochemistry
draft guidelines for laboratory testing of diabetes. We would like to acknowledge and thank those organizations and indi-
viduals who reviewed and commented on the draft guidelines. For those organizations that were able to send a representa-
tive to the Arnold O. Beckman Conference or provide written comments, the name of the representative is listed with the
organization.
Appendix Table 1. Organizations and individuals participating in the public commenting of the NACB Diabetes Mel-
litus Guidelines

Organizations:
ARUP Laboratories Association of Public Health Laboratories
William Roberts, MD, PhD www.aphl.org
http://www.aruplab.com/
Agency for Healthcare Research and Quality Bayer HealthCare
www.ahrq.gov Donald Parker, PhD
http://www.bayerhealthcare.com/scripts/pages/en/index.php
American Academy of Family Physicians Centers for Disease Control and Prevention
www.aafp.org www.cdc.gov
Jane Kelly, MD
American Association of Clinical Endocrinologists Centers for Medicare and Medicaid Services
www.aace.com http://www.cms.gov/
American Association of Diabetes Educators College of American Pathologists
www.aadenet.org www.cap.org
Amparo Gonzalez RN, CDE Peter Howanitz, MD
Karen Fitzner, PhD
American College of Obstetricians and Gynecologists Department of Veterans Affairs
www.acog.org www.va.gov
Donald Coustan, MD Leonard Pogach, MD
American College of Physicians Diabetes UK
www.acponline.org www.diabetes.org.uk
Merri Pendergrass, MD
American Diabetes Association The Endocrine Society
www.diabetes.org www.endo-society.org
M. Sue Kirkman, MD Lisa Marlow
Association for Clinical Biochemistry European Association for the Study of Diabetes
www.acb.org.uk www.easd.org
Garry John, MD Jonathan Levy, MD
Food and Drug Administration Siemens Healthcare Diagnostics
www.fda.gov Roma Levy, MS
Arleen Pinkos Tricia Bal, MD
Susan Selgren, PhD
http://www.medical.siemens.com/webapp/wcs/stores/servlet/
StoreCatalogDisplay~q_catalogId~e_-101~a_langId~e_-
101~a_storeId~e_10001.htm
International Diabetes Federation Lifescan Inc
www.idf.org John Mahoney, BA
http://www.lifescan.com/
International Federation of Clinical Chemistry and Laboratory National Institute of Diabetes and Digestive and Kidney
Medicine Diseases (of the National Institutes of Health)
www.ifcc.org www.nih.gov
Mauro Panteghini, MD

59
60 Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

International Society of Diabetes and Vascular Disease National Medical Association

http://www.intsocdvd.com/ http://www.nmanet.org
Italian SIBioC-SIMeL Study Group on Diabetes North American Nursing Diagnosis Association (NANDA-
http://www.simel.it/en/ International)
http://www.sibioc.it/ www.nanda.org
Mary Ann Lavin, ScD, RN, FAAN
Juvenile Diabetes Research Foundation Roche Diagnostics
www.jdrf.org Theresa Bush, PhD
http://www.roche.com/index.htm

Individuals:
Phillip Bach, Primary Childrens Medical Center, Salt Lake City, John Mahoney, Lifescan, USA
USA
Jim Boyd, University of Virginia, USA Matthew Meerkin, University of Notre Dame, Australia
Yu Chen, Dr. Everett Chalmers Regional Hospital/Horizon Andrea Mosca, University of Milan, Italy
Health Network, Canada
Rob Christenson, University of Maryland Medical Center, USA Christian Perier, Hospital Nord, Saint-Etienne, France
Edgard Delvin, CHU Ste-Justine, Montreal, Canada Leonard Pogach, VA New Jersey Healthcare System, USA
Kent Dooley, LifeLabs, British Columbia, Canada Chris Price, University of Oxford, UK
Raymond Gambino, Quest Diagnostics Inc, USA Kastoori Ramakrishnan, ProdConcepts, LLC
Mary Lou Gantzer, Siemens Healthcare Diagnostics, USA Maria del Patrocinio Chueca Rodriguez, Hospital Reina Sofia,
Spain
Eswari Gudipati, USA (patient view) Kareena Schnabl, DynaLIFEDx, Canada
Trefor Higgins, DynaLifeDx, Canada Dhastagir Sheriff, Al Arab Medical University, Benghazi, Libya
Stephen Kahn, Loyola University, USA Robbert Slingerland, Isala Klinieken, The Netherlands
Raymond Karcher (retired), Beaumont Hospital, USA John Tayek, Harbor UCLA Medical Center, USA
Eric Kilpatrick, Hull Royal Infirmary, UK Joseph Watine, Hpital de la Chartreuse, Villefranche-de-
Rouergue, France
Ben Kukoyi, Houston, USA Shirley Welch, Kaiser Permanente, USA
Phillip Lee, University of Texas Medical Branch Galveston, USA William E. Winter, University of Florida, USA
Randie Little, University of Missouri-Columbia School of
Medicine, USA
Appendix 61

Appendix Table 2: Criteria for prioritization of key questions

Prioritization criteria Explanatory notes Examples

A: The test has high A1: The test or its characteristics (e.g. its Glucose cut-off values for diagnosing DM,
impact on clinical out- diagnostic or target value or range) are di- IFG or IGT
comes (e.g. morbidity, rectly or indirectly linked to important clinical The impact of maternal glycemia on preg-
mortality, prognosis) outcomes nancy outcomes (direct link to outcome);
The test is a surrogate (indirect) measure of OGTT diagnostic criteria to detect GDM
important clinical outcomes (indirect link to outcome)
HbA1c is a surrogate measure of morbidity
and mortality
A2: The test and its result have a major impact Diagnostic criteria for DM to guide initiation
on clinical management decisions of treatment
HbA1c values in guiding decision on chang-
ing treatment
Albuminuria results guiding decisions on
initiating therapy with ACE-inhibitors
A3: There is current controversy on the use of OGTT vs FPG for the diagnosis of DM
the test in practice Diagnostic criteria for GDM
A4: There is wide variation in practice with Differing criteria for diagnosing DM or GDM
unfavorable outcomes (e.g. misdiagnosis of Variations in the use of random or timed
the condition) specimens and albumin concentration or
albumin excretion rate vs ACR for diagnos-
ing albuminuria
A5: New and substantial evidence has SMBG in type 2 DM
emerged since the publication of the 2002 HAPO study in GDM
NACB guideline
B: The test has high im- B1: High volume testing with uncertain impact SMBG in type 2 DM
pact on organizational
B2: There is public/commercial/ professional/ Use of portable meters in groceries,
outcomes
governmental pressure on testing by patients, etc.
Changing the expression of HbA1c values
due to standardization
C: The test has high C1: Testing is associated with high costs SMBG
impact on economic
C2: New and substantial evidence has
outcomes
emerged on the cost-effectiveness of the
test since the publication of the 2002 NACB
guideline
Abbreviations: ACE: Angiotensin Converting Enzyme; ACR: Albumin Creatainine Ratio; DM: Diabetes Mellitus; FPG: Fasting Plasma Glucose;
GDM: Gestational Diabetes Mellitus; HAPO: Hyperglycemia and Adverse Pregnancy Outcome; IFG: Impaired Fasting Glucose; IGT: Impaired
Glucose Tolerance; NACB: National Academy of Clinical Biochemistry; OGTT: Oral Glucose Tolerance Test; SMBG: Self-Monitoring of Blood
Glucose
Appendix Table 3: Evidence table 62
Chapter 1: GLUCOSE

No 1. NACB 2002 2. NACB 2011 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
recommenda- updated/ it necessary new recommendation design evidence(2) evidence(2)
tion and its new recom- to modify (high-mod- (high-moderate-
grade(1) mendation the recom- erate-low) low-very low)
with its grade mendation?
and quality of
evidence(2)
DOES GLUCOSE NEED TO BE MEASURED IN PLASMA FOR THE DIAGNOSIS OF DIABETES MELLITUS? (3)
Priority: 3 (B2, C1)
1.a Glucose should When glucose Clarification American Diabetes Association. Guideline Low High Direct relationship between glucose and
be measured is used to Standards of medical care in dia- expert complications of diabetes has been shown
in plasma in establish the di- betes2010. Diab Care 2010; 33 opinion in earlier high quality studies incorporated
an accredited agnosis of dia- (Suppl 1):S11-61 in ADA and WHO guidelines. Difficult to
laboratory to betes, it should World Health Organization, Definition Guideline Low evaluate quality of evidence as plasma glu-
establish the be measured in and Diagnosis of Diabetes Mellitus cose has been sole diagnostic criterion for
diagnosis of venous plasma and Intermediate Hyperglycermia: diabetes for many years of clinical practice.
diabetes A (high) Report of a WHO/IDF Consultation. Glucometers are not accurate enough to
Level A Geneva: World Health Organization, diagnose diabetes. This represents strong
2006 agreement of experts.

Engelgau MM, et al. Comparison cross-sec- High WHO recommends "venous plasma
of fasting and 2-hour glucose and tional popu- glucose" should be standard, but due to
HbA1c levels for diagnosing diabetes. lation-based wide-spread use of capillary sampling
Diagnostic criteria and performance sample (especially in under-resourced countries)
revisited. Diab Care 1997;20(5):785- capillary samples are accepted as a prag-
91. matic solution. However, evidence does
NOT support use of capillary samples.
McCance DR, e al. Comparison Cross sec- High Provides evidence on the relation between
of tests for glycated haemo-globin tional and complications and concomitant results of
and fasting and two hour plasma longitudinal the three tests.
glucose concentrations as diagnostic analysis Recommendation upgraded for direct link
methods for diabetes. BMJ. 1994; between glucose and DM complications
308(6940): 1323-8. Erratum in: BMJ and outcomes.
1994; 309(6958):841
(1)
 acks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus.
S
Clin Chem 2002;48:436-72.
(2)
Explanations for grading the level and quality of evidence and for grades of recommendations are given in Tables 1 and 2 in the Preamble.
(3)
For priority codes, see Appendix Table 2
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 2. NACB 2011 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
recommenda- updated/ it necessary new recommendation design evidence(2) evidence(2)
tion and its new recom- to modify (high-moder- (high-moderate-
grade(1) mendation the recom- ate-low) low-very low)
Appendix

with its grade mendation?

and quality of
evidence(2)
(3)
DOES GLUCOSE NEED TO BE MEASURED IN PLASMA FOR THE SCREENING OF DIABETES MELLITUS? Priority: 3 (B2, C1)
1.b Glucose should When glucose Former American Diabetes Association. Guideline Low Moderate WHO accepts glucometers for screening,
be measured is used for recommen- Standards of medical care in dia- expert for pragmatic reasons i.e., lack of access
in plasma in screening dation was betes --2010. Diab Care 2010; 33 opinion to an accredited central lab in underdevel-
an accredited of high-risk split for clari- (Suppl 1):S11-61 oped countries. This represents a strong
laboratory for individuals, fication and consensus view that it is better than doing
screening of it should be re-grading nothing.
high-risk indi- measured in World Health Organization, Definition Guideline Low
viduals venous plasma and Diagnosis of Diabetes Mellitus
Level E B (moderate) and Intermediate Hyperglycermia:
Report of a WHO/IDF Consultation.
Geneva: World Health Organization,
2006.
Jesudason DR, et al. Macro-vascular Popu- Moderate -
risk and diagnostic criteria for type lation- high
2 diabetes: implications for the use based
of FPG and HbA1c for cost-effective analysis
screening. Diab Care 2003;
26:485-90.
Knowler WC, et al. Reduction in the RCT High Recommendation downgraded for indirect-
incidence of type 2 diabetes with ness outcome was to reduce DM with
lifestyle intervention or metformin. N treatment/lifestyle changes.
Engl J Med 2002; 346:393-403.
Tuomilehto J, et al. Prevention of RCT High
type 2 diabetes mellitus by changes
in lifestyle among subjects with
impaired glucose tolerance. N Engl J
Med 2001; 344:1343-50.
1.c Plasma glu- Former Consensus of experts
cose should recommen-
be measured dation was
in an accred- split for clari-
ited laboratory fication and
when used for re-grading
diagnosis of or
screening for
diabetes
GPP
63
Chapter 1: GLUCOSE (Cont'd) 64

No 1. NACB 2. NACB 2011 3. Why was 4. Key references supporting the new recom- 5. Study design 6. Level of evi- 7. Quality of 8. Comments
2002 rec- updated/new it neces- mendation dence(2) (high- evidence(2)
ommenda- recommenda- sary to moderate-low) (high-moderate-
tion and its tion with its modify the low-very low)
grade(1) grade and recommen-
quality of evi- dation?
dence(2)
(3)
ARE SCREENING PROGRAMS FOR DIABETES MELLITUS EFFECTIVE? Priority: NOT LISTED
1.d Outcome stud- New recom- Kahn R, et al. Age at initiation and frequency of Cost-effective- High Moderate No evidence so far that
ies are needed mendation screening to detect type 2 diabetes: a cost-ef- ness study screening has benefit.
to determine the based on fectiveness analysis. Lancet 2010;375:1365-74 Quality of evidence down-
effectiveness of additional graded for indirectness.
screening evidence Glumer C, et al. What determines the cost-ef- Cost-effective- Moderate
C (moderate) fectiveness of diabetes screening? Diabetologia ness modeling
2006; 49:1536-44. study
Icks A, et al. Cost-effectiveness of type 2 diabe- Review and cost- Moderate - low
tes screening: results from recently published effectiveness
studies. Gesundheitswesen 2005; 67 Suppl analysis
1:S167-71
Hoerger TJ, et al. Screening for type 2 diabe- Cost-effective- Moderate
tes mellitus: a cost-effectiveness analysis. Ann ness analysis by
Intern Med 2004; 140:689-99. Markov model
Dallo FJ, Weller SC. Effectiveness of diabetes Cross-sectional High
mellitus screening recommendations. Proc Natl analysis of
Acad Sci USA 2003; 100:10574-9. population-based
study
Jesudason DR, et al. Macro-vascular risk and Population-based Moderate - high
diagnostic criteria for type 2 diabetes: implica- analysis
tions for the use of FPG and HbA1c for cost-
effective screening. Diab Care 2003; 26:485-90.
Perry RC, et al. HbA1c measurement improves RCT High
the detection of type 2 diabetes in high-risk
individuals with nondiagnostic levels of fasting
plasma glucose: the Early Diabetes Intervention
Program (EDIP). Diab Care 2001; 24:465-71
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 recom- 2. NACB 2011 updated/new 3. Why was 4. Key references supporting 5. Study 6. Level of 7. Quality of 8. Comments
mendation and its recommendation with its it necessary the new recommendation design evidence(2) evidence(2)
grade(1) grade and quality of evi- to modify (high-mod- (high-moder-
dence(2) the recom- erate-low) ate-low-very
Appendix

mendation? low)
(3)
DOES GLUCOSE NEED TO BE MEASURED IN PLASMA FOR THE MONITORING OF DIABETES MELLITUS? Priority: 3 (B2, C1)
1.e Routine measurement Routine measurement of plasma No change American Diabetes Association. Guideline ex- Low Low
of plasma glucose glucose concentrations in an Standards of medical care in pert opinion
concentrations in an accredited laboratory is not diabetes2010. Diab Care 2010;
accredited laboratory recommended as the primary 33 (Suppl 1):S11-61.
is not recommended means of monitoring or evaluat-
as the primary means ing therapy in individuals with
of monitoring or diabetes
evaluating therapy B (low)
in individuals with
diabetes.
Level E
(3)
WHAT ARE THE PRE-ANALYTICAL CONSIDERATIONS IN GLUCOSE TESTING? Priority: NOT LISTED
1.f Blood for fasting Blood for fasting plasma glucose Clarification WHO Definition and Diagnosis of Guideline Low Low Evidence reveals a diurnal
plasma glucose analy- analysis should be drawn in the Diabetes Mellitus and Intermedi- variation in FPG, with mean
sis should be drawn morning after the individual has ate Hyperglycemia: Report of a FPG higher in the morn-
after the subject has fasted overnight (at least 8 h) WHO/IDF Consultation. Geneva: ing than in the afternoon,
fasted overnight (at B (low) World Health Organization, 2006 indicating that many cases
least 8 h). Troisi RJ, et al. Diurnal variation Retro- High of diabetes would be missed
Level B in fasting plasma glucose: impli- spective in patients seen in the af-
cations for diagnosis of diabetes population- ternoon. No RCT compared
in patients examined in the after- based study morning vs afternoon testing
noon. JAMA 2000; 284:3157-9. in terms of diagnostic accu-
racy or outcomes. Therefore
American Diabetes Association. Guideline Low quality of evidence is down-
Report of the Expert Committee graded for indirectness.
on the Diagnosis and Classifica- However, there is strong
tion of Diabetes Mellitus. Diab consensus of experts that
Care 1997; 20:1183-97. a fasting plasma specimen
drawn in the morning should
be used.
65
Chapter 1: GLUCOSE (Cont'd) 66

No 1. NACB 2002 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
recommenda- new recommendation with it necessary new recommendation design evidence(2) evidence(2)
tion and its its grade and quality of to modify (high-mod- (high-moder-
grade(1) evidence(2) the recom- erate-low) ate-low-very
mendation? low)
1.g Plasma should To minimize glycolysis, one Clarification Gambino R et al. Acidification of Observa- High Moderate A consistent body of good evidence
be separated should place the sample tube blood is superior to sodium fluoride tional that delay in sample processing
from the cells immediately in an icewater alone as an inhibitor of glycolysis. leads to reduction in glucose in
within 60 min; slurry, and the plasma should Clin Chem 2009;55:1019-21. sample, and thus strong consen-
if this is not be separated from the cells sus that this may alter diagnostic
possible, a within 30 min. If that cannot accuracy. However, no study is
tube contain- be achieved, a tube contain- available to determine if this leads to
ing a glycolytic ing a rapidly effective glycoly- unfavorable outcomes or increased
inhibitor such as sis inhibitor, such as citrate rate of complications. Therefore
sodium fluoride buffer, should be used for quality of evidence is downgraded
should be used collecting the sample. Tubes for indirectness.
for collecting the with only enolase inhibitors,
sample such as sodium fluoride,
Level B should not be relied on to
prevent glycolysis
B (moderate)
Bruns DE, Knowler WC. Stabiliza- Editorial Low In vitro decrease of glucose may
tion of glucose in blood samples: lead to missed diagnoses of
Why it matters. Clin Chem [Editorial] diabetes in the large proportion of
2009;55:850-2. the population who have glucose
concentrations near the diagnostic
cut points for diabetes.
Sacks DB. Carbohydrates. In: Burtis Review Moderate-
CA, Ashwood ER, Bruns DE, eds. (book low
Tietz Textbook of Clinical Chemis- chapter)
try and Molecular Diagnostics, 4th
ed. St. Louis: Elsevier Saunders,
2006:837
Boyanton BL, Jr., Blick KE. Stability Observa- High
studies of twenty-four analytes in hu- tional
man plasma and serum. Clin Chem
2002; 48:2242-7
Stahl M, et al. Optimization of Observa- High
preanalytical conditions and analysis tional
of plasma glucose. 1. Impact of the
new WHO and ADA recommenda-
tions on diagnosis of diabetes mel-
litus. Scand J Clin Lab Invest 2001;
61:169-79
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2. NACB 2011 up- 3. Why was it neces- 4. Key references supporting the 5. Study 6. Level of 7. Qual- 8. Comments
2002 dated/new recom- sary to modify the new recommendation design evidence(2) ity of
recom- mendation with its recommendation? (high-mod- evidence(2)
menda- grade and quality of erate-low) (high-mod-
Appendix

tion and evidence(2) erate-low-

its grade(1) very low)
Chan AY, et al. Effectiveness of Observa- High
sodium fluoride as a preservative tional
of glucose in blood. Clin Chem
1989; 35:315-7.
Ladenson JH. Nonanalytical Review Moderate-
sources of variation in clinical (book low
chemistry results. In: Sonnenwirth chapter)
A, Jarett L, eds. Clinical Labora-
tory Methods and Diagnosis.
St. Louis, MO: C.V. Mosby Co.,
1980:149
(3)
DO ANALYTICAL GOALS FOR GLUCOSE ANALYSIS NEED TO CHANGE/IMPROVE WITH THE LOWERED CUTOFF FOR IFG? Priority: 2 (A1-3, B2)
1.h On the basis of biolog- New recommendation Ricos C et al. Current databases Review Moderate Low Quality of evidence is downgraded for indi-
ical variation, glucose for setting analytical on biological variation: pros, cons rectness to outcomes and for lack of primary
measurement should performance goals and progress. Scand J Clin Lab studies linking analytical performance to
have an analytical for achieving better Invest. 1999;59:491-500 outcomes. However, there is strong expert
imprecision 2.9%, diagnostic accuracy Fraser CG. The necessity of Expert Low consensus that analytical uncertainty of glu-
a bias 2.2%, and a around diagnostic achieving good laboratory perfor- opinion cose measurement could result in misclas-
total error 6.9%. To thresholds. mance. Diabet Med 1990; 7:490-3. sification of patients. The related recom-
avoid misclassification mendation therefore was upgraded to reflect
of patients, the goal this potential impact on patient centered
for glucose analysis outcomes.
should be to minimize
total analytical error,
and methods should
be without measurable
bias
B (low)
67
Chapter 2: GLUCOSE METERS 68

No 1. NACB 2002 rec- 2. NACB 2011 updated/ 3. Why was it neces- 4. Key references supporting 5. Study 6. Level 7. Quality of 8. Comments
ommendation and new recommendation sary to modify the the new recommendation design of evi- evidence(2)
its grade(1) with its grade and qual- recommendation? dence(2) (high-moder-
ity of evidence(2) (high- ate-low-very
moder- low)
ate-low)
(3)
SHALL PORTABLE METERS BE USED IN DIAGNOSIS AND SCREENING OF DIABETES MELLITUS? Priority: 2 (A3-4, B2, C1)
2.a There are no There are insufficient New evidence Dungan K, et al. Glucose mea- Review Low Moderate WHO recommends plasma, but
published data to published outcome data emerged since 2002 surement: Confounding issues accepts capillary whole blood
support a role for to support a role for and clarification. Prior in setting targets for inpatient using glucometer.
portable meters in portable meters and skin- recommendation was management. Diab Care 2007; WHO accepts meters for
the diagnosis of prick (finger-stick) blood split into two separate 30(2): 403-409. screening for practical and
diabetes or for popu- samples in diagnosis of recommendations for The Diabetes Research in Observa- High financial reasons. This rep-
lation screening. diabetes or for population clarity and regarding. Children Network (DirecNet) tional resents a strong consensus
The imprecision of screening Study Group. Accuracy of newer (Analyti- view that it is better than doing
the meters, coupled C (moderate) generation home blood glucose cal evalu- nothing.
with the substantial meters in a Diabetes Research ations)
differences among Glucometers are not accurate
in Children Network (DirecNet)
meters, precludes enough to diagnose diabetes.
inpatient exercise study. Diabe-
their use in the di- This represents strong agree-
tes Technology and Therapeutics
agnosis of diabetes ment of experts.
2005; 7(5): 675-680.
and limits their use- Quality of evidence down-
fulness in screening graded for inconsistency and
for diabetes indirectness of evidence.
Level E
(3)
HOW SHOULD PORTABLE METERS BE USED IN MONITORING TYPE 1 DIABETES MELLITUS? Priority: NOT LISTED
2.c SMBG is recom- Self-monitoring of blood Clarification American Diabetes Association. Guideline Low High Intensive glycemic control in
mended for all glucose (SMBG) is Standards of medical care in dia- expert patients with type 1 diabetes
insulin-treated recommended for all betes2010. Diab Care 2010;33 opinion was achieved in the DCCT by
patients with dia- insulin-treated patients (Suppl 1):S11-61 participants performing SMBG
betes. For type 1 with diabetes DCCT Research Group. The RCT High at least four times per day,
patients, SMBG is A (high) effect of intensive treatment of hence the ADA recommenda-
recommended three diabetes on the development tion and a strong consensus for
or more times a and progression of long-term SMBG to be performed three
day. SMBG may be complications in insulin-depen- or more times per day in type 1
desirable in patients dent diabetes mellitus. N Engl J diabetes.
treated with sulfo- Med 1993;329:977-986.
nylureas or other in-
sulin secretagogues
and in all patients
not achieving goals
Level B
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 rec- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the new 5. Study 6. Level 7. Quality of 8. Comments
ommendation and new recommendation it necessary recommendation design of evi- evidence(2) (high-
its grade(1) with its grade and to modify the dence(2) moderate-low-
quality of evidence(2) recommenda- (high- very low)
Appendix

tion? moder-
ate-low)
(3)
SHOULD PORTABLE METERS BE USED IN MONITORING TYPE 2 DM? Priority: 2 (A3, A5, B1-2, C1)
2.d In patients with type In patients with type 2 New evidence Allemann S, Houriet C, Diem P, Stettler C. Systematic High High In spite of the number of
2 diabetes, SMBG diabetes treated with emerged since Self-monitoring of blood glucose in non-in- Review high quality new studies
may help achieve diet and oral agents, the 2002 publi- sulin treated patients with type 2 diabetes: and evidence reviews,
better control, SMBG may help cation a systematic review and meta-analysis. there is insufficient evi-
particularly when achieve better con- Curr Med Res Opin 2009;25:2903-13 dence to claim improved
therapy is initiated or trol, particularly when Poolsup N, Suksomboon N, Rattana- Systematic High outcomes for SMBG in
changed. However, therapy is initiated or sookchit S. Meta-analysis of the benefits Review type 2 DM. Therefore
there are no data to changed. Data are of self-monitoring of blood glucose on clear recommendations
support this concept. insufficient, however, glycemic control in type 2 diabetes pa- for or against SMBG in
The role of SMBG in to claim an associated tients: an update. Diabetes Technol Ther. type 2 DM cannot be
patients with stable improvement of health 2009;11:775-84 made at this stage.
type 2 diabetes con- outcomes. The role of
trolled by diet alone SMBG in patients with Farmer A, et al. Impact of self monitoring RCT High
is not known stable type 2 diabetes of blood glucose in the management of
Level C controlled by diet alone patients with non-insulin treated diabetes:
is not known open parallel group randomised trial. BMJ
C (high) 2007;21;335:132
Martin S, at al. The ROSSO Study Group. Epidemio- Moder-
Self-monitoring of blood glucose in type lo-gical co- ate
2 diabetes and long-term outcome: an hort study
epidemio-logical study. Diabetologia
2006;49:2718.
Karter AJ, et al.Longitudinal study of new High
and prevalent use of self-monitoring of
blood glucose. Diab Care 2006;29:1757
63.
Welschen LMC, et al. Self-monitoring of Observa- High Systematic review of 6
blood glucose in patients with type 2 dia- tional study RCTs
betes mellitus who are not using insulin.
Cochrane Database of Systematic Re-
views 2005;Issue 2. Art. No.: CD005060.
Welschen LMC, et al. Self-moni-toring of Systematic
blood glucose in patients with type 2 dia- review
betes who are not using insulin: a system-
atic review. Diab Care 2005;28:15107.
69
Chapter 2: GLUCOSE METERS (Cont'd) 70

No 1. NACB 2002 recom- 2. NACB 2011 3. Why was it 4. Key references supporting the 5. Study design 6. Level of 7. Quality of 8. Comments
mendation and its updated/new necessary to new recommendation evidence(2) evidence(2)
grade(1) recommendation modify the rec- (high-mod- (high-moder-
with its grade ommendation? erate-low) ate-low-very
and quality of low)
evidence(2)
Davidson MB. Counter-point: Self- Expert opinion Low
Monitoring of Blood Glucose in Type
2 Diabetic Patients not Receiving
Insulin: A waste of money. Diab Care
2005;28:1531-3.
Franciosi M, et al., the QuED Study Observational study High
Group. Self-monitoring of blood glu-
cose in non-insulin-treated diabetic
patients: a longitudinal evaluation of
its impact on metabolic control. Diab
Med 2005;22:9006.
Guerci B, et al., the ASIA Group. Multi-center, pro- Moderate
Self-monitoring of blood glucose sig- spective open label,
nificantly improves metabolic control randomized trial
in patients with type 2 diabetes mel-
litus: the Auto-Surveillance Interven-
tion Active (ASIA) study. Diabetes
Metab 2003; 29:58794.
Harris MI. Frequency of blood Cross-sectional High NHANES study
glucose monitoring in relation to gly- study
cemic control in patients with type 2
diabetes. Diab Care 2001;24:979-82.
Coster S, et al. Self-monitoring in Meta-analysis High Meta-analysis of 8 RCTs
Type 2 diabetes mellitus: a meta-
analysis. Diab Med 2000;17:755-761.
Faas A, et al. The efficacy of Systematic review High 11 studies reviewed,
self-monitoring of blood glucose including 6 RCTs
in NIDDM subjects. Diab Care
1997;20:1482-1486.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 recom- 2. NACB 2011 updated/new 3. Why was it 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
mendation and its recommendation with its necessary to new recommendation design evidence(2) evidence(2)
grade(1) grade and quality of evi- modify the rec- (high-mod- (high-moder-
dence(2) ommendation? erate-low) ate-low-very
Appendix

low)
(3)
WHAT ARE THE PRE-ANALYTICAL CONSIDERATIONS FOR GLUCOSE METERS? Priority: 2 (A2-3, B1-2, C1)
2.e Patients should be Patients should be instructed Clarification and Kristensen GB, et al. Standard- Observa- High Moderate
instructed in the correct in the correct use of glucose new data ized evaluation of nine instru- tional
use of glucose meters, meters, including quality ments for self-monitoring of blood
including quality control. control. Comparison between glucose. Diab Technol and Therap
Comparison between SMBG and concurrent labora- 2008;10:467-77.
SMBG and concurrent tory glucose analysis should be Kristensen GB, et al. Standard- Observa- High
laboratory glucose analy- performed at regular intervals to ized evaluation of instruments for tional
sis should be performed evaluate the performance of the self-monitoring of blood glucose by
at regular intervals to meters in the patients hands patients and a technologist. Clin
evaluate the accuracy of B (moderate) Chem 2004; 50:1068-71.
patient results.
Level B Kabadi UM, et al. The effect of Observa- Moderate
recurrent practice at home on the tional
acceptability of capillary blood
glucose readings. Accuracy of self
blood glucose testing. Diab Care
1994;10:1110-23.
(3)
WHAT ARE THE ANALYTICAL CONSIDERATIONS FOR GLUCOSE METERS? Priority: 2 (A2-3, B1-2, C1)
2.f Multiple performance Multiple performance goals for Clarification and Kristensen GB, et al. Standard- Observa- High Low Performance goal tar-
goals for portable portable glucose meters have new data ized evaluation of nine instru- tional gets vary widely and
glucose meters have been proposed. These targets ments for self-monitoring of blood are highly controver-
been proposed. These vary widely and are highly glucose. Diab Technol and Therap sial. No evidence is
targets vary widely and controversial. Manufacturers 2008;10:467-77. available that the ADA
are highly controversial. should work to improve the im- The Diabetes Research in Children Obser- Moderate targets of less than
No published study precision of current meters, with Network (DirecNet) Study Group. vational 5% total error can be
has achieved the goals an intermediate goal of limiting Accuracy of newer generation home (Analyti- achieved in practice.
proposed by the ADA. total error for 95% of samples blood glucose meters in a Diabe- cal evalu- Downgraded evi-
Manufacturers should to 15% at glucose concentra- tes Research in Children Network ation) dence for inconsisten-
work to improve the tions 5.6 mmol/L (100 mg/dL) (DirecNet) Inpatient Exercise Study. cy, indirectness and
imprecision of current and to <0.8 mmol/L (15 mg/dL) Diab Technol Ther 2005;7:675-83. lack of consensus of
meters at glucose concentrations <5.6 experts.
Level E mmol/L (100 mg/dL). Lower Bohme P, et al. Evolution of Observa- High
total error would be desirable Analytical Performance in Portable tional
and may prove necessary in Glucose Meters in the Last Decade
tight glucose-control protocols Diab Care 2003;26:1170-5.
and for avoiding hypoglycemia Skeie S, et al. Instruments for self- Observa- High
in all settings monitoring of blood glucose: com- tional
C (low) parisons of testing quality achieved
by patients and a technician. Clin
Chem 2002;48:994-1003.
71
Chapter 2: GLUCOSE METERS (Cont'd) 72

No 1. NACB 2002 recom- 2. NACB 2011 up- 3. Why was it 4. Key references supporting the 5. Study design 6. Level of 7. Quality of 8. Comments
mendation and its dated/new recom- necessary to new recommendation evidence(2) evidence(2)
grade(1) mendation with its modify the rec- (high-mod- (high-moder-
grade and quality of ommendation? erate-low) ate-low-very
evidence(2) low)
Weitgasser R, et al. Newer portable Observational High
glucose meters - analytical improve-
ment compared with previous
generation devices? Clin Chem
1999;45:1821-1825.
American Diabetes Association. Guideline Low
Self-monitoring of blood glucose.
Diab Care 1996;19 (S 1):S62-66.
Novis DA, Jones BA. Interinstitu- Observational High Q-probe
tional comparison of bedside blood
glucose monitoring program char-
acteristics, accuracy perfor-mance,
and quality control documentation.
Arch Pathol Lab Med 1998;122:495-
502.
Barr JT, et al. Ancillary (bedside) Guideline Low
blood glucose testing in acute and
chronic care facilities. NCCLS
1994;14:1-14.
2.g We recommend meters Meters should No change, Expert consensus Low Very low
that measure and report measure and report rewording
plasma glucose con- plasma glucose
centrations to facilitate concentrations to
comparison with assays facilitate comparison
performed in accredited with assays per-
laboratories. formed in accredited
Level E laboratories
GPP
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 recommenda- 2. NACB 2011 up- 3. Why was 4. Key references supporting the new 5. Study 6. Level of 7. Quality of 8. Comments
tion and its grade(1) dated/new recom- it necessary recommendation design evidence(2) evidence(2)
mendation with its to modify (high-mod- (high-moder-
grade and quality the recom- erate-low) ate-low-very
Appendix

of evidence(2) mendation? low)

(3)
ARE GLUCOSE METERS ADEQUATE FOR WIDESPREAD USE IN INTENSIVE CARE UNITS? Priority: 2 (A1-3, B2, C1)
2.h Clinical studies are needed to Studies are needed Clarification Meynaar IA, et al. Accuracy of AccuChek Observational High Moderate-low
determine the analytic goals to determine the and expan- glucose measurement in intensive care patients. study
for glucose meters. At a mini- analytical goals sion of scope Crit Care Med 2009;37:2691-6.
mum, the end points should (quality specifica- of recom-
be glycated hemoglobin and tions) for glucose mendation
frequency of hypoglycemic meters in SMBG to intensive
episodes. Ideally, outcomes and in intensive care setting
(e.g., long-term complications care units
and hypoglycemia) should also C (moderate)
be examined
Level E
2.i Recommendations Boyd JC, Bruns DE. Monte Carlo simulation in Simulation Moderate
for future research: establishing analytical quality requirements for modeling
Important end points clinical laboratory tests meeting clinical needs.
in studies of SMBG Methods Enzymol 2009;467:411-33.
should include, at a Scott MG, et al. Tight glucose control in the Expert opinion Low
minimum, hemoglo- intensive care unit: Are glucose meters up to the
bin A1c (Hb A1c) and task? Clin Chem 2009; 55:18-20.
frequency of hypo-
glycemic episodes Scott MG, et al. Tight glucose control in critically Expert opinion Low
to ascertain whether ill adults [Letter]. JAMA 2008; 300(23):2726-7.
improved meters Wiener RS, et al. Benefits and risks of tight Systematic Moderate
enable patients glucose control in critically ill adults. JAMA review and
to achieve better 2008;300(8):933-944. meta-analysis
glucose control. For
studies of meter Hoedemaekers CW, et al. Accuracy of bed- Observational High
use in intensive side glucose measurement from three gluco- study
or critical care, meters in critically ill patients. Crit Care Med
important end points 2008;36(11):3062-6.
include mean blood Dungan K, et al. Glucose measurement: con- Narrative Low
glucose, frequency founding issues in setting targets for inpatient review
of hypoglycemia, management. Diabetes Care 2007;30:403-9.
and variation of
Finkielman J, et al: Agreement between bedside Observational Low
glucose control. Ide-
blood and plasma glucose measurement in the study
ally, outcomes (e.g.,
ICU setting. Chest 2005;127:1749-51.
long-term complica-
tions) should also van den Berghe G, et al. Intensive insulin RCT Moderate
be examined therapy in the critically ill patients. N Engl J Med.
GPP 2001;345(19):1359-1367.
73
Chapter 3: CONTINUOUS MINIMALLY-INVASIVE GLUCOSE ANALYSES 74

No 1. NACB 2002 recommen- 2. NACB 2011 up- 3. Why was it neces- 4. Key references sup- 5. Study 6. Level of 7. Quality of 8. Comments
dation and its grade(1) dated/new recom- sary to modify the porting the new recom- design evidence(2) evidence(2)
mendation with its recommendation? mendation (high-moder- (high-moder-
grade and quality of ate-low) ate-low-very
evidence(2) low)
(3)
ARE THERE ADEQUATE WELL CONTROLLED STUDIES DEMONSTRATING THE IMPACT OF CONTINUOUS GLUCOSE MONITORS Priority: 2 (A1, A3, B2, C1)
ON INTERMEDIATE OUTCOMES (E.G. HbA1c) TO JUSTIFY WIDESPREAD ADOPTION OF THE TECHNOLOGY?
(3)
GIVEN THE HIGH COSTS OF THE TECHNOLOGY, ARE THERE EVIDENCE-BASED SELECTION CRITERIA FOR ITS USE AND POTEN- Priority: 2 (A3, C1)
TIAL REIMBURSEMENT?
3.a Noninvasive glucose Real-time continuous Gluco Watch technol- The Juvenile Diabetes RCT High High Three age subgroups pre-spec-
analyses cannot be recom- glucose monitoring ogy is no longer on Research Foundation ified for outcome assessment
mended as replacements (CGM) in conjunction market and has been Continuous Glucose
for SMBG or glucose with intensive insulin supplanted by subcuta- Monitoring Study
measurements by an ac- regimens can be a neous CGM devices. Group: N.Engl.J.Med.
credited laboratory. useful tool to lower Additional evidence 2008;359:1464-1476
Ongoing developments Hb A1c in selected is available about ef-
in the field, such as use adults (age >25 fectiveness of real-time
of the new Gluco Watch years) with type 1 CGM.
Biographer, may influence diabetes
this recommendation. A (high)
Level E
3.b Although the New recommendation The Juvenile Diabetes RCT Moderate Moderate This was a per-protocol post-
evidence for lower- based on additional Research Foundation hoc analysis of the relationship
ing Hb A1c is not as evidence Continuous Glucose between HbA1c lowering and
strong for children, Monitoring Study days per week of use, not an
teens, and younger Group: N.Engl.J.Med. intention-to-treat analysis or
adults, real-time 2008;359:1464-1476 the primary outcome. Therefore
CGM may be helpful the quality of evidence and the
in these groups. Suc- strength of recommendation
cess correlates with were downgraded.
adherence to ongoing
use of the device
B (moderate)
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 recommendation 2. NACB 2011 updated/ 3. Why was it nec- 4. Key references sup- 5. Study 6. Level of 7. Quality of 8. Comments
and its grade(1) new recommendation essary to modify porting the new recom- design evidence(2) evidence(2)
with its grade and quality the recommenda- mendation (high-mod- (high-moder-
of evidence(2) tion? erate-low) ate-low-very
Appendix

low)
3.c Real-time CGM may be New recommenda- Garg S, et al. Improve- RCT Moderate Low Comparison of
a supplemental tool to tion based on ad- ment in glycemic excur- real-time vs. blinded
SMBG in individuals with ditional evidence sions with a transcutane- CGM (outcomes were
hypoglycemia unaware- ous, real-time continuous patients time in hy-
ness and/or frequent glucose sensor - a ran- per-glycemic and hy-
episodes of hypoglycemia domized controlled trial. poglycemic ranges).
B (low) Diab Care 2006;29:44-50 Evidence is indirect
as the outcome was
a surrogate biochemi-
cal marker (although
patient-related), i.e.
not clinical episodes
of hypoglycemia.
(3)
ARE CONTINUOUS GLUCOSE MONITORS SUFFICIENTLY ACCURATE FOR CLINICAL USE BY PATIENTS? Priority: 1 (A1-4, B1-2, C1)
3.d Patients require exten- New recommenda- Clinical ex- Low Very low FDA labeling of the
sive training in using the tion perience and device (for trend
device. Available devices FDA labeling assessment, not
must be calibrated with of the device treatment decisions -
SMBG readings, and the use SMBG for insulin
latter are recommended for dosing)
making treatment changes
GPP
75
Chapter 4: NON-INVASIVE GLUCOSE ANALYSIS 76

No 1. NACB 2002 rec- 2. NACB 2011 updated/ 3. Why was it 4. Key references supporting the new 5. Study 6. Level of 7. Quality of 8. Comments
ommendation and new recommendation necessary to recommendation design evidence(2) evidence(2)
its grade(1) with its grade and modify the rec- (high-mod- (high-moder-
quality of evidence(2) ommendation? erate-low) ate-low-very
low)
(3)
SHOULD PRESENT NON-INVASIVE GLUCOSE SENSING TECHNOLOGY BE RECOMMENDED FOR MONITORING GLYCEMIA? Priority: 3 (A3, A5, B2)
4.a Noninvasive No noninvasive sensing New recom- Arnold MA, et al. Selectivity assessment of Animal Low Very low Demonstration of selectiv-
glucose analyses technology is currently mendation and noninvasive glucose measurements based model ity issues. Downgraded
cannot be recom- approved for clinical clarification on analysis of multivariate calibration vec- for indirectness
mended as replace- glucose measurements tors. J Diabetes Sci Technol 2007;1:454-62.
ments for SMBG or of any kind. Major Tura A, et al. Non-invasive glucose moni- Review of Low Review with assessment
glucose measure- technological hurdles toring: assessment of technologies and technolo- of feasibility of each ap-
ments by an ac- must be overcome devices according to quantitative criteria. gies proach
credited laboratory. before noninvasive Diabetes Res Clin Pract 2007;77:16-40.
Ongoing develop- sensing technology will
ments in the field, be sufficiently reliable Arnold MA, Small GW. Noninvasive glu- Review of Low Review with listing of criti-
such as use of the to replace exist- cose sensing. Anal Chem 2005;77:4529- technolo- cal analytical parameters
new Gluco Watch ing portable meters, 39. gies
Biographer may implantable biosensors, Khalil OS. Non-invasive glucose measure- Review of Low Review with assessment
influence this rec- or minimally invasive ments at the dawn of the new millen- technolo- of feasibility of each ap-
ommendation. technologies nium: An update. Diabetes Technol Ther gies proach
Level E C (very low) 2004;6:660-697.
Gutman S, et al. Regulatory aspects of Consen- Low Listing of anticipant FDA
noninvasive glucose measurements. Dia- sus state- requirements for approval
betes Technol Ther 2002;4:779-81. ment of any future non-invasive
sensing technology.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 5: GESTATIONAL DIABETES MELLITUS (GDM)
No 1. NACB 2. NACB 2011 up- 3. Why was 4. Key references supporting the new 5. Study design 6. Level of 7. Quality of 8. Comments
2002 rec- dated/new recom- it necessary recommendation evidence(2) evidence(2)
Appendix

ommenda- mendation with its to modify the (high-mod- (high-moder-

tion and grade and quality recommenda- erate-low) ate-low-very
its grade(1) of evidence(2) tion? low)
(3)
WHAT ARE THE STRATEGIES FOR DETECTION AND DIAGNOSIS OF GESTATIONAL DIABETES MELLITUS? Priority: 1 (A5, B2)
5.a All pregnant New recom- American Diabetes Association. Standards Guideline, posi- High High Based on the HAPO study and
women not previ- mendation of medical care in diabetes2011. Diab Care tion statement the IADPS criteria, ADA recom-
ously known to have based on 2011;34 (Suppl 1):S11-61 mends that women with risk
diabetes should additional factors for type 2 diabetes are
undergo testing for evidence of screened for diabetes at the first
gestational diabetes associations prenatal visit.
mellitus (GDM) at of maternal International Association of Diabetes and Guideline, ex- High Expert Consensus Panel appoint-
2428 weeks of glycemia Pregnancy Study Groups. International as- pert consensus ed by IADPSG recommended
gestation and perinatal sociation of diabetes and pregnancy study outcome based criteria for the
A (high) outcome and groups recommendations on the diagnosis classification of glucose concen-
RCT results and classification of hyperglycemia in preg- trations in pregnancy.
showing ben- nancy. Diab Care 2010;33:676-82.
efit from treat-
ing mild GDM Hyperglycemia and Adverse Pregnancy Prospective High
and expert Outcome (HAPO) Study Cooperative Re- observational
consensus. search Group: Hyperglycemia and Adverse study of a multi-
Pregnancy Outcome (HAPO) Study: Associa- center cohort
tions with neonatal anthropometrics. Diabetes
2009;58:453-459.
Landon MB, et al. A multicenter, randomized RCT High This RCT does not deal with the
trial of treatment for mild gestational diabetes. diagnosis of GDM directly but pro-
N Engl J Med 2009;361:1339 vides evidence that treating mild
GDM improves outcome.
Hyperglycemia and Adverse Pregnancy Out- Prospective High Strong evidence for continuous
come (HAPO) Study Cooperative Research observational association between maternal
Group (Metzger BE, HAPO Study PI). Hyper- study of multi- glucose levels and pregnancy
glycemia and Adverse Pregnancy Outcomes. center cohort outcome
N Engl J Med 2008;358:1991-2002
Crowther CA, et al. Effect of treatment of ges- RCT High This RCT does not deal with the
tational diabetes mellitus on pregnancy out- diagnosis of GDM directly but pro-
comes. N Engl J Med 2005;352:2477 vides evidence that treating mild
GDM improves outcome.
77
Chapter 5: GESTATIONAL DIABETES MELLITUS (GDM) (Cont'd) 78

No 1. NACB 2. NACB 2011 up- 3. Why was 4. Key references supporting the new 5. Study design 6. Level of 7. Quality of 8. Comments
2002 rec- dated/new recom- it necessary recommendation evidence(2) evidence(2)
ommenda- mendation with its to modify the (high-mod- (high-moder-
tion and grade and quality recommenda- erate-low) ate-low-very
its grade(1) of evidence(2) tion? low)
5.b GDM should be New recom- International Association of Diabetes and Guideline, ex- High Moderate* This guideline was based on the
diagnosed by a 75-g mendation Pregnancy Study Groups. International as- pert consensus HAPO study and on the opinions
OGTT according to based on sociation of diabetes and pregnancy study of the IADPSG Consensus Panel
the IADPSG criteria additional groups recommendations on the diagnosis members because associations
derived from the evidence and classification of hyperglycemia in preg- between maternal glycemia and
HAPO study and expert nancy. Diab Care 2010;33:676-82. clinical outcomes were continu-
A (moderate) consensus. ous with no obvious thresholds at
which risks increased. Therefore
a consensus was required to
translate these results into clinical
practice.
Hyperglycemia and Adverse Pregnancy Out- Prospective High The study of 25,000 participants
come (HAPO) Study: associations with neo- multi-national revealed strong, graded, pre-
natal anthropometrics. Diabetes 2009;58:453 epidemiologic dominantly linear and continuous
study associations between maternal
glycemia and primary study
outcomes
Metzger, et al. Summary and Recommenda- Conference Moderate- Opinion of world-wide experts
tions of the Fifth International Workshop- review low based on findings of the HAPO
Conference on Gestational Diabetes Mellitus. outcome study.
Diab Care 2007;30:S251-S260.
*NB: The HAPO study and the subsequent guideline published suggest setting diagnostic thresholds at OR 1.75, but OR 1.5 and 2.0 were also considered.
The authors themselves suggest the followings:

It is likely that additional well-designed randomized controlled trials and other clinical studies will be needed to determine
1
cost-effective therapeutic strategies for treatment of GDM diagnosed by the IADPSG Consensus Panelrecommended criteria;
2
optimal glycemic treatment targets;
3
appropriate follow-up of mothers to determine risks for later development of diabetes, other metabolic disorders, or CVD risk factors; and
4
follow-up of children to assess potential associations of maternal glycemia with long-term risks of obesity, altered glucose metabolism, and CVD risk factors.

Therefore recommendations are likely to change as more evidence becomes available or modified locally for resource considerations. Therefore the quality of evidence is downgraded to
moderate but, due to strong consensus on the current criteria, the strength of recommendation is A.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 6: URINARY GLUCOSE
No 1. NACB 2002 recommen- 2. NACB 2011 updat- 3. Why was it nec- 4. Key references supporting 5. Study design 6. Level of 7. Quality of 8. Comments
dation and its grade(1) ed/new recommenda- essary to modify the new recommendation evidence(2) evidence(2)
Appendix

tion with its grade and the recommenda- (high-mod- (high-moder-

quality of evidence(2) tion? erate-low) ate-low-very
low)
(3)
IS THERE A ROLE FOR URINE GLUCOSE TESTING IN THE MANAGEMENT OF DIABETES MELLITUS? Priority: NOT LISTED
6.a Semi-quantitative urine Semiquantitative urine No change Goldstein DE, et al. Tests of Guideline Low Low Downgraded for low
glucose testing is not rec- glucose testing is not glycemia in diabetes. Diab Care quality and indirectness
ommended for routine care recommended for 2004;27:1761-73. of evidence. However,
of patients with diabetes routine care of patients American Diabetes Association. Guideline Low consensus is strong
mellitus with diabetes mellitus Tests of glycemia in diabetes. against the use of this
Level C B (low) Diab Care 1999;22:S77-9. test. IDF supports urine
glucose monitoring
where blood glucose is
not available or afford-
able.
79
Chapter 7: KETONE TESTING 80

No 1. NACB 2002 recommenda- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
tion and its grade(1) dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
mendation with its to modify the (high-mod- (high-moder-
grade and quality of recommenda- erate-low) ate-low-very
evidence(2) tion? low)
(3)
WHICH PATIENTS SHOULD BE ADVISED TO MEASURE URINE OR BLOOD KETONES AT HOME, AND UNDER WHAT CIRCUMSTANCES? Priority: 2 (A2-4)
7.a Ketones should be measured in Ketones measured No change ADA: Standards of Medical Care in Guideline Low Very low Expert opinion, clinical
urine or blood by patients with in urine or blood in Diabetes2009; Diab Care 2009; 32 expert experience
diabetes in the home setting the home setting by (Suppl 1):S13-S61 opinion
and in the clinic/hospital setting patients with diabetes ADA: Hyperglycemic crises in diabe- Guideline Low
as an adjunct to the diagnosis and in the clinic/hos- tes (position statement). Diab Care expert
of diabetic ketoacidosis pital setting should 2004; 27 (Suppl 1):S94-102 opinion
Level E be considered only
an adjunct to the
diagnosis of diabetic
ketoacidosis (DKA)
GPP
7.b Urine ketone determinations Urine ketone mea- No change ADA Tests of glycemia position Guideline Low Very low Based on lack of
should not be used to diagnose surements should not statement, DiaB Care 2001; 23 expert measurement of beta-
or monitor the course of DKA be used to diagnose (Suppl 1):S80-82). opinion hydroxybutyrate by
Level A or monitor the course nitroprusside
of DKA
GPP

(3)
ARE DIRECT MEASUREMENTS OF HBA PREFERABLE TO NITROPRUSSIDE MEASUREMENTS OF KETONES? Priority: 3 (A2)
7.c Blood ketone determinations Blood ketone No change Wiggam MI, et al. Treatment of RCT Moderate Moderate Outcome not clinically
that rely on the nitroprusside determinations that diabetic ketoacidosis using normal- meaningful
reaction should be used only rely on the nitroprus- ization of blood 3-hydroxybutyrate Downgraded for indi-
as an adjunct to diagnose side reaction should concentration as the end point of rectness of evidence
DKA and should not be used be used only as an emergency management. A random-
to monitor treatment of DKA. adjunct to diagnose ized controlled study. Diabetes Care
Specific measurement of HBA DKA and should not 1997;20:1347-52.
in blood can be used for diag- be used to monitor Umpierrez GE, et al. Clinical utility of Observa- Moderate Comparison of two
nosis and monitoring of DKA. DKA treatment. Spe- beta-hydroxybutyrate determined by tional cohort strategies of monitor-
Further studies are needed to cific measurement of reflectance meter in the manage- study ing DKA
determine if the test offers any -hydroxybutiric acid ment of diabetic ketoacidosis. Diab
clinical advantage over more in blood can be used Care 1995;18:137-8.
traditional management ap- for diagnosis and
proaches (e.g., measurements monitoring of DKA Noyes KJ, et al. Hydroxybutyrate Observa- Moderate Comparison of two
of serum CO2, anion gap, or B (moderate) near-patient testing to evaluate a tional cohort strategies of monitor-
pH). new endpoint for intravenous insulin study ing DKA
Level E therapy in the treatment of diabetic
ketoacidosis in children. Pediatr
Diabetes 2007;8:150-156
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 8: HEMOGLOBIN A1c
No 1. NACB 2002 rec- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
ommendation and dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
Appendix

its grade(1) mendation with its to modify the (high-mod- (high-moder-

grade and quality recommenda- erate-low) ate-low-very
of evidence(2) tion? low)
(3)
HOW GLYCATED HEMOGLOBIN SHOULD BE USED IN MONITORING DIABETES MELLITUS? Priority: NOT LISTED
8.a Glycated hemoglo- Hb A1c should be Clarification American Diabetes Association. Guideline Moderate Moderate The DCCT and UKPDS had de-
bin (GHb) should be measured routinely Standards of medical care in diabe- termined the relationship between
measured routinely in all patients with tes2010. Diab Care 2011;34 (Suppl the results of a specific GHb test
in all patients with diabetes mellitus 1):S11-61. (HbA1c) and long-term complications
diabetes mellitus to document their Nathan DM, et al. Management of Consensus Low in patients with type 1 and type 2
to document their degree of glycemic hyperglycaemia in type 2 diabetes: a statement diabetes, respectively
degree of glycemic control consensus algorithm for the initiation HbA1c has become a surrogate
control. A (moderate) and adjustment of therapy. A consen- outcome measure in DM but this
Level A sus statement from the American Dia- represents indirect evidence and
betes Association and the European therefore of moderate quality.
Association for the Study of Diabetes. However there is strong consensus
Diabetologia 2006;49:1711-21. for measuring HbA1c routinely in DM
U.K. Prospective Diabetes Study RCT High monitoring. Therefore the recom-
(UKPDS) Group. Intensive blood- mendation is upgraded.
glucose control with sulphonyl-ureas
or insulin compared with conventional
treatment and risk of complications
in patients with type 2 diabetes
(UKPDS 33). UK Prospective Diabe-
tes Study (UKPDS) Group. Lancet
1998;352:837-53
DCCT. The effect of intensive treat- RCT High
ment of diabetes on the development
and progression of long-term compli-
cations in insulin-dependent diabetes
mellitus. N Engl J Med 1993;329:
977-86.
81
Chapter 8: HEMOGLOBIN A1c (Cont'd) 82

