Original Article

No Increased Risk of Cerebrovascular Accident With

Beach-Chair Versus Lateral Positioning for Shoulder
Arthroscopy
Conor N. O’Neill, M.D., Kimberly McFarland, B.S., Austin Bowyer, M.D.,
James Satalich, M.D., Bryant Tran, M.D., and Alexander R. Vap, M.D.

Purpose: To assess the incidence of adverse cerebrovascular events following shoulder arthroscopy in the beach-chair
position when compared with the lateral position. Methods: Records of 5 shoulder surgeons were searched using
Current Procedural Technology codes to identify patients who underwent arthroscopic shoulder surgery in both the
beach-chair and lateral positions between 2015 and 2020. Using both Current Procedural Technology codes for cere-
brovascular accident (CVA) imaging as well as the International Classification of Diseases, Tenth Revision, codes for CVA and
late neurologic sequela, patient charts were analyzed in the 30-day postoperative period. The anesthesiology record also
was queried for data regarding the blood pressure management intraoperatively, recording mean arterial pressures
(MAPs), and vasopressor administration. Patient demographics, comorbidities, and complications were compared between
the 2 cohorts using the Student 2-tailed t-test for continuous variables and c2 analysis for categorical variables. Signifi-
cance was set at P < .05. Results: There were 711 patients included in the analysis, with 471 in the beach-chair cohort
and 240 in the lateral cohort. Baseline demographics were similar between groups, except for age and American Society of
Anesthesiologists physical status classification, with the lateral group being significantly younger (P < .001) and lower
American Society of Anesthesiologists physical status classification (P ¼ .001) than the beach-chair group. Mean body
mass index, history of CVA, transient ischemic attack, hypertension, and peripheral vascular disease were not significantly
different. There were no documented CVAs in either cohort. There was no significant difference in the number of
postoperative radiologic scans to evaluate for CVA (P ¼ .77) or neurologic sequelae (P ¼ .48) between groups. The beach-
chair cohort had fewer instances of MAP <65 mm Hg, greater mean minimum MAP, but a greater percentage of patients
who received blood pressure support. Conclusions: There were no significant differences identified in the incidence of
CVA between patients undergoing arthroscopic shoulder surgery in the beach-chair and lateral positions. Level of
Evidence: Level III, retrospective cohort study.

T he many benefits of beach-chair positioning for

both open shoulder surgery and shoulder arthros-
copy have led to its more frequent use as compared
include anatomic positioning, improved visibility, ease
of setup and intraoperative repositioning, and simple
conversion to an open procedure (in the case of
with lateral decubitus positioning. These benefits arthroscopic surgery). However, this positioning is not
without risks, and although rare, the associated com-
plications can be catastrophic. With patients sitting in
From the Department of Orthopaedic Surgery, Virginia Commonwealth an upright position, the sequelae are typically second-
University, Richmond, Virginia, U.S.A. (C.N.O., J.S., A.R.V.); Department of ary to hypotensive events leading to cerebrovascular
Anesthesiology, Virginia Commonwealth University, Richmond, Virginia, hypoperfusion. Anecdotes and case reports of patients
U.S.A. (B.T.); and School of Medicine, Virginia Commonwealth University, experiencing neurovascular events, such as cortical in-
Richmond, Virginia, U.S.A. (K.M., A.B.).
Received February 2, 2023; accepted October 26, 2023.
farcts, hemispheric watershed injuries, and spinal cord
Address correspondence to Conor N. O’Neill, M.D., Department of Ortho- and medullary infarcts following surgery in this position
paedic Surgery, VCU Medical Center, West Hospital, 1200 E Broad St., 9th have been noted.1
Floor, Richmond, Virginia 23298, U.S.A. E-mail: [email protected] Moerman et al.2 demonstrated in a prospective study
Ó 2023 THE AUTHORS. Published by Elsevier Inc. on behalf of the using near infrared-spectroscopy that cerebral desatu-
Arthroscopy Association of North America. This is an open access article under
the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
ration events had an 80% incidence rate in the beach-
2666-061X/2358 chair position in their study population. However,
https://doi.org/10.1016/j.asmr.2023.100826 observed cerebrovascular desaturation events have not

Arthroscopy, Sports Medicine, and Rehabilitation, Vol 5, No 6 (December), 2023: 100826 1

2 C. N. O’NEILL ET AL.

been shown to have regular, anticipated clinical corre- lateral and the beach-chair positions. This was verified
lations. Yadeau et al.3 demonstrated in their retro- through chart review of operative reports. This study
spective and prospective analysis that despite frequent received institutional review board approval (IRB
episodes of hypotension, no strokes were observed in HM20020293 Short-term Incidence of Stroke following
an ambulatory setting with 4,169 patients. In addition, Beach Chair Positioning for Orthopaedic Surgery).
studies have illustrated that these desaturation event These charts were then queried for the CPT codes for
incidence rates were greater with general anesthesia, CVA work-up imaging, including computed tomogra-
yet patients had no neurologic events and had no dif- phy scan of the head, magnetic resonance imaging/
ference compared with the control group in cognitive angiography of the brain, and magnetic resonance
testing at follow-up.4,5 angiography of the neck (Appendix Table 1, available at
In a safety review article from 2019, Murphy et al.6 www.arthroscopyjournal.org) as well as the Interna-
indicated that multiple studies have suggested an tional Classification of Diseases, Tenth Revision (ICD-10)
imbalance in supply and demand for cerebral oxygen- diagnosis codes relevant to CVA (Appendix Table 2,
ation, but the actual association between this imbalance available at www.arthroscopyjournal.org) in the 30-
and negative outcomes are not clearly understood. day postoperative period. We also queried the ICD-10
Furthermore, they stated additional studies, “are codes for late neurologic sequelae in the 90-day post-
needed to define the incidence of adverse neurological operative period (Appendix Table 3, available at www.
adverse events in the beach chair position....” arthroscopyjournal.org).
The objective of this present study is to assess the The anesthesia record also was queried for the intra-
incidence of adverse cerebrovascular events following operative blood pressure readings as well as the peri-
shoulder arthroscopy in the beach-chair position when operative medication administration, including
compared with the lateral position. We hypothesized preoperative regional anesthesia and any vasopressor
that there would be no difference in rates of cerebro- medications, and their frequency, used during each
vascular accident (CVA) between the 2 groups. case.
Statistical analysis was performed using R-studio
Methods software, version 1.0.143 (R Foundation for Statistical
All patients who underwent shoulder arthroscopy at a Computing, Vienna, Austria) with the assistance of our
single facility by 5 fellowship-trained sports medicine institutional Biostatistics Consulting Laboratory. Patient
shoulder surgeons between 2015 and 2020 were demographics, comorbidities, and complications were
identified using Current Procedural Technology (CPT) compared between the 2 cohorts using the Student 2-
codes. Not all surgeons were active at our institution for tailed t test for continuous variables and chi-square
the full study period. Per known surgeon preference, analysis for categorical variables. Significance was set
we were able to identify procedures performed in the at P < .05.

Table 1. CPT Codes Included in Each Cohort

CPT Codes Code Description

Lateral
23455 Capsulorrhaphy, anterior; with labral repair (eg, Bankart procedure)
23465 Capsulorrhaphy, glenohumeral joint, posterior, with or without bone block
23466 Capsulorrhaphy, glenohumeral joint, any type multidirectional instability
29806 Arthroscopy, shoulder, surgical; capsulorrhaphy
Beach chair
29805 Arthroscopy, shoulder, diagnostