British Journal of Anaesthesia, 130 (3): 262e271 (2023)

doi: 10.1016/j.bja.2022.12.010
Advance Access Publication Date: 25 January 2023
Review Article

CLINICAL PRACTICE

A systematic review of perioperative clinical practice guidelines for

care of older adults living with frailty
Jake S. Engel1, Jason Tran1, Noha Khalil2, Emily Hladkowicz2, Manoj M. Lalu2,3,4,5, Allen Huang4,6,
Camilla L. Wong7, Brian Hutton2,5, Jugdeep K. Dhesi8,9 and Daniel I. McIsaac2,3,4,5,*
1
Faculty of Medicine, University of Ottawa, Ottawa, Canada, 2Clinical Epidemiology Program, Ottawa Hospital Research
Institute, Ottawa, Canada, 3Department of Anesthesiology & Pain Medicine, University of Ottawa, Ottawa, Canada, 4The
Ottawa Hospital, Ottawa, Canada, 5School of Epidemiology and Public Health, University of Ottawa, Ottawa,
Canada, 6Division of Geriatric Medicine, University of Ottawa, Ottawa, Canada, 7St. Michael’s Hospital, Division of
Geriatric Medicine, University of Toronto, Toronto, Canada, 8Guy’s and St. Thomas Hospital, King’s Partners Health Trust,
London, England, UK and 9King’s College London, London, England, UK

*Corresponding author. E-mail: [email protected]

Abstract
Background: Perioperative frailty is prevalent and requires complex management, which could be guided by clinical
practice guidelines (CPGs). The objective of this systematic review was to identify and synthesise CPGs that provide
perioperative recommendations specific to older adults living with frailty.
Methods: After protocol registration, we performed a systematic review of CPGs. MEDLINE, Embase, CINAHL, and 14 grey
literature databases were searched (January 1, 2000 until December 22, 2021). We included all CPGs that contained at least
one frailty-specific recommendation related to any phase of the perioperative period. We compiled all relevant recom-
mendations, extracted underlying strength of evidence, and categorised them by perioperative phase of care. Within
each phase, recommendations were synthesised inductively into themes. Quality of CPGs was assessed using the
Appraisal of Guidelines for Research and Evaluation (AGREE) II instrument.
Results: From 4707 citations, 13 guidelines were included; 8/13 were focused on the perioperative care of older surgical
patients in general. Among 110 recommendations extracted, 37 themes were generated, with the majority pertaining to
preoperative care. Four themes were supported by strong evidence: performing preoperative frailty assessments, using
multidimensional frailty instruments, reducing urinary catheter use, and following multidisciplinary care and
communication throughout the perioperative period. Per AGREE II, most guidelines (8/13; 62%) were recommended for
use with modifications.
Conclusions: Despite increasing numbers of patients living with frailty, few guidelines exist that address frailty-specific
perioperative care. Given the lack of strong evidence-based recommendations, particularly outside the preoperative
period, high-quality primary research is required to underpin future guidelines and better inform the care of older
surgical patients with frailty.
Systematic review protocol: PROSPERO CRD42022320149.

Keywords: comprehensive geriatric assessment; CPGs; frail; guidelines; surgery; surgical

Received: 31 August 2022; Accepted: 11 December 2022

© 2022 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.
For Permissions, please email: [email protected]

262
 Perioperative frailty guidelines systematic review - 263

Study eligibility
Editor’s key points We included all peer-reviewed publications, institutional
publications, and working papers that were consistent with
 Frailty is common in surgical patients and is associ-
the Institute of Medicine (IOM) definition of CPGs (see
ated with the risk of adverse outcomes.
Supplementary Appendix 2 for criteria).9 Given that few
 The authors conducted a systematic review of prac-
guidelines adhered to all IMO criteria, we included those that
tice guidelines for perioperative patients with frailty.
met at least 50% of criteria. Guidelines were then included if
High-quality and consistent recommendations sup-
they contained at least one recommendation relevant to older
ported preoperative, multidimensional frailty
adults living with frailty in any phase of the perioperative
assessment, avoidance of urinary catheterisation,
period and were published during or after the year 2000. This
and multidisciplinary perioperative care and
date was chosen as the two main conceptual frailty frame-
communication.
works (phenotype and accumulating deficits) were first pub-
 Researchers should develop and evaluate
lished in 2001.2,11 Guidelines were excluded if they only
intraoperative and postoperative interventions for
discussed the management of distinct, but related conditions,
surgical patients with frailty. Future efforts should
such as multimorbidity, disability, or where frailty was
aim to include patients and caregivers and should
mentioned but not explicitly incorporated into a guideline. In
focus on implementation.
addition, CPGs describing the assessment of frailty diagnostic
or screening tools without making explicit recommendations
on their use were excluded.
Frailty is a multidimensional state attributed to the accumu-
lation of age- and disease-related deficits.1,2 The prevalence of
frailty increases exponentially with age, and older people Search strategy
living with frailty are vulnerable to stressors.3 Surgery is a A peer-reviewed systematic search strategy12 was developed
major stressor and is commonly undertaken in older adults, by an information specialist and applied (on December 22,
who represent the fastest growing segment of the surgical 2021) to MEDLINE (Ovid), Embase, and CINAHL (see search
population.4 Given the stress of surgery and vulnerability strategies, Supplementary Appendix 3). In addition, best
inherent in living with frailty, adverse outcomes are common, practice structured searches of 14 grey literature databases
including a greater than two-fold increase in rates of (Google [‘perioperative frailty guideline’]; Grey Matters: Trip-
morbidity, mortality, loss of independence, and patient- database, Emergency Care Research Institute [ERCI], Guideline
reported disability.5e7 Health resource use is also signifi- Central, NHS Evidence/National Institute for Health and Care
cantly increased.8 Excellence [NICE], Toward Optimized Practice, British
Clinical practice guidelines (CPGs) are systematically Columbia [BC] Guidelines, Canadian Medical Association
developed statements to support delivery of quality patient [CMA] Infobase, College of Physicians and Surgeons of Ontario
care informed by systematic review of evidence, and an [CPSO], Nursing Best Practice Guidelines, Academy of Medi-
assessment of benefits and harms of alternative care options.9 cine of Malaysia, Best Practice Advocacy Centre New Zealand,
Review and synthesis of CPGs allows current recommenda- French National Authority for Health, and National Health and
tions to be systematically identified, compared, and assessed Medical Research Council [NHMRC]) were applied.13 Refer-
for quality. The resulting new knowledge can be used to ences of eligible studies were screened to identify records that
identify strongly supported and widely recommended prac- may have been missed from the original search. Non-English
tices, further refine future guidelines, and describe the key guidelines were included after online translation tools were
knowledge gaps to be addressed through future research. used to provide English translations.
Our overarching objective was to systematically identify
and synthesise existing CPGs that (1) provide perioperative
recommendations specifically for older patients living with Study selection
frailty, or (2) contain frailty-specific recommendations within We used DistillerSR® (Evidence Partners, Ottawa, ON, Can-
a guideline related to perioperative care. Our specific objec- ada), a web-based systematic review software, to manage our
tives included identification and synthesis of relevant rec- screening process. The process to review titles and abstracts
ommendations from CPGs, collation of recommendations used the DistillerAI machine-learning algorithm, based on
within each phase of perioperative care, appraisal of the recommendations for use of artificial intelligence in knowl-
strength of evidence underlying recommendations, and edge synthesis.14 A full description of the screening process is
appraisal of the quality of existing guidelines. Furthermore, we provided in Supplementary Appendix 4. Briefly, after creating
aimed to identify the specific actors and actions targeted by a training set from independent duplicate review of the first
guideline-based recommendations. 50% of citations, we followed a liberal accelerated screening
approach where a human reviewer was involved in every de-
cision to exclude any citation, but the DistillerAI algorithm
Methods
could advance a citation to full text review. The full text review
We conducted a systematic review of CPGs informed by rec- stage was completed in duplicate by two independent re-
ommended best practices.10 A detailed protocol was developed viewers (JSE and JT) with conflicts resolved by the senior
a priori and registered in the International Prospective Register investigator (DIM).
of Systematic Reviews (PROSPERO ID: CRD42022320149). Pro-
tocol amendments are described in Supplementary Appendix
Data extraction
1. This systematic review is reported using the Preferred
Reporting Items for Systematic Review and Meta-analyses We extracted data using Microsoft Excel® (Microsoft Corpo-
(PRISMA) guidelines. ration, Redmond, WA, USA) after piloting forms designed for
 264 - Engel et al.

this review. Three independent reviewers extracted data. We were different by >2 points on the 7-point scale. These were
extracted background information from each guideline reviewed with the senior author (DIM) and a consensus score
including the focus of the guideline, publication information, was applied.
and type and urgency of surgical intervention. In addition, A recommendation for guideline use was provided based
reviewers extracted all frailty-related perioperative recom- on the overall assessment of each guideline. As one of the
mendations from the included citations. Recommendations study authors was also involved in the production of several
that adhered to at least one of the following criteria were included guidelines (JKD), all quality assessments were
extracted: (1) all those related to any phase of the perioperative completed, reviewed, and finalised independent of this
period from frailty-specific guidelines, or (2) all those related to author.
frailty from perioperative-targeted guidelines. Perioperative
recommendations were categorised using the Action, Actor,
Context, Target, and Time (AACTT) framework with the Time
Results
domain as the primary consideration.15 Where reported, the Study selection
strength of evidence for each recommendation was extracted
After removal of duplicates, 4707 citations underwent initial
as reported by the authors of the guidelines. A second reviewer
review. Thirteen guidelines met inclusion criteria (Fig 1).
independently verified each extracted data point for accuracy
and completeness with discrepancies resolved by the senior
investigator (DIM). Guideline characteristics
Descriptive and background information for each included
Data synthesis guideline is summarised in Table 1. Guidelines were published
between 2012 and 2022, and originated from the UK,19e23
Informed by methodological considerations for systematic
Italy,24e26 Austria,27 the USA,28e30 or from a combination of
reviews of practice guidelines,10 we produced tabular sum-
European countries.31 Three (23%)20,21,29 guidelines were
maries of recommendations within each perioperative phase
frailty-specific, among which two (15%)20,29 were specific to
of care (i.e. pre-, intra-, post-, or perioperative). To further
perioperative care. The third21 frailty-specific guideline was
summarise results, we used thematic analysis to synthesise
focused on inpatient diabetes management but included
recommendations into themes within each phase of the
perioperative recommendations. The remaining 10
perioperative period; inductive and deductive approaches
(77%)19,22e24,26e28,30,31 included guidelines were perioperative-
were used. Specifically, the first author followed recom-
focused, mostly on the care of older surgical patients. Three
mended methods for qualitative analysis: (1) familiarisation
(23%)24,26,28 guidelines were specific to general surgery, and
with data; (2) generation of initial codes within each periop-
one guideline (8%)23 was targeted for orthopaedic surgery; no
erative period; (3) search for themes; (4) review of themes with
other guidelines were intended for a specific surgical
the senior author; (5) define and name themes; (6) reporting of
population.
themes.16 Generation of these themes was inductive and
informed by the available data. Next, each theme was deduc-
tively characterised by at least one of the following actions Quality assessment
based on the AACTT framework: assessment, therapy, moni- The appraisal of guidelines using the AGREE II framework17,18
toring, and disposition. Given that most guidelines did not is summarised in Supplementary Appendix 5. Overall, four
routinely address actors, we focused only on actions. We guidelines19e21,27 were recommended for use, eight guide-
noted each synthesised theme that was supported by high- lines23e26,28e31 were recommended for use with modifications,
quality evidence; that is a theme that was developed and and one guideline22 was not recommended for use. Upon
was based on at least one recommendation supported by comparison of domains, most guidelines demonstrated high
strong evidence according to the quality assessment scale scores in editorial independence (median [range]: 100%
used by the respective guideline. Finally, we identified themes [92e100]), scope and purpose (92% [61e97]), and clarity of
that were recommended by more than two guidelines, which presentation (89% [58e94]). Conversely, scores were generally
we thought would be worthwhile in highlighting where over- lower in applicability (50% [31e75]), stakeholder involvement
lapping agreements existed between included CPGs. (58% [39e92]), and rigour of development (60% [17e82]).

Study quality Synthesis of recommendations

We used the Appraisal of Guidelines for Research and Evalu- We identified 110 total recommendations from the 13 included
ation (AGREE) II17 framework to assess each guideline. The guidelines. Recommendations were synthesised into 37
AGREE II tool consists of 23 items, each receiving a score of 1 themes, with four (11%) themes representing recommenda-
(strongly disagree) to 7 (strongly agree). Items are organised tions that were supported by strong evidence.
into six domains: (1) scope and purpose, (2) stakeholder
involvement, (3) rigour of development, (4) clarity of presen-
Preoperative period
tation, (5) applicability, and (6) editorial independence. Scores
for each of the six domains in each guideline were calculated Sixteen themes were produced from 68 recommendations
as per the AGREE II user manual,18 based on the evaluation of specific to the preoperative period. Most preoperative themes
two independent appraisers, who were trained using the (13/16; 81%) related to the action of preoperative patient
AGREE II user’s manual, as recommended.17 As the AGREE II assessment. The two themes supported by strong evidence
guidance suggests that between two and four reviewers were recommendations to perform a frailty assessment dur-
participate, but provides no specific approach to discordant ing preoperative evaluation of older adults, and when doing so
assessments, we identified any ratings where reviewer scores to use a validated, multidimensional frailty instrument. Other
 Perioperative frailty guidelines systematic review - 265

Identification of studies via databases and registers

Identification Records identified

Records removed before
screening:
Databases (n=2366)
Duplicate records removed
Other sources (n=2739)
(n=398)

Records screened Records excluded after title and

(n=4707) abstract screening (n=4505)
Screening

Reports sought for retrieval

Reports not retrieved (n=0)
(n=202)

Reports assessed for eligibility Reports excluded with reasons

(n=202) (n=189):
Not in keeping with Institute
of Medicine definition of
Guidelines (n=105)
No perioperative frailty-related
recommendations (n=69)
Duplicate records (n=15)
Included

Studies included in review

(n=13)

Fig. 1. Study selection process.

preoperative themes recommended by more than two guide- assessment of risk and prevention of delirium by a multidis-
lines were optimisation of frailty through assessment and ciplinary team in the preoperative period; however, none of
referral of comorbidities, referral of those living with frailty for these was supported by strong evidence (Table 2;
comprehensive geriatric assessment (CGA), and initiating Supplementary Appendix 6).

Table 1 Patient characteristics and study characteristics of included guidelines. *Listed as both elective and emergent if not specified
by the guideline.

Authors, year Location of Frailty-specific Perioperative Surgery urgency* Surgical

origin guideline phase subspecialty
of care

Beecroft and colleagues, 202219 UK No Mixed Elective and Emergent Not specified
Partridge and colleagues, 202120 UK Yes Mixed Elective and Emergent Not specified
Peden and colleagues, 202128 USA No Preoperative Emergent General surgery
Podda and colleagues, 202124 Italy No Mixed Elective and Emergent General surgery
Griffiths and colleagues, 202123 UK No Mixed Emergent Orthopaedic surgery
Aceto and colleagues, 202025 Italy No Mixed Elective Not specified
Sinclair and colleagues, 202021 UK Yes Mixed Elective and Emergent Not specified
Pisano and colleagues, 201926 Italy No Preoperative Elective General surgery
Alvarez-Nebreda and USA Yes Preoperative Elective Not specified
colleagues, 201829
De Hert and colleagues, 201827 Austria No Preoperative Elective Not specified
Kozek-Langenecker and Europe No Mixed Elective and Emergent Not specified
colleagues, 201731
Griffiths and colleagues, 201422 UK No Mixed Elective and Emergent Not specified
Chow and colleagues, 201230 USA No Preoperative Elective Not specified
 266 - Engel et al.

Table 2 Synthesised preoperative themes for older adults living with frailty. Recommendations used to synthesise preoperative
themes can be found in Supplementary Appendix 6. *A theme supported by high-quality evidence is classified as one which is sup-
ported by at least one recommendation carrying strong evidence based on the quality assessment scales used by the respective
guidelines. CGA, comprehensive geriatric assessment.

Themes Action Number of related Number of Theme

recommendations guidelines with supported by
recommendation high-quality
evidence (Yes/
No)*

Preoperative evaluation of functional status for Assessment 6 5 Yes

patients 65 should include a frailty
screen19,20,24,27,29
Frailty assessment should be completed using Assessment 10 8 Yes
validated multimodal scales19,20,24e28,30
Optimise older patients living with frailty for Assessment 6 3 No
surgery by assessing for comorbidities and
making appropriate referrals19e21
A positive frailty screen should be followed a Assessment 4 3 No
CGA19,20,29
Shared-decision making regarding considerations Assessment 5 2 No
of the risk, benefits, and alternative management
options should begin upon contemplation of
surgery20,29
Early discussion is required to establish who will be Assessment 9 1 No
involved in medical decision-making and
provide support20
Medications should be optimised using START/ Assessment 1 1 No
STOPP tool20
A detailed diabetic plan must be individualised, Assessment 3 1 No
consider pre-admission support, and
documented on admission21
Emergency staff should obtain a collateral history Assessment 2 1 No
and be aware of atypical presentations20
Assess patients living with frailty who suffer from Assessment 1 1 No
hip fractures with the Nottingham Hip Fracture
score23
Key domains of optimisation include Assessment 1 1 No
prehabilitation, shared-decision making, and and
interdisciplinary management29 therapy
Hypothermia should be assessed by ambulance Assessment 1 1 No
personnel and managed appropriately during and
transit22 therapy
Screening, assessment, and interventions used to Assessment 5 4 No
reduce the risk of postoperative cognitive and
disorders, such as delirium, should begin therapy
preoperatively and involve a multidisciplinary
team19,20,22,23
Education regarding multimodal prehabilitation Therapy 6 2 No
should be provided, including physical, lifestyle,
and psychological optimisation20,29
A watch-and-wait strategy can be used in select Monitoring 1 1 No
frail patients with low-rectal tumours24
Admission and discharge planning should begin in Disposition 7 2 No
the preoperative period through collaboration
with the various services involved in the
patient’s care20,21

Intraoperative and postoperative periods themes in either the intra- or postoperative periods were
recommended by more than two guidelines.
Eight themes were generated from eight recommendations
relevant to the intraoperative period, all of which were
designated as actions for patient therapy. One theme sup-
Perioperative period
ported by strong evidence endorsed limiting urinary catheter Six themes were developed from 25 recommendations related
use in patients living with frailty (Table 3; Supplementary to multiple phases of the perioperative period or to the peri-
Appendix 7). Seven themes were synthesised from nine post- operative period in general, all of which were associated with
operative recommendations, none of which were supported patient therapy. One theme was supported by strong evidence
by strong evidence (Table 4; Supplementary Appendix 8). No that suggested a multidisciplinary team be involved early and
 Perioperative frailty guidelines systematic review - 267

Table 3 Synthesised intraoperative themes for older adults living with frailty. Recommendations used to synthesise intraoperative
themes can be found in Supplementary Appendix 7. *A theme supported by high-quality evidence is classified as one which is sup-
ported by at least one recommendation carrying strong evidence based on the quality assessment scales used by the respective
guidelines. UTI, urinary tract infection.

Themes Action Number of related Number of Theme

recommendations guidelines with supported by
recommendation high-quality
evidence (Yes/
No)*

Ensure physiological homeostasis (normothermia, Therapy 1 1 No

MAP within 20% of preoperative range)20
Use strategies when moving and positioning patients Therapy 1 1 No
with frailty to protect the musculoskeletal and
integumentary systems20
Re-consider pharmacological strategies: avoid Therapy 1 1 No
anticholinergics, benzodiazepines, cyclizine,
tramadol; use glycopyrrolate instead of atropine20
Bilateral knee replacement should be avoided in Therapy 1 1 No
patients with frailty31
Avoid use of unnecessary urinary catheterisation or Therapy 1 1 Yes
remove catheters at the soonest appropriate time to
avoid hospital-acquired catheter-related UTIs20
Ensure senior surgeon and anaesthetist input, in Therapy 1 1 No
particular for emergency cases20
Consider local excision rather than Total Mesorectal Therapy 1 1 No
Excision for T1 rectal cancer24
Consider depth of general anaesthesia and use of Therapy and 1 1 No
regional anaesthesia to reduce postoperative opioid monitoring
use20

Table 4 Synthesised postoperative themes for older adults living with frailty. Recommendations used to synthesise postoperative
themes can be found in Supplementary Appendix 8. *A theme supported by high-quality evidence is classified as one which is sup-
ported by at least one recommendation carrying strong evidence based on the quality assessment scales used by the respective
guidelines. CGA, comprehensive geriatric assessment.

Themes Action Number of related Number of Theme

recommendations guidelines with supported by
recommendation high-quality
evidence
(Yes/No)*

Document individualised discharge criteria before Assessment 2 1 No

leaving recovery area and handover to ward20
Be aware of patients who develop frailty Assessment 1 1 No
postoperatively and may require CGA referral20 and
monitoring
Screen for delirium daily with validated tools and Assessment, 1 1 No
implement delirium guidelines20 therapy, and
monitoring
Timely discharge documentation for the primary Assessment, 2 2 No
care team should include, but is not limited to: monitoring,
new diagnoses, changes to current condition/ and
medications, in-hospital complications, and disposition
future care plans20,21
Use strategies to minimise hospital-acquired Therapy 1 1 No
deconditioning20
Multi-faceted interventions for venous Therapy 1 1 No
thromboembolism prophylaxis are
recommended after knee or hip replacements31
Monitor for common postoperative Monitoring 1 1 No
complications20
 268 - Engel et al.

communicate throughout all phases of perioperative care for care of older people living with frailty.20,29 The first from
older patients living with frailty. This theme was also the only Alvarez-Nebreda and colleagues,29 originating from the USA,
one recommended by more than two guidelines (Table 5; focused on preoperative assessment and optimisation, and
Supplementary Appendix 9). was limited to elective surgery. The second originated from
the Centre of Perioperative Care in the UK and the British
Geriatrics Society,20 which addressed both elective and
Discussion emergent surgeries and all phases of perioperative care.
This systematic review synthesised and evaluated 13 CPGs Although both provide important insights into key aspects of
that provided 110 recommendations regarding management perioperative care for older people living with frailty, including
of frailty in the perioperative setting. However, only two a focus on preoperative assessment using a multimodal in-
guidelines were explicitly targeted at the perioperative care of strument, these guidelines also highlight important gaps in
people living with frailty. The majority of recommendations the evidence base, including a lack of procedure-specific rec-
were specific to the preoperative period, and few were sup- ommendations, and a lack of strong evidence-based recom-
ported by high-quality evidence. Guideline quality varied mendations for care beyond the preoperative period.
substantially, with little focus on stakeholder inclusion or More than 60% of all recommendations and three of four
implementation of recommendations. Two distinct, but themes supported by strong evidence, were pertinent to the
related, gaps appear to be present in terms of perioperative preoperative period. Specifically, strong evidence and high-
CPGs for older people with frailty. First, primary research that quality guidelines recommend routine frailty assessment,
includes patients living with frailty are clearly required to use of a multidimensional tool when assessing frailty, and
address the lack of strong evidence-based, perioperative rec- initiation of multidisciplinary care when frailty was present.
ommendations for this surgical cohort. Second, guideline These preoperatively focused themes probably reflect several
development must adhere to best practices, including mean- important considerations. First, high-certainty evidence sup-
ingful stakeholder engagement and a focus on guideline ports a strong association between the presence of frailty
application.9 17 before surgery and a variety of adverse postoperative out-
Older patients are the fastest growing segment of the sur- comes.32,33 Second, consistent evidence demonstrates that
gical population,4 and accordingly, 11 of the 13 guidelines that preoperative frailty assessment is feasible and provides
we identified were targeted to the care of older surgical pa- unique prognostic information.7,34,35 Third, any clinical pro-
tients, regardless of frailty status. However, as frailty distin- cess aimed at improving outcomes for people with frailty must
guishes a clinically meaningful high-risk stratum of the older necessarily start with identification of frailty. Fourth, where
surgical population, it is important to recognise that there are implemented to date, evidence suggests that routine frailty
at least two guidelines specifically addressing perioperative assessment is associated with improved outcomes.34

Table 5 Synthesised perioperative themes for older adults living with frailty. Recommendations used to synthesise perioperative
themes can be found in Supplementary Appendix 9. *A theme supported by high-quality evidence is classified as one which is sup-
ported by at least one recommendation carrying strong evidence based on the quality assessment scales used by the respective
guidelines.

Themes Action Number of related Number of Theme

recommendations guidelines with supported by
recommendation high-quality
evidence
(Yes/No)*

Multidisciplinary teams should be involved Assessment and 8 4 Yes

early on in assessment and continue to therapy
communicate and collaborate daily on
management plans19,20,22,29
Ensure utilisation of appropriate aids and Therapy 5 1 No
strategies during planning of admission
and all phases of perioperative care for
patients with physical, sensory, or
cognitive impairments20
Perioperative analgesia should be Therapy 5 2 No
individualised using validated tools,
involve of a multidisciplinary team, and
attempt to limit opioid and NSAID use20,22
Adhere to Enhanced Recovery Therapy and 3 2 No
programmes20,24 monitoring
Diabetic patients with frailty should have Therapy and 3 1 No
blood glucose targets of 7.8e10 mmol L 1 monitoring
with monitoring of associated
complications during interventions21
Perioperative haemoglobin levels in older Therapy and 1 1 No
surgical patients living with frailty should monitoring
be kept above approximately 90 g L 123
 Perioperative frailty guidelines systematic review - 269

Together, themes supported by strong evidence further link methodological approaches to systematic reviews of CPGs.10
to themes included in 3 guidelines but that lacked strong Our search strategy was developed and peer-reviewed with
evidence, including use of CGA, concurrent assessment of the assistance of an information specialist, then applied to
comorbidities, and assessment of delirium risk. In fact, despite three major databases. Best practices were followed to review
lacking strong direct evidence in people with frailty, each of the grey literature, including structured searches of 14 rele-
these themes are supported by indirect evidence. CGA is an vant databases. However, use of an alternative approach to
evidence-based cornerstone in geriatric medicine that uses a grey literature searching could result in some different
multidimensional approach to diagnosis and assessment of guidelines being identified. Quality was assessed using a
comorbid, cognitive, and functional limitations to develop a validated instrument,17,18 and recommendations were syn-
personalised plan of care.36 This care plan can (and likely thesised using a validated framework specific to imple-
should) inform multidisciplinary perioperative care for people mentation of healthcare provider behaviour change (AACT).15
living with frailty.37 As more than two-thirds of people living All stages of screening, extraction, and synthesis were per-
with frailty also live with multimorbidity,38 CGA can further formed in duplicate, with consensus reached with the senior
guide assessment and treatment of co-existing comorbidities. author as required. Limitations were also present. First, our
Finally, frailty is associated with a greater than four-fold in- synthesised themes may not fully capture granular differ-
crease in delirium risk, and moderate-certainty evidence ences between recommendations that were combined within
supports the use of multi-component interventions to reduce themes. Next, although we pursued an extensive formal and
the risk of delirium in hospitalised patients.39 Moving forward, grey literature search, some institutional, jurisdictional, or
future guidelines would be enhanced by production and sub- society guidelines that were not peer reviewed or captured by
sequent inclusion of more direct evidence, but this will require web and repository searches could have been missed. Finally,
greater inclusion of surgical patients living with frailty, who our synthesis reflects recommendations provided in CPGs,
have largely been excluded from clinical trials to date.40e43 In which are directly informed by the availability of evidence
the interim, clinicians will continue to be relied upon to make specific to perioperative care of people living with frailty.
treatment decisions with individual patients under their care, Currently, there are recognised limitations in high-quality
and will need to integrate indirect evidence, opinion, and evidence supporting frailty-specific interventions, which
experience in decision-making processes. may highlight where generation of further evidence is
Although frailty-specific evidence continues to be gener- required.
ated, future guidelines and updates should also focus on
improving quality in the guideline development processes.
Using the AGREE II tool, we found that guidelines generally Conclusions
had well-defined objectives, clear methodology for developing Few clinical practice guidelines exist that specifically address
and organising recommendations, and lacked external influ- the perioperative management of older patients living with
ence. However, consistent with other reviews of CPGs,44,45 frailty, and most existing guidelines addressing care for older
applicability scores, which reflect inclusion of barriers and surgical patients with frailty have methodological limitations.
facilitators, advice and tools for implementation, resource Despite these gaps, there are strong and agreed-upon recom-
implications, and monitoring criteria, were generally low. This mendations to perform a preoperative frailty assessment us-
ultimately means that guidelines typically do not provide ing a multidimensional instrument, to reduce urinary catheter
guidance on how to execute and implement their recom- use, and to follow multidisciplinary care and communication
mendations. A lack of guidance may further reflect a lack of throughout the perioperative period. Moving forward, pro-
available data regarding barriers and facilitators specific to duction of high-quality guidelines to inform the care of sur-
perioperative care for people living with frailty46 and the gical patients living with frailty will require two, sequential
absence of core outcome sets specific to this population.47 To steps. First, there is a need to conduct high-quality primary
further support applicability, guideline developers should look research across the perioperative journey to better inform the
beyond the AGREE II tool and consult the AGREE- care of surgical patients living with frailty. Second, future
Recommendations Excellence (AGREE-REX), a validated tool guidelines developed to incorporate such evidence should
focused on ensuring that recommendations are credible and focus on improving and describing the applicability of their
implementable.48 Finally, given the importance of patient- recommendations, and include stakeholders, such as older
centred care amongst older people living with frailty, the individuals living with frailty, their families, and caregivers.
impact of guidelines would also likely be enhanced by greater
stakeholder involvement. Although guideline development
groups appeared to meaningfully involve individuals from Authors’ contributions
relevant professional groups, only three19e21 scored highly in
the stakeholder involvement domain, highlighting the need Conception: JSE, EH, JKD, DIM.
for more meaningful involvement of members of the target Study design: JSE, JT, MML, BH, DIM.
population, such as patients living with frailty and their Data acquisition: JSE, JT, NK, DIM.
Data analysis: JSE, DIM.
caregivers.
Data interpretation: JSE, JT, EH, MML, AH, CLW, JKD, DIM.
Drafting of the manuscript: JSE, JT, EH, MML, AH, CLW, BH, JKD,
Strengths and weaknesses
DIM.
Our synthesis of CPGs for perioperative management of older Revision of the manuscript: JSE, JT, EH, MML, AH, CLW, BH,
adults living with frailty should be considered based on its JKD, DIM.
strengths and limitations. Strengths include use of a protocol Approval of the final manuscript version: JSE, JT, EH, MML, AH,
that was pre-specified and registered, following best-practice CLW, BH, JKD, DIM.
 270 - Engel et al.

Acknowledgements Agency for Drugs and Technologies in Health, Ottawa,

http://www.cadth.ca; 2008. accessed 16 August 2022
We acknowledge Risa Shorr for her assistance in developing
13. Grey matters: a practical tool for searching health-related grey
and executing our search strategy. We also acknowledge
literature, 2018, Canadian Agency for Drugs and Technologies
support from The Ottawa Hospital Department of Anesthesi-
in Health; Ottawa https://www.cadth.ca/grey-matters-
ology & Pain Medicine for use of DistillerSR.
practical-tool-searching-health-related-grey-literature
(Accessed 16 August 2022).
14. Hamel C, Hersi M, Kelly SE, et al. Guidance for using
Declaration of interest
artificial intelligence for title and abstract screening while
No authors report any real or perceived conflicts of interest. conducting knowledge syntheses. BMC Med Res Methodol
2021; 21: 1e12
15. Presseau J, McCleary N, Lorencatto F, Patey AM,
Funding Grimshaw JM, Francis JJ. Action, actor, context, target,
DIM and MML receive salary support from The Ottawa Hos- time (AACTT): a framework for specifying behaviour.
pital Anaesthesia Alternate Funds Association and Clinical Implement Sci 2019; 14: 1e13
Research Chairs from the University of Ottawa Faculty of 16. Braun V, Clarke V. Using thematic analysis in psychology.
Medicine. MML receives further support from a Canadian Qual Res Psychol 2006; 3: 77e101
Anesthesiologists’ Society Career Scientist Award. 17. Brouwers MC, Kho ME, Browman GP, et al. Agree II:
advancing guideline development, reporting and evalua-
tion in health care. Can Med Assoc J 2010; 182: E839e42
Appendix A. Supplementary data 18. AGREE Next Steps Consortium, The AGREE II instrument
[Electronic version], 2009, Published http://www.agreetrust.
Supplementary data to this article can be found online at
org (Accessed 20 May 2022).
https://doi.org/10.1016/j.bja.2022.12.010.
19. Beecroft C, Joachim S, Gooneratne M, Connor D, Owen N.
Guidelines for the provision of anaesthesia services for
the perioperative care of elective and urgent care pa-
References
tients, 2021: 1e30. https://rcoa.ac.uk/sites/default/files/
1. Rockwood K, Song X, MacKnight C, et al. A global clinical documents/2021-06/GPAS-2021-02-PERIOPCARE_0.pdf.
measure of fitness and frailty in elderly people. Can Med [Accessed 25 April 2022].
Assoc J 2005; 173: 489e95 20. Partridge J, Dhesi J, Selwyn D. Guideline for perioperative care
2. Fried LP, Tangen CM, Walston J, et al. Frailty in older for people living with frailty undergoing elective and emergency
adults: evidence for a phenotype. J Gerontol A Biol Sci Med surgery. 1.0. Centre for Perioperative Care; 2021. Available
Sci 2001; 56: M146e56 from, https://www.bgs.org.uk/sites/default/files/content/
3. Rockwood K, Song X, Mitnitski A. Changes in relative CPOC-BGS-Guideline for Perioperative Care for People
fitness and frailty across the adult lifespan: evidence from Living with Frailty Undergoing Elective and Emergency
the Canadian National Population Health Survey. Can Med Surgery.pdf
Assoc J 2011; 183: E487e94 21. Sinclair AJ, Dashora U, George S, Dhatariya K. Joint British
4. Yang R, Wolfson M, Lewis MC. Unique aspects of the diabetes societies for inpatient care (JBDS-IP) clinical
elderly surgical population: an anesthesiologist’s guideline inpatient care of the frail older adult with dia-
perspective. Geriatr Orthop Surg Rehabil 2011; 2: 56e64 betes: an executive summary. Diabet Med 2020; 37: 1981e91
5. Lin HS, Watts JN, Peel NM, Hubbard RE. Frailty and post- 22. Griffiths R, Beech F, Brown A, et al. Peri-operative care of
operative outcomes in older surgical patients: a system- the elderly: association of anaesthetists of great britain
atic review. BMC Geriatr 2016; 16: 157 and Ireland. Anaesthesia 2014; 69: 81e98
6. Makary MA, Segev DL, Pronovost PJ, et al. Frailty as a 23. Griffiths R, Babu S, Dixon P, et al. Guideline for the man-
predictor of surgical outcomes in older patients. J Am Coll agement of hip fractures 2020: guideline by the Associa-
Surg 2010; 210: 901e8 tion of Anaesthetists. Anaesthesia 2021; 76: 225e37
7. McIsaac DI, Taljaard M, Bryson GL, et al. Frailty as a pre- 24. Podda M, Sylla P, Baiocchi G, et al. Multidisciplinary
dictor of death or new disability after surgery: a pro- management of elderly patients with rectal cancer: rec-
spective cohort study. Ann Surg 2020; 271: 283e9 ommendations from the SICG (Italian society of geriatric
8. Mrdutt MM, Papaconstantinou HT, Robinson BD, Bird ET, surgery), SIFIPAC (Italian society of surgical pathophysi-
Isbell CL. Preoperative frailty and surgical outcomes ology), SICE (Italian society of endoscopic surgery and new
across diverse surgical subspecialties in a large health technology. World J Emerg Surg 2021; 16: 35
care system. J Am Coll Surg 2019; 228: 482e90 25. Aceto P, Antonelli Incalzi R, Bettelli G, et al. Perioperative
9. Graham R, Mancher M, Wolman DM, Greenfield S, Management of Elderly patients (PriME): recommenda-
Steinberg E, editors. Clinical practice guidelines we can trust. tions from an Italian intersociety consensus. Aging Clin
Washington, DC: National Academies Press; 2011 Exp Res 2020; 32: 1647e73
10. Johnston A, Kelly SE, Hsieh S-C, Skidmore B, Wells GA. 26. Pisano M, Ceresoli M, Cimbanassi S, et al. 2017 WSES and
Systematic reviews of clinical practice guidelines: a SICG guidelines on acute calcolous cholecystitis in elderly
methodological guide. J Clin Epidemiol 2019; 108: 64e76 population. World J Emerg Surg 2019; 14: 10
11. Mitnitski AB, Mogilner AJ, Rockwood K. Accumulation of 27. De Hert S, Staender S, Fritsch G, et al. Pre-operative
deficits as a proxy measure of aging. Sci World 2001; 1: evaluation of adults undergoing elective noncardiac sur-
323e36 gery. Eur J Anaesthesiol 2018; 35: 407e65
12. Sampson M, McGowan J, Lefebvre C, Moher DGJ. PRESS: 28. Peden CJ, Aggarwal G, Aitken RJ, et al. Guidelines for
peer review of electronic search strategies. Canadian perioperative care for emergency laparotomy enhanced
 Perioperative frailty guidelines systematic review - 271