No 1. NACB 2002 recommen- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
dation and its grade(1) new recommendation it necessary new recommendation design evidence(2) evidence(2)
with its grade and to modify the (high-mod- (high-moder-
quality of evidence(2) recommenda- erate-low) ate-low-very
tion? low)
(3)
WHAT ARE THE ANALYTICAL CONSIDERATIONS AND GOALS FOR HbA1c MEASUREMENT? Priority: 2 (A1)
8.b Laboratories should use Laboratories should Clarification Hanas R, John G. 2010 consen- Consensus Moderate Low Differences in HbA1c re-
only GHb assay methods use only Hb A1c as- and addition of sus statement on the worldwide statement ported led to an agreement
that are certified by the say methods that are new recom- standardization of the hemoglobin among IFCC and the major
National Glycohemoglobin certified by the National mendation A1c measurement. Clin Chem diabetes organizations to
Standardization Program Glycohemoglobin Stan- based on ex- 2010;56:1362-4 report HbA1c results as
as traceable to the DCCT dardization Program pert consensus Weykamp C, et al. The IFCC ref- Progress Moderate the IFCC result and as the
reference. In addition, (NGSP) as traceable erence measurement system for report equivalent NGSP DCCT-
laboratories that measure to the DCCT reference. HbA1c: a 6-year progress report. aligned result. Some, but
GHb should participate in a The manufacturers of Clin Chem 2008;54:240-8 not all, organizations have
proficiency-testing program, Hb A1c assays should agreed to report HbA1c as
such as the CAP Glyco- also show traceability Goldstein DE, et al. Tests of Positions Low the DCCT-aligned percent-
hemoglobin Survey, that to the IFCC reference glycemia in diabetes. Diab Care statement age and the IFCC value.
uses fresh blood samples method 2004;27:1761-73
Impact on patient out-
with targets set by the GPP Hoelzel W, et al. IFCC reference Method- High comes is unknown and
National Glycohemoglobin system for measurement of hemo- comparison indirect, therefore quality of
Standardization Program globin A1c in human blood and the study evidence is downgraded.
Laboratory Network national standardization schemes However, there is strong
Level B in the United States, Japan, and consensus of experts on
Sweden: a method-comparison HbA1c reporting.
study. Clin Chem 2004;50:166-74.
8.c Laboratories that Jeppsson JO, et al. Approved Method de- High
measure Hb A1c should IFCC reference method for the velopment
participate in a profi- measurement of HbA1c in hu-
ciency-testing program, man blood. Clin Chem Lab Med
such as the College of 2002;40:78-89.
American Pathologists Little RR, et al. The national gly- Analytical Moderate Retrospective analysis of
(CAP) Hb A1c survey, cohemoglobin standardization pro- study analytical performance of
that uses fresh blood gram: a five-year progress re-port. the NGSP network and
samples with targets Clin Chem 2001;47:1985-92. clinical labs in HbA1c mea-
set by the NGSP Labo- surement
ratory Network
Little RR, Goldstein DE. Stan- Analytical Low
GPP
dardization of glycohemoglobin study
measurements. AACC Endo
1995;13:109-24
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 recommen- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
dation and its grade(1) new recommendation it necessary new recommendation design evidence(2) evidence(2)
with its grade and qual- to modify the (high-mod- (high-moder-
ity of evidence(2) recommenda- erate-low) ate-low-very
Appendix

tion? low)
8.d Laboratories should be Laboratories should Clarifica- Ziemer DC, et al. Glucose-indepen- Cross-sec- Moderate Low Quality of evidence
aware of potential interfer- be aware of potential tion and new dent, black-white differences in he- tional study downgraded for indi-
ences, including hemoglo- interferences, including recommenda- moglobin A1c levels: a cross-sectional rectness
binopathies that may affect hemoglobinopathies,that tion based on analysis of 2 studies. Ann Intern Med
GHb test results. In select- may affect Hb A1c test experience 2010;152:770-7
ing assay methods, labora- results, depending on the and published Selvin E, et al. Glycated hemoglobin, Observa- High
tories should consider the method used. In selecting reports. diabetes, and cardiovascular risk in tional cohort
potential for interferences assay methods, laborato- nondiabetic adults. N Engl J Med study
in their particular patient ries should consider the 2010;362:800-11
population potential for interferences
Level A in their particular patient Bry L, et al. Effects of hemoglobin Review Low
population. In addition, variants and chemically modified
disorders that affect derivatives on assays for glyco-
erythrocyte turnover may hemoglobin [Review]. Clin Chem
cause spurious results, 2001;47:153-63.
regardless of the method Schnedl WJ, et al. Evaluation of HbA1c Test com- Moderate
used determination methods in patients parison
GPP with hemoglobinopathies. Diab Care study
2000;23:339-44.
Roberts WL, et al. Glycohemo-globin Test com- Moderate
results in samples with hemoglobin parison
C or S trait: a comparison of four test study
systems. Clin Chem 1999;45:906-9
Weykamp CW, et al. Influence of Multi/center Moderate
hemoglobin variants and derivatives method
on glycohemoglobin determinations, comparison
as investigated by 102 laborato- study
ries using 16 methods. Clin Chem
1993;39:1717-23.
83
Chapter 8: HEMOGLOBIN A1c (Cont'd) 84

No 1. NACB 2002 recommen- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level 7. Quality of 8. Comments
dation and its grade(1) dated/new recom- it necessary new recommendation design of evi- evidence(2)
mendation with its to modify the dence(2) (high-moder-
grade and quality of recommenda- (high- ate-low-very
evidence(2) tion? moderate- low)
low)
8.e Laboratories should use Desirable specifica- Clarification Little RR, et al. Status of HbA1c mea- Review Moderate Low This study used the refer-
GHb assay methods with tions for Hb A1c mea- and rewording surement and goals for improvement: ence change value (also
an interassay CV<5% surement are an in- of recommen- From chaos to order for improving called critical difference)
(ideally <3%). At least two tralaboratory CV <2% dations diabetes care. Clin Chem 2011;in to calculate an appropriate
control materials with dif- and an interlaboratory press analytical goal
ferent mean values should CV <3.5%. At least 2 Sacks DB. CAP Surveys: Partici- National Moderate The body of evidence is of
be analyzed as an inde- control materials with pant Summary for Glycohemoglobin survey low quality for indirectness
pendent measure of assay different mean values Survey 2010 Set GH2-A. Northfield, (<10% from of the data to clinical out-
performance. Laboratories should be analyzed IL: College of American Pathologists, outside US) comes, but there is strong
should verify specimens as an independent 2010. consensus of experts for
below the lower limit of measure of assay appropriate analytical speci-
the reference interval or performance fications to avoid unfavor-
greater than 15% by re- B (low) able outcomes of misclassifi-
peat testing. If Schiff base cations and mismanagement
(labile pre-HbA1c) interferes of patients.
with the assay method, it
should be removed prior Therefore the recommenda-
to assay tion was upgraded.
8.f Level C Samples with Goodall I, et al. Desirable perfor- Consensus Low
Hb A1c results below mance standards for HbA(1c) analysis statement
the lower limit of the - precision, accuracy and standardi-
reference interval or sation: consensus statement of the
>15% Hb A1c should Australasian Association of Clinical
be verified by repeat Biochemists (AACB), the Australian
testing Diabetes Society (ADS), the Royal
B (low) College of Pathologists of Australasia
(RCPA), Endocrine Society of Austra-
lia (ESA), and the Australian Diabetes
Educators Association (ADEA). Clin
Chem Lab Med 2007;45:1083-97.
8.g Hb A1c values that are Bry L, et al. Effects of hemoglobin Review Low
inconsistent with the variants and chemically modified
clinical presentation derivatives on assays for glyco-
should be investi- hemoglobin [Review]. Clin Chem
gated further 2001;47:153-63
GPP Marshall SM, Barth JH. Standardiza- Consensus Low
tion of HbA1c measurements: a con- statement
sensus statement. Ann Clin Biochem
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

2000;37:45-6
No 1. NACB 2002 recommen- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting 5. Study 6. Level of 7. Quality of 8. Comments
dation and its grade(1) new recommendation it necessary the new recommendation design evidence(2) evidence(2)
with its grade and qual- to modify the (high-mod- (high-moder-
ity of evidence(2) recommenda- erate-low) ate-low-very
Appendix

tion? low)
(3)
WHAT ARE THE HbA1c TREATMENT GOALS IN DIABETES MELLITUS? Priority: 2 (A1, A2)
8.h Treatment goals should Treatment goals should Clarification ADA. Standards of medical care Guideline Moderate High Converging validity of
be based on ADA recom- be based on American in diabetes2010. Diab Care several controlled clinical
mendations which include Diabetes Association 2010;33 (Suppl 1):S11-61. trials on patient-centered
maintaining GHb concen- recommendations, Duckworth W, et al. Glucose con- RCT High outcomes in type 1 and
trations <7% and reevalu- which include gener- trol and vascular complications in type 2 diabetes. Up-
ation of the treatment regi- ally maintaining Hb A1c veterans with type 2 diabetes. N graded for directness and
men for GHb values > 8%. concentrations at <7% Engl J Med 2009;360:129-39 consistency and strong
(Note that these values and more-stringent goals consensus of experts and
are applicable only if the in selected individual Gerstein HC, et al. Effects of RCT High several clinical organiza-
assay method is certified patients if they can be intensive glucose lowering in tions.
as traceable to the DCCT achieved without sig- type 2 diabetes. N Engl J Med
reference.) nificant hypoglycemia or 2008;358:2545-59
Level B other adverse treatment Patel A, et al. Intensive blood RCT High
effects. Somewhat higher glucose control and vascu-
intervals are recom- lar outcomes in patients with
mended for children and type 2 diabetes. N Engl J Med
adolescents and may be 2008;358:2560-72.
appropriate for patients
Berg AH, Sacks DB. Haemoglobin Review Low
with a limited life expec-
A1c analysis in the management
tancy, extensive comor-
of patients with diabetes: from
bid illnesses, a history of
chaos to harmony. J Clin Pathol
severe hypoglycemia, or
2008;61:983-7.
advanced complications
(note that these values Qaseem A, et al. Glycemic control Guideline, Moderate
are applicable only if and type 2 diabetes mellitus: the consensus
the NGSP has certified optimal hemoglobin A1c targets. statement
the assay method as A guidance statement from the
traceable to the DCCT American College of Physicians.
reference) Ann Intern Med 2007;147:417-22
A (high) ADA. Implications of the Diabetes Position Low
Control and Complications Trial statement
(position statement). Diab Care
2000;23 (Suppl 1):S24-6
DCCT. The relationship of glyce- RCT High
mic exposure (HbA1c) to the risk of
development and progression of
retinopathy in the diabetes control
and complications trial. Diabetes
1995;44:968-83
85
Chapter 8: HEMOGLOBIN A1c (Cont'd) 86

No 1. NACB 2002 2. NACB 2011 updated/ 3. Why was 4. Key references supporting 5. Study 6. Level of 7. Quality of 8. Comments
recommendation new recommendation it necessary the new recommendation design evidence(2) evidence(2)
and its grade(1) with its grade and qual- to modify the (high-mod- (high-moder-
ity of evidence(2) recommenda- erate-low) ate-low-very
tion? low)
(3)
WHAT SHOULD BE THE FREQUENCY OF HbA1c MONITORING IN DIABETES MELLITUS? Priority: NOT LISTED
8.i GHb testing Hb A1c testing should No change ADA. Standards of medical Guideline Moderate Low 240 patients; followed x1 year; 50%
should be per- be performed at least care in diabetes2010. Diab had HbA1c measured every 3 months;
formed at least biannually in all patients Care 2010;33 Suppl 1:S11-61. 50% no HbA1c measured. Does not
biannually in all and quarterly for patients Larsen ML, Horder M, Mo- RCT Moderate directly evaluate frequency only test-
patients and quar- whose therapy has gensen EF. Effect of long-term ing vs no testing. Moreover, the best
terly for patients changed or who are not monitoring of glycosylated correlations of HbA1c with complica-
whose therapy meeting treatment goals hemoglobin levels in insulin- tions have been based on quarterly
has changed or B (low) dependent diabetes mellitus. HbA1c testing for capturing overall gly-
are not meeting N Engl J Med 1990;323:1021-5 cemic exposure. However, there is no
treatment goals consensus on the optimal frequency of
Level B HbA1c testing. Most recommendations
are based on strong expert consensus.
(3)
SHOULD HbA1c BE USED FOR SCREENING AND DIAGNOSIS OF DIABETES MELLITUS? Priority: 1 (A1-5, B2, C1)
8.j Hb A1c may be used for New recom- ADA. Standards of medical Guideline Moderate Moderate The data supporting the use of
the diagnosis of diabetes, mendation care in diabetes2010. Diab HbA1c, i.e. its relationship with risk of
with values 6.5% being based on addi- Care 2010;33 (Suppl 1):S11- retinopathy, is similar to the data that
diagnostic. An NGSP-cer- tional evidence 61. support glucose testing as the means
tified method should be and consensus American Association of Clini- Guideline Moderate of diagnosis. These are definitional is-
performed in an accred- of experts cal Endocrinologists/American sues. Both the ADA and the American
ited laboratory. Analo- College of Endocrinology Endocrinology societies endorsed the
gous to its use in the statement on the use of he- HbA1c test for diagnosis.
management of diabetes, moglobin A1c for the diagnosis Other international organizations,
factors that interfere with of diabetes. Endocr Pract including the WHO and IDF, are con-
or adversely affect the Hb 2010;16:155-6 sidering HbA1c for diabetes diagnosis
A1c assay will preclude its and screening, therefore there is an
use in diagnosis Cheng YJ, et al. Association Population- High emerging strong consensus on the
A (moderate) of A1C and fasting plasma based cross topic, which resulted in upgrading the
glucose levels with diabetic sectional recommendation.
retinopathy prevalence in the
U.S. population: Implications
for diabetes diagnostic thresh-
olds. Diab Care 2009;32(11):
2027-32
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
recommenda- dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
tion and its mendation with its to modify the (high-moder- (high-moderate-
grade(1) grade and quality of recommenda- ate-low) low-very low)
Appendix

evidence(2) tion?
Nathan DM et al. for the International Expert con- Low A HbA1c value of 6.5% or greater
Expert Committee on the Diagnosis sensus was considered diagnostic based
of Diabetes. Report on the Role of the on the observed relationship with
Glycated Hemoglobin (A1C) Assay in retinopathy in more than 28,000
the Diagnosis of Diabetes. Diab Care persons. This represents direct
2009;32:1327-34 relationship to outcomes and thus
quality of evidence is upgraded.
Sabanayagam C, et al. Relation- Population- High
ship between glycated haemoglobin based cross
and microvascular complications: is sectional
there a natural cut-off point for the
diagnosis of diabetes? Diabetologia
2009;52(7):1279-89.
Ito C, et al. Importance of OGTT Population- High
for diagnosing diabetes mellitus based cross
based on prevalence and incidence sectional
of retinopathy. Diab Res Clin Pract.
2000;49(2-3): 181-6
8.k Point-of-care Hb A1c New recom- American Diabetes Association. Guideline Moderate Moderate The ADA cautions that POCT de-
assays are not suf- mendation Standards of medical care in diabetes vices for HbA1c should not be used
ficiently accurate to 2011. Diab Care 2011;34 (Suppl for diagnosis.
use for the diagnosis 1):S11-61
of diabetes Lenters-Westra E, Slingerland RJ. Six Analytical Moderate
B (moderate) of eight hemoglobin A1c point-of-care study
instruments do not meet the general
accepted analytical performance
criteria. Clin Chem 2010;56:44-52
87
Chapter 9: GENETIC MARKERS 88

No 1. NACB 2002 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
recommenda- dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
tion and its mendation with its to modify the (high-moder- (high-moderate-
grade(1) grade and quality of recommenda- ate-low) low-very low)
evidence(2) tion?
(3)
IS THERE A ROLE FOR GENETIC TESTING IN TYPE 1 DIABETES MELLITUS? Priority: NOT LISTED
9.a Routine mea- Routine measure- New informa- Concannon P, et al. Genetics of Review Moderate Moderate Useful review of genetic factors
surement of ment of genetic tion is available type 1A diabetes. N Engl J Med outside the HLA region.
genetic mark- markers is not of on mutations 2009;360:1646
ers is not of value at this time in the proinsu- Murphy R, et al. Clinical implications Linkage High Monogenic diabetes below the age
value at this for the diagnosis lin and other of a molecular genetic classification analyses of six needs to be considered for
time for the or management of genes that are of monogenic beta-cell diabetes. monogenic diabetes
diagnosis or patients with type linked to neo- Nat Clin Pract Endocrinol Metab
management 1 diabetes. For se- natal diabetes 2008;4:200-13.
of patients lected diabetic syn-
with type 1 dromes, including Edghill EL, et al. Insulin mutation Linkage High Many mutations than known hith-
diabetes. neonatal diabetes, screening in 1,044 patients with analyses erto affect the human preproinsulin
For selected valuable informa- diabetes: mutations in the INS gene in multiple gene
diabetic tion can be obtained are a common cause of neonatal familes
syndromes, with definition of diabetes but a rare cause of diabetes
valuable in- diabetes-associated diagnosed in childhood or adulthood.
formation can mutations Diabetes 2008;57:1034
be obtained A (moderate) Sty J, et al. Neonatal Diabetes Inter- Linkage Moderate Diabetes below the age of six
with definition national Collaborative Group. Insulin analyses months needs to be considered for
of diabetes- gene mutations as a cause of per- monogenic diabetes.
associated manent neonatal diabetes. Proc Natl
mutations Acad Sci USA. 2007;104(38):15040-4
Level E
Hagopian WA, et al. TEDDYThe Observa- High In contrast to other studies, the
Environ-mental Determinants of tional study TEDDY study has sufficient sta-
Diabetes in the Young: an observa- tistical power to answer questions
tional clinical trial. Ann N Y Acad Sci related to environmental triggers
2006;1079:320-6. for islet autoimmunity and type 1
diabetes.
Barker JM, et al. Clinical charac- Screening Moderate Early diagnosis may prevent hos-
teristics of children diagnosed with study of pitalization with ketoacidosis and
type 1 diabetes through intensive children at preserve residual beta cells. More
screening and follow-up. Diab Care risk for type outcome studies are needed to
2004;27:1399-404. 1 diabetes prove this.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study design 6. Level of 7. Quality of 8. Comments
recommendation dated/new recom- it neces- new recommendation evidence(2) evidence(2)
and its grade(1) mendation with its sary to (high-moder- (high-moder-
grade and quality of modify the ate-low) ate-low-very
Appendix

evidence(2) recommen- low)

dation?
Graham J, et al. Genetic effects on Population- Moderate First time INS VNTR were found to
age-dependent onset and islet cell based case- be associated with INS VNTR.
auto- antibody markers in type 1 dia- control study
betes. Diabetes 2002;51:1346-55
Fajans SS, et al. Molecular mecha- Review Low Careful analysis of family history of
nisms and clinical pathophysiology of diabetes is important to the detec-
maturity-onset diabetes of the young. tion of monogenic diabetes.
N Engl J Med 2001;345:971-80
Kukreja A, Maclaren NK. Auto-immu- Review Moderate
nity and diabetes. J Clin Endocrinol
Metab 1999;84:4371
Rewers M, et al. Newborn scree-ning Screening study Moderate It is possible to screen new-
for HLA markers associated with of children at born children to identify those at
IDDM: diabetes autoimmunity study risk for type 1 increased risk for developing type
in the young (DAISY). Diabetologia diabetes 1 diabetes. This strategy cannot
1996;39:807 be recommended until there is a
proven intervention available to
delay or prevent the disease.
Ziegler AG, et al. Prophylactic insulin Review Moderate
treatment in relatives at high risk for
type 1 diabetes. Diabetes Metab Rev
1993;9:289
(3)
IS THERE A ROLE FOR GENETIC TESTING IN TYPE 2 DIABETES MELLITUS? Priority: NOT LISTED
9.b There is no There is no role No change Meigs JB, et al. Genotype score in Genome wide Moderate Moderate Risk alleles in these loci all have
role for routine for routine genetic addition to common risk factors for association relatively small effects (odds ratios
genetic testing testing in patients prediction of type 2 diabetes. N Engl case-control 1.1 to 1.3) and do not significantly
in patients with with type 2 diabe- J Med 2008;359:2208-19. study enhance our ability to predict risk of
type 2 diabetes. tes. These studies type 2 diabetes
These stud- should be confined Scott LJ, et al. A genome- wide as- Genome wide Moderate
ies should be to the research set- sociation study of type 2 diabetes in association
confined to the ting and evaluation Finns detects multiple susceptibility case-control
research setting of specific syn- variants. Science 2007;316:1341 study
and evaluation of dromes
specific syn- A (moderate) Saxena R, et al. Genome wide asso- Genome wide Moderate
dromes ciation analysis identifies loci for type association
Level E 2 diabetes and triglyceride levels. case-control
Science 2007;316: 1331 study
89
Chapter 10: AUTOIMMUNE MARKERS 90

No 1. NACB 2002 recommenda- 2. NACB 2011 updated/ 3. Why was it nec- 4. Key references support- 5. Study 6. Level 7. Qual- 8. Comments
tion and its grade(1) new recommendation with essary to modify ing the new recommenda- design of evi- ity of evi-
its grade and quality of the recommenda- tion dence(2) dence(2)
evidence(2) tion? (high- (high-
moder- moderate-
ate-low) low-very
low)
(3)
SHOULD GAD65, IA-2 OR INSULIN AUTOANTIBODIES BE USED FOR THE DIAGNOSIS, SCREENING, MONITORING OF TYPE 1 Priority: 1.5 (A1-5, C1)
(3)
AND TYPE 2 DIABETES? Priority: 3 (A3-4, C1)
10.a Islet cell autoantibodies are Islet cell autoantibodies are Considerable Bingley PJ,et al. Measure- Analytical Moderate Low International workshops
recommended for screening recommended for screen- progress has been ment of islet cell antibod- test evalu- using serum exchange ex-
of non-diabetic family mem- ing nondiabetic family made to standard- ies in the Type 1 Diabetes ation ercises provide measures
bers who wish to donate part members who wish to do- ize islet cell auto- Genetics Consortium: efforts of inter-laboratory variation.
of their pancreas for trans- nate part of their pancreas antibody tests. to harmonize procedures Quality of evidence is
plantation to a relative with for transplantation into a among the laboratories. downgraded for indirect-
end stage, immune-mediated relative with end-stage type Clin Trials. 2010;7(1 ness.
(type 1) diabetes. Islet cell 1 diabetes Suppl):S56-64.
autoantibodies are not recom- B (low)
mended for routine diagnosis
of diabetes nor for screening
Level E
10.b Islet cell autoantibodies Trn C, et al. Participat- Analytical Moderate
are not recommended for ing Laboratories. Diabetes test evalu-
routine diagnosis of diabe- Antibody Standardization ation
tes, but standardized islet Program: evaluation of as-
cell autoantibody tests may says for autoantibodies to
be used for classification glutamic acid decarboxylase
of diabetes in adults and and islet antigen-2. Diabeto-
in prospective studies of logia. 2008;51(5):846-52.
children at genetic risk for
type 1 diabetes after HLA
typing at birth
B (low)
10.c Screening from GAD65 anti- Screening patients with Considerable Rolandsson O, Palmer Review Low Low Review suggesting that
bodies in patients diagnosed type 2 diabetes for islet progress has been JP. Latent autoimmune islet autoantibody positivity
with type 2 diabetes is not cell autoantibodies is not made to standard- diabetes in adults (LADA) is should suffice to classify
recommended at present to recommended at pres- ize islet autoan- dead: long live autoimmune adult diabetes patients with
be reclassified with type 1 ent. Standardized islet cell tibody tests. It is diabetes! Diabetologia. autoimmune diabetes
diabetes. autoantibodies are tested not clear to what 2010;53(7):1250-3. is GAD65 autoantibody
Level E in prospective clinical extent a positive positive.
studies of type 2 diabetes islet autoantibody Strength of recommenda-
patients to identify possible test would suffice tion is upgraded for strong
mechanisms of secondary to alter diagnostic consensus
failures of treatment of type criteria.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus

2 diabetes
B (low)
No 1. NACB 2002 recommen- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level 7. Qual- 8. Comments
dation and its grade(1) new recommendation it necessary new recommendation design of evi- ity of evi-
with its grade and quality to modify dence(2) dence(2)
of evidence(2) the recom- (high- (high-mod-
Appendix

mendation? moder- erate-low-

ate-low) very low)
10.d Screening of relatives of Screening for islet cell Clarification Patterson CC, et al. Incidence Multicentre Moder- Low Epidemiology data
patients with type 1 diabetes autoantibodies in rela- and addi- trends for childhood type 1 diabe- prospective ate
or of persons in the gen- tives of patients with type tion of new tes in Europe during 1989-2003 registration
eral population for islet cell 1 diabetes or in persons recommen- and predicted new cases 2005-20: study
autoantibodies is not recom- from the general popula- dation based a multicentre prospective regis-
mended at present tion is not recommended on new tration study. Lancet 2009;373:
Level E at present. Standardized evidence 2027-33
islet cell autoantibodies are Maclaren N, et al. Only multiple Review Low
tested in prospective clini- autoantibodies to islet cells (ICA),
cal studies insulin, GAD65, IA-2 and IA-2beta
B (low) predict immune-mediated (Type 1)
diabetes in relatives. J Autoimmun
1999;12:279-87
Verge CF, et al. Prediction of type Multicentre Moder- Data only applicable to first
I diabetes in first- degree relatives prospective ate degree relatives who com-
using a combination of insulin, registration prise only 10-15% of newly
GAD, and ICA512bdc/IA-2 autoan- study diagnosed type 1 diabetes
tibodies. Diabetes 1996;45:926-33. children.
Quality of the overall body of
evidence was downgraded for
lack of suitably powered stud-
ies or RCTs investigating the
value of islet cell autoantibody
testing for screening purposes
10.e There is currently no role for There is currently no role No change Sosenko JM et al. Glucose excur- Prospec- Moder- Low Data on first degree relatives
measurement of islet cell au- for measurement of islet sions between states of glycemia tive family ate suggest an important contribu-
toantibodies in the monitoring cell autoantibodies in the with progression to type 1 diabetes study of islet tion of insulin sensitivity on
of patients in clinical practice. monitoring of patients in in the diabetes prevention trial-type autoantibody glucose tolerwance.
Islet cell autoantibodies are clinical practice. Islet cell 1 (DPT-1). Diabetes Prevention positive Quality of the overall body of
measured in research proto- autoantibodies are mea- Trial-Type 1 Study Group. Diabe- subjects evidence was downgraded
cols and some clinical trials sured in research protocols tes. 2010;59(10):2386-9. for lack of sufficient data from
as surrogate endpoints and in some clinical trials multiple studies
Level E as surrogate end points
B (low)
91
Chapter 10: AUTOIMMUNE MARKERS 92