recovery after surgery (ERAS) society recommendations: comorbidity: implications for improved targeting and
Part 1dpreoperative: diagnosis, rapid assessment and care. J Gerontol Ser A Biol Sci Med Sci 2004; 59: M255e63
optimization. World J Surg 2021; 45: 1272e90 38. Vetrano DL, Palmer K, Marengoni A, et al. Frailty and
29. Alvarez-Nebreda ML, Bentov N, Urman RD, et al. Rec- multimorbidity: a systematic review and meta-analysis.
ommendations for preoperative management of frailty J Gerontol Ser A Biol Sci Med Sci 2019; 74: 659e66
from the society for perioperative assessment and 39. Siddiqi N, Harrison JK, Clegg A, et al. Interventions for
quality improvement (SPAQI). J Clin Anesth 2018; 47: preventing delirium in hospitalised non-ICU patients.
33e42 Cochrane Database Syst Rev 2016; 3: CD005563
30. Chow WB, Rosenthal RA, Merkow RP, Ko CY, Esnaola NF. 40. Hempenius L, Slaets JPJ, Boelens MAM, et al. Inclusion of
Optimal preoperative assessment of the geriatric surgical frail elderly patients in clinical trials: solutions to the
patient: a best practices guideline from the American problems. J Geriatr Oncol 2013; 4: 26e31
College of Surgeons national surgical quality improve- 41. Rehman HU. Under-representation of the elderly in clin-
ment program and the American geriatrics society. ical trials. Eur J Intern Med 2005; 16: 385e6
American College of Surgeons National Surgical Quality 42. Van Deudekom FJ, Postmus I, Van Der Ham DJ, et al.
Improvement Program; American Geriatrics Society J Am External validity of randomized controlled trials in older
Coll Surg 2012; 215: 453e66 adults, a systematic review. PLoS One 2017; 12: 1e8
31. Kozek-Langenecker S, Fenger-Eriksen C, Thienpont E, 43. McIsaac DI, Jen T, Mookerji N, Patel A, Lalu MM. In-
Barauskas G. European guidelines on perioperative terventions to improve the outcomes of frail people
venous thromboembolism prophylaxis. Eur J Anaesthesiol having surgery: a systematic review. PLoS One 2017; 12,
2018; 35: 116e22 e0190071
32. Aucoin SD, Hao M, Sohi R, et al. Accuracy and feasibility of 44. Jolliffe L, Lannin NA, Cadilhac DA, Hoffmann T. System-
clinically applied frailty instruments before surgery: a atic review of clinical practice guidelines to identify rec-
systematic review and meta-analysis. Anesthesiology 2020; ommendations for rehabilitation after stroke and other
133: 78e95 acquired brain injuries. BMJ Open 2018; 8, e018791
33. Alkadri J, Hage D, Nickerson LH, et al. A systematic review 45. Ciapponi A, Tapia-Lo  pez E, Virgilio S, Bardach A. The
and meta-analysis of preoperative frailty instruments quality of clinical practice guidelines for preoperative care
derived from electronic health data. Anesth Analg 2021; using the AGREE II instrument: a systematic review. Syst
133: 1094e106 Rev 2020; 9: 1e16
34. Hall DE, Arya S, Schmid KK, et al. Association of a frailty 46. Hladkowicz E, Dorrance K, Bryson GL, et al. Identifying
screening initiative with postoperative survival at 30, 180, barriers and facilitators to routine preoperative frailty
and 365 days. JAMA Surg 2017; 152: 233e40 assessment: a qualitative interview study. Can J Anaesth
35. McIsaac DI, Harris EP, Hladkowicz E, et al. Prospective 2022; 69: 1375e89
comparison of preoperative predictive performance be- 47. Abdellatif S, Hladkowicz E, Lalu MM, Boet S, Gagne S,
tween 3 leading frailty instruments. Anesth Analg 2020; McIsaac DI. Patient prioritization of routine and patient-
131: 263e72 reported postoperative outcome measures: a prospec-
36. Ellis G, Whitehead MA, Robinson D, O’Neill D, tive, nested cross-sectional study. Can J Anesth 2022; 69:
Langhorne P. Comprehensive geriatric assessment for 693e703
older adults admitted to hospital: meta-analysis of rand- 48. Brouwers MC, Spithoff K, Kerkvliet K, et al. Development
omised controlled trials. BMJ 2011; 343: d6553 and validation of a tool to assess the quality of clinical
37. Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. practice guideline recommendations. JAMA Netw Open
Untangling the concepts of disability, frailty, and 2020; 3: 1e13

Handling editor: Jonathan Hardman