No 1. NACB 2002 recommen- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level 7. Qual- 8. Comments
dation and its grade(1) new recommendation it necessary new recommendation design of evi- ity of evi-
with its grade and quality to modify dence(2) dence(2)
of evidence(2) the recom- (high- (high-mod-
mendation? moder- erate-low-
ate-low) very low)
10.f It is important that autoanti- It is important that islet Clarifica- Bonifacio E, et al Harmonization Analytical Moder- Moderate Standardization was pos-
bodies be measured only in cell autoantibodies be tion, but no of glutamic acid decarboxylase test evalu- ate sible between three expert
an accredited laboratory with measured only in an ac- change and islet antigen-2 autoantibody ation laboratories.
an established quality control credited laboratory with an assays for national institute of
program and participation in established quality-control diabetes and digestive and kidney
a proficiency testing program program and participation diseases consortia. J Clin Endocri-
Level E in a proficiency-testing nol Metab. 2010;95(7):3360-7.
program
GPP
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 11: LOW LEVELS OF ALBUMINURIA (FORMERLY MICROALBUMINURIA)
No 1. NACB 2002 recom- 2. NACB 2011 updated/ 3. Why was it 4. Key references supporting the 5. Study 6. Level 7. Qual- 8. Comments
mendation and its new recommendation necessary to new recommendation design of evi- ity of evi-
Appendix

grade(1) with its grade and quality modify the rec- dence(2) dence(2)
of evidence(2) ommendation? (high- (high-
moder- moderate-
ate-low) low-very
low)
(3)
WHEN TESTING FOR LOW LEVELS OF ALBUMINURIA IS INDICATED? Priority: 1 (A5, A1-2)
11.a Annual microalbumin Annual testing for Clarification American Diabetes Association. Guideline Low Moderate There is a higher incidence
testing of patients albuminuria in patients Standards of medical care in diabetes expert of obesity and metabolic
without clinical protein- without clinical proteinuria 2010. Diab Care 2010; 33 (Suppl opinion derangements that accom-
uria should begin in should begin in pubertal 1):S11-61. pany this problem including
pubertal or postpubertal or postpubertal individuals Vassalotti JA, et al. Testing for chronic Position Low an increase in cardiovascu-
individuals five years 5 years after diagnosis of kidney disease: a position statement statement lar risk. Low levels of albu-
after diagnosis of type type 1 diabetes and at the from the National Kidney Foundation. minuria is a risk marker for
1 diabetes and at the time of diagnosis of type Am J Kidney Dis 2007;50 (2):169-180 cardiovascular events and
time of diagnosis of type 2 diabetes, regardless of predictive of cardiovascular
2 diabetes. The role treatment KDOQI Clinical Practice Guidelines Guideline Moderate events. This is especially
of testing is unclear in B (moderate) and Clinical Practice Recommenda- true in diabetes.
patients under treatment tions for Diabetes and Chronic Kidney
with angiotensin-con- Disease. Am J Kidney Dis 2007;49 (2
verting enzyme inhibitors Suppl 2):S12-154
and in those with short Klausen KP, et al. Very low level of Cohort study Low
life expectancy. microalbumin-uria is associated with
Level E increased risk of death in subjects
with cardio-vascular or cerebro-
vascular diseases. J Intern.Med.
2006;260 (3):231-237
Klausen KP, et al. New definition of Observation- Low
microalbuminuria in hyper-tensive al study
subjects: association with incident
coronary heart disease and death.
Hypertension 2005;46 (1):33-37
Kistorp K, et al. N-terminal pro- Meta-anal- Moderate
brain natriuretic peptide, C-reactive ysis
protein, and urinary albumin levels as
predictors of mortality and cardiovas-
cular events in older adults. JAMA
2005;293:1609-1616.
Gansevoort RT, et al. The validity Observation- Moderate Study in the Netherlands of
of screening based on spot morn- al study more than 30,000 people
ing urine samples to detect subjects
with microalbuminuria in the general
population. Kidney Int.Suppl 2005;
93

(94):S28-S35
No 1. NACB 2002 2. NACB 2011 up- 3. Why was it 4. Key references supporting the new 5. Study design 6. Level of 7. Quality of 8. Comments 94
recommenda- dated/new recom- necessary to recommendation evidence(2) evidence(2)
tion and its mendation with its modify the rec- (high-moder- (high-moder-
grade(1) grade and quality ommendation? ate-low) ate-low-very
of evidence(2) low)
Ibsen H, et al. Reduction in albuminuria Post hoc analysis Moderate Post hoc analysis of
translates to reduction in cardiovascular clinical cardiovascular
events in hypertensive patients: losartan outcome trials
intervention for end point reduction in hy-
pertension study. Hyper-tension 2005;45
(2):198-202
Arnlov J, et al. Low-grade albuminuria O b s e r v a t i o n a l Moderate Study of cardiovascular
and incidence of cardiovascular disease study outcomes
events in nonhypertensive and nondia-
betic individuals: the Framingham Heart
Study. Circulation 2005;112 (7):969-975
Chobanian AV, et al. Seventh report of the Guideline state- Moderate
Joint National Committee on Prevention, ment from NIH
Detection, Evaluation, and Treatment
of High Blood Pressure. Hypertension.
2003;42 (6):1206-1252
Lepore G, et al. Cost-effectiveness of two Cost-effective- Moderate
screening programs for microalbuminuria ness analysis
in type 2 diabetes. Diab Care 2002;25
(11):2103-2104
(3)
WHAT IS THE RELATIONSHIP BETWEEN ALBUMINURIA AND CARDIOVASCULAR OUTCOMES? Priority: 1 (A5, A1-2)
11.b Urine albumin at New recommen- G. Pambianco, et al. The prediction of O b s e r v a t i o n a l Moderate Moderate This was an observational
concentrations 30 dation major outcomes of type 1 diabetes: a cohort study study in patients with type
mg/g creatinine 12-year prospective evaluation of three 1 diabetes followed for 12
should be consid- separate definitions of the metabolic years.
ered a continuous syndrome and their components and
risk marker for estimated glucose disposal rate: the
cardiovascular Pittsburgh Epidemiology of Diabetes
events Complications Study experience. Diab
B (moderate) Care 2007;30(5):1248-1254.
Klausen KP, et al. Very low level of micro- Cohort study Moderate
albuminuria is associated with increased
risk of death in subjects with cardiovascu-
lar or cerebro-vascular diseases. J Intern
Med 2006;260 (3):231-237.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 11: LOW LEVELS OF ALBUMINURIA (FORMERLY MICROALBUMINURIA) (Cont'd)
No 1. NACB 2002 2. NACB 2011 up- 3. Why was it 4. Key references supporting the 5. Study design 6. Level of 7. Quality of 8. Comments
recommenda- dated/new recom- necessary to new recommendation evidence(2) evidence(2)
Appendix

tion and its mendation with its modify the rec- (high-moder- (high-moder-
grade(1) grade and quality ommendation? ate-low) ate-low-very
of evidence(2) low)
Ratto E, et al. Microalbuminuria and Observational Low The study evaluated level of
cardiovascular risk assessment in pri- cohort study microalbuminuria relative to
mary hypertension: should threshold development of left ventricular
levels be revised? Am J Hypertension hypertrophy; not cardiovascu-
2006;19 (7):728-734 lar outcome
Klausen KP, et al. New definition of Observational Low
microalbuminuria in hyperten-sive cohort study
subjects: association with incident
coronary heart disease and death.
Hypertension 2005;46 (1):33-37
K. Wachtell, et al. Albuminuria and Prospective ran- High This clinical trial evaluated
cardiovascular risk in hypertensive domized trial changes in albuminuria over
patients with left ventricular hypertro- a 5 year period in high risk
phy: the LIFE study. Ann.Intern.Med. patients for cardiovascular
2003;139 (11):901-906. events all of whom had left
ventricular hypertrophy.
R. Rachmani, et al. Considerations Observational Moderate This was an 8 year follow-up
about the threshold value of micro- cohort study of 599 people with diabetes
albuminuria in patients with diabetes evaluating changes in cardio-
mellitus: lessons from an 8-year vascular risk markers including
follow-up study of 599 patients. Diab. microalbuminuria
Res.Clin. Pract. 2000;49 (2-3):187-
194.
(3)
WHAT ARE THE ANALYTICAL CONSIDERATIONS WHEN TESTING FOR LOW LEVELS OF ALBUMINURIA? Priority: NOT LISTED
11.c The analytical The analytical CV No change Sarafidis PA, et al. A comparative Randomized Moderate Moderate Comparative studies of differ-
CV of methods of methods to mea- evaluation of various methods for study ent validated assays
to measure sure albuminuria microalbuminuria screening. Am.J
micro-albumin- should be <15% Nephrol. 2008;28 (2):324-329.
uria should be B (moderate) Gansevoort RT, et al. The validity Observational Moderate
<15% of screening based on spot morn- study
Level E ing urine samples to detect sub-
jects with microalbuminuria in the
general population. Kidney Int.Suppl
2005;(94):S28-S35
Incerti J, et al. Evaluation of tests Observational Moderate
for microalbuminuria screening in study
patients with diabetes. Nephrol Dial.
Transplant. 2005;20 (11):2402-2407
95
 96
No 1. NACB 2002 recom- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
mendation and its dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
grade(1) mendation with its to modify (high-moder- (high-moder-
grade and quality of the recom- ate-low) ate-low-very
evidence(2) mendation? low)
Meinhardt U, et al. Microalbumin-uria Observa- Moderate
in diabetes mellitus: efficacy of a new tional study
screening method in comparison with
timed overnight urine collection J Diab
Compli-cations 2003;17 (5):254-257
11.d Semiquantitative or Semiquantitative or No change Sarafidis PA,et al. A comparative evalu- Randomized Moderate Moderate Most recent studies do have
qualitative screening qualitative screen- ation of various methods for microal- study >95% for Hemocue and Im-
tests for microalbumin- ing tests should be buminuria screening. Am.J Nephrol. munodip but only one study
uria should be positive positive in >95% 2008;28 (2):324-329. confirmed against standard
in >95% of patients with of patients with lab for Hemocue
microalbuminuria to be albuminuria to be Shaikh A, et al. Comparison between Analytical Moderate Recommendation down-
useful for screening. useful for screen- immunoturbidimetry, size-exclusion study graded for indirectness of
Positive results must be ing. Positive results chromatography, and LC-MS to analytical data to clinical
confirmed by analysis in must be confirmed quantify urinary albumin. Clin Chem outcomes
an accredited laboratory by analysis in an ac- 2008;54 (9):1504-1510
Level E credited laboratory
GPP
11.e Currently available New recom- Gansevoort RT, et al. The validity of Observa- Moderate Moderate There is no convincing
dipstick tests do not mendation screening based on spot morning urine tional study evidence in multiple studies
have adequate ana- according samples to detect subjects with micro- for any specific test achieving
lytical sensitivity to to recent albuminuria in the general population. >95% diagnostic sensitivity in
detect albuminuria literature on Kidney Int.Suppl 2005; (94):S28-S35. two or more different studies.
B (moderate) the topic Incerti J, et al. Evaluation of tests for Observa- Moderate Due to this, no specific
microalbuminuria screening in patients tional study screening test can be recom-
with diabetes. Nephrol Dial. Transplant. mended. Dipstick tests for
2005;20(11):2402-2407. microalbuminuria cannot be
Davidson MB, et al. ImmunoDip: Observa- Moderate recommended as replace-
an improved screening method for tional study ment for the quantitative
microalbuminuria. Am J Nephrol tests.
2004;24:284-8.
Meinhardt U, et al. Microalbumin-uria Observa- Moderate
in diabetes mellitus: efficacy of a new tional study
screening method in comparison with
timed overnight urine collection. J Diab
Comp-lications 2003;17 (5): 254-257.
Fernandez Fernandez I, et al. Rapid Observa- Moderate
screening test evaluation for microal- tional study
buminuria in diabetes mellitus. Acta
Diabetol 1998; 35:199-202
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 11: LOW LEVELS OF ALBUMINURIA (FORMERLY MICROALBUMINURIA) (Cont'd)
No 1. NACB 2002 recom- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
mendation and its dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
Appendix

grade(1) mendation with its to modify (high-moder- (high-moder-

grade and quality of the recom- ate-low) ate-low-very
evidence(2) mendation? low)
Leong SO, et al. The use of semi-quan- Randomized Moderate
titative urine test-strip (Micral Test) for trial
microalbuminuria screening in patients
with diabetes mellitus. Singapore Med
J 1998;39:101-3.
Poulsen PL, et al. Evaluation of a dip- Observa- Low
stick test for micro-albuminuria in three tional study
different clinical settings, including the
correlation with urinary albumin excre-
tion rate. Diabetes Metab 1992;18:395-
400.
(3)
WHAT ARE THE PREANALYTICAL CONSIDERATIONS WHEN TESTING FOR LOW LEVELS OF ALBUMINURIA? Priority: 3 (A3-4)
11.f Acceptable samples Acceptable samples to No change, Lambers Heerspink HJ, et al. Com- Prospective High Moderate The albumin:creatinine
to test for increased test for increased uri- but new parison of different measures of cohort ratio is the superior method
urinary albumin excre- nary albumin excretion evidence urinary protein excretion for prediction to predict renal events
tion are timed (e.g., 12 are timed collections supports of renal events. J Am Soc Nephrol in patients with type 2
or 24 hour) collections (e.g., 12 or 24 h) for recommen- 2010;21:1355-60 diabetes
for measurement of measurement of the dation Ibsen H, et al. Reduction in albuminuria Observa- Moderate
albumin concentra- albumin concentration translates to reduction in cardiovas- tional study
tion and timed or and timed or untimed cular events in hypertensive patients:
untimed samples for samples for measure- losartan intervention for end point
measurement of the ment of the albumin reduction in hypertension study. Hyper-
albumin:creatinine creatinine ratio tension 2005;45:198-202.
ratio. For screening, B (moderate)
an untimed sample for Gansevoort RT, et al. The validity of Observa- Moderate
albumin measurement screening based on spot morning urine tional study
(without creatinine) samples to detect subjects with micro-
may be considered if a albuminuria in the general population.
concentration cutoff is Kidney Int.Suppl 2005;(94):S28-S35
used that allows high Meinhardt U, et al. Microalbumin-uria Observa- Moderate
sensitivity for detection in diabetes mellitus: efficacy of a new tional study
of an increased albumin screening method in comparison
excretion rate. with timed overnight urine collectionJ
Level E Diabetes Compli-cations 2003;17
(5):254-257
Hishiki S, et al. Circadian variation of Observa- Low
urinary microalbumin excretion and tional study
ambulatory blood pressure in patients
with essential hypertension. J Hyper-
97

tens 1998;16:2101-8.
No 1. NACB 2002 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments 98
recommendation new recommendation with it necessary new recommendation design evidence(2) evidence(2)
and its grade(1) its grade and quality of to modify (high-moder- (high-moder-
evidence(2) the recom- ate-low) ate-low-very
mendation? low)
Howey JE, et al. Biologic variation of Observa- Moderate
urinary albumin: consequences for tional study
analysis, specimen collection, inter-
pretation of results, and screening pro-
grams. Am J Kidney Dis 1989;13:35-7.
(3)
WHAT IS THE OPTIMAL TIME OF DAY TO MEASURE ALBUMINURIA? Priority: 2 (A2-4)
11.g The optimal time for spot New recom- Witte EC, Lambers Heerspink HJ, de Prospective Moderate Low Collected three different
urine collection is the mendation Zeeuw D, Bakker SJ, de Jong PE, and non-random- urines and analyzed in
early morning. All collections Gansevoort R. First morning voids are ized three different ways.
should be at the same time more reliable than spot urine samples One study only that
of day to minimize variation. to assess microalbuminuria. J Am.Soc. investigates this topic.
The patient should not have Nephrol. 2009;20 (2):436-443. Recommendation down-
ingested food within the pre- graded for indirectness
ceding 2 h but should be well of evidence and lack of
hydrated (i.e., not volume more data.
depleted).
GPP
(3)
HOW FREQUENTLY ALBUMINURIA SHOULD BE MEASURED? Priority: 1 (A5, A1-2)
11.h Low urine albumin concentra- New recom- Levey AS, et al. The definition, classifi- Consensus Moderate Moderate Strong consensus of
tions (i.e., <30 mg/g creati- mendation cation and prognosis of chronic kidney report experts upgraded the
nine) are not associated with disease: a KDIGO Controversies Con- recommendation
high cardiovascular risk if ference report. Kidney Int 2011;in press
the eGFR is >60 mL . min21. Yuyun MF, et al. Micro-albuminuria Prospective Moderate
(1.73 m2)21 and the patient is independently predicts all-cause and cohort
normotensive. If the eGFR is cardiovascular mortality in a British
<60 mL . min21. (1.73 m2)21 population: The European Prospective
and/or the level of albumin- Investigation into Cancer in Norfolk
uria is 30 mg/g creatinine (EPIC-Norfolk) population study. Int.J
on a spot urine sample, a Epidemiol. 2004;33 (1):189-198
repeat measurement should
be taken within the year
to assess change among
people with hypertension
A (moderate)
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
Chapter 12: MISCELLANEOUS POTENTIALLY IMPORTANT ANALYTES
No 1. NACB 2002 rec- 2. NACB 2011 up- 3. Why was it 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
ommendation and dated/new recom- necessary to new recommendation design evidence(2) evidence(2)
Appendix

its grade(1) mendation with its modify the rec- (high-moder- (high-moder-
grade and quality of ommendation? ate-low) ate-low-very
evidence(2) low)
IS THERE A ROLE FOR MEASUREMENT OF INSULIN AND C-PEPTIDE CONCENTRATIONS TO DISTINGUISH TYPE 1 FROM TYPE 2 (3)Priority: 2 (A3-4)
DIABETES MELLITUS?
12.a There is no role for There is no role for Changed word- Rutter MD, et al. Use of Alternative Cohort study Moderate Moderate Models of predictive baseline
routine testing for routine testing for ing. thresholds defining insulin resistance measures of insulin resis-
insulin, C-peptide, insulin, C-peptide, Many groups, to predict incident type 2 diabetes tance (which include mea-
or proinsulin in or proinsulin in most including ADA, and cardiovascular disease. Circula- sures of insulin) in a large
most patients with patients with diabe- are moving tion. 2008;117:1003-1009. population. Surrogate IR
diabetes. Differentia- tes. Differentiation beyond the cat- measures (which all included
tion between type 1 between type 1 and measures of insulin) had
egorical concept
and type 2 diabetes type 2 diabetes may modest performance at the
(diagnosis)
may, in most cases, be made in most 76th centile, with no thresh-
of metabolic
be made based on cases on the basis of syndrome to that old effects. Prediction was
the clinical presenta- the clinical presenta- of continuous particularly poor for CVD.
tion and subsequent tion and the subse- Cohort study Moderate Models of predictive baseline
and more global Wilson PW et al. Prediction of
course. There is no quent course. These measures of risk incident diabetes mellitus in middle- values in a large popula-
role for measurement assays are useful tion. Factors easily obtain-
for diabetes and aged adults: The Framingham
of insulin concentra- primarily for research cardiovascular Offspring Study. Arch Intern Med able on history, exam, or
tion in the diagnosis purposes. Occasion- disease. 2007;167:1068-74. standard lab tests (glucose,
of the metabolic ally, C-peptide mea- lipids) predicted incident DM
syndrome because surements may help strongly. Addition of more
knowledge of this distinguish type 1 complex factors, including
value does not alter from type 2 diabetes fasting insulin, did not add
the management of in ambiguous cases, significantly.
these patients. such as patients who
Level E have a type 2 phe- Despres J-P et al. Hyperinsulinemia Case-control Moderate Case-control study looking at
notype but present in as an independent risk factor for study baseline fasting insulin levels
ketoacidosis ischemic heart disease. N Engl J in Quebec Heart Study. High
B (moderate) Med 1996;334:952-7. fasting insulin levels ap-
peared to be an independent
risk factor for IHD. However,
only excluded clinically diag-
nosed DM (in early 1990s,
probably many undiagnosed)
and did not adjust for any
measures of glycemia or BMI
99
 100
Chapter 12: MISCELLANEOUS POTENTIALLY IMPORTANT ANALYTES (Cont'd)
No 1. NACB 2002 recommen- 2. NACB 2011 updated/ 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
dation and its grade(1) new recommendation it necessary new recommendation design evidence(2) evidence(2)
with its grade and to modify the (high-moder- (high-moder-
quality of evidence(2) recommenda- ate-low) ate-low-very
tion? low)
12.a These assays are use- These assays are useful New evidence Balasubramanyam A et al. Accuracy Obser- Moderate Moderate- Investigation of
ful primarily for research primarily for research regarding us- and predictive value of classification vational low patients presenting
purposes and, in rare cases, purposes. Occasionally, ing C-peptide schemes for ketosis-prone diabetes. prognostic/ with ketosis, with
to identify patients with an C-peptide measure- to clarify Diab Care 2006; 29:2575-9. diagnostic absent or preserved
absolute requirement for ments may help distin- diagnosis study C-peptide function
insulin before switching guish type 1 from type 2 at one year the
to oral agents, or to assist diabetes in ambiguous outcome.
patients in obtaining insur- cases, such as patients Unclear how direct
ance coverage for continu- who have a type 2 the outcome is,
ous subcutaneous infusion phenotype but present whether this is better
pumps. in ketoacidosis. than current care
Level E B (moderate)
A possible role for measure- None Prior recom- American College of Obstetrics and Guideline/ Low Very low Prior recommen-
ment of fasting insulin or the mendation Gynecology. ACOG practice bulletin. Expert con- dation was also
assessment of insulin re- deleted. No Polycycstic ovary syndrome. Number sensus supported by expert
sistance is in the evaluation evidence that 41, December 2002. Int J Gynecol opinion only
of patients with polycystic this is better Obstet 2003; 80:335-48
ovary syndrome who may than clinical
be candidates for treatment evaluation for
aimed at lowering insulin signs of insulin
resistance in the absence resistance; not
of overt diabetes or glucose recommended
intolerance by ACOG or
Level E other groups.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 2. NACB 2011 up- 3. Why was 4. Key references supporting the new 5. Study 6. Level of 7. Quality of 8. Comments
recommendation dated/new recom- it necessary recommendation design evidence(2) evidence(2)
and its grade(1) mendation with its to modify (high-moder- (high-moder-
grade and quality of the recom- ate-low) ate-low-very
Appendix

evidence(2) mendation? low)

(3)
IS THERE A ROLE FOR MEASUREMENT OF INSULIN CONCENTRATIONS OR INDIRECT MEASURES OF INSULIN RESISTANCE IN THE Priority: 2 (A3)
ASSESSMENT OF PATIENTS CARDIOMETABOLIC RISK OR TO DETERMINE USE OF INSULIN SENSITIZING DRUGS IN DIABETIC OR
NON-DIABETIC PATIENTS?
12.b There is no role for New recom- Rutter MD, et al. Use of Alternative Cohort study Moderate Moderate
measurement of mendation thresholds defining insulin resistance
insulin concentration to predict incident type 2 diabetes and
in the assessment of cardiovascular disease. Circulation.
cardiometabolic risk, 2008;117:1003-9.
because knowledge Wilson PW et al. Prediction of incident Cohort study Moderate
of this value does not diabetes mellitus in middle-aged adults:
alter the management The Framingham Offspring Study. Arch
of these patients Intern Med 2007;167:1068-74.
B (moderate)
Despres J-P et al. Hyperinsulinemia Case-control Moderate
as an independent risk factor for study
ischemic heart disease. N Engl J Med
1996;334:952-7.
(3)
DO INSULIN MEASUREMENTS NEED TO BE HARMONIZED? Priority: 2 (A3)
12.c Because current mea- Staten M, et al, for the Insulin Standard- Expert con- Low Low Commentary summarizes the
sures of insulin are ization Workgroup. Insulin assay stan- sensus above papers and calls for
poorly harmonized, a dardization: leading to measures of insulin a standardized insulin assay
standardized insulin sensitivity and secretion for practical based on above.
assay should be clinical care. Diab Care 2010;33:205-6
developed to encour- Miller WG, et al for the Insulin Standard- Investigation Moderate Most assays can achieve
age the development ization Work Group. Toward standardiza- of alternate consistent performance with
of measures of insulin tion of insulin immunoassays. Clin Chem preparation for calibration traceability based
sensitivity that will be 2009;55:1011-8 insulin refer- on individual serum samples
practical for clinical ence materials with insulin concentrations
care set by isotope dilution mass
GPP spectrometry.
Marcovina S, et al. Standardization of Comparison Moderate Current FDA-approved com-
insulin immunoassays: report of the of different mercially available insulin as-
American Diabetes Association Work- insulin assays says provide a wide range of
group. Clin Chem 2007; 53:711-6 currently on values for the same samples.
the market in There clearly is a need to
the US. standardize the reference
system and protocols to
enable all assays to achieve
consistent and uniform
results and to report insulin in
101

identical units.
Chapter 12: MISCELLANEOUS POTENTIALLY IMPORTANT ANALYTES (Cont'd) 102

No 1. NACB 2002 recom- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
mendation and its dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
grade(1) mendation with its to modify the (high-moder- (high-moder-
grade and quality of recommenda- ate-low) ate-low-very
evidence(2) tion? low)
(3)
IS THERE A ROLE FOR INSULIN AUTOANTIBODY TESTING IN MANAGING PATIENTS WITH DIABETES MELLITUS? Priority: NOT LISTED
12.d There is no published There is no published No change Bingley PJ,et al. Measurement of Analytical test Moderate Very low International workshops
evidence to support evidence to support islet cell antibodies in the Type 1 evaluation using serum exchange ex-
the use of insulin an- the use of insulin Diabetes Genetics Consortium: ercises provide measures of
tibody testing for rou- antibody testing for efforts to harmonize procedures inter-laboratory variation.
tine care of patients routine care of pa- among the laboratories. Clin Trials Standardization was pos-
with diabetes tients with diabetes. 2010;7(1 Suppl):S56-64. sible between three expert
Level E C (very low) Bonifacio E, et al Harmonization of Analytical test Moderate laboratories.
glutamic acid decarboxylase and evaluation Quality of evidence and
islet antigen-2 autoantibody assays strength of recommendation
for national institute of diabetes are downgraded for indirect-
and digestive and kidney diseases ness.
consortia. J Clin Endocri-nol Metab.
2010;95(7):3360-7.
Trn C, et al. Participating Laborato- Analytical test Moderate
ries. Diabetes Antibody Standardiza- evaluation
tion Program: evaluation of assays
for autoantibodies to glutamic acid
decarboxylase and islet antigen-2.
Diabetologia. 2008;51(5):846-52.
Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus
No 1. NACB 2002 recom- 2. NACB 2011 up- 3. Why was 4. Key references supporting the 5. Study 6. Level of 7. Quality of 8. Comments
mendation and its dated/new recom- it necessary new recommendation design evidence(2) evidence(2)
grade(1) mendation with its to modify the (high-moder- (high-moder-
grade and quality of recommenda- ate-low) ate-low-very
Appendix

evidence(2) tion? low)

(3)
IS THERE A ROLE FOR AMYLIN AND LEPTIN TESTING IN MANAGING PATIENTS WITH DIABETES MELLITUS? Priority: NOT LISTED
Assays for amylin are None The evidence
not clinically useful in accumulated
the management of in the last six
diabetes. These stud- to seven years
ies should be confined has failed to
to the research setting identify any
Level E clinical value
in measuring
these analytes
in patients with
diabetes.
Routine measure- None Recommenda-
ment of plasma leptin tion removed
concentrations is not for reasons
of value at this time mentioned
for the evaluation above
or management of
patients with diabetes
or obesity
Level E
103
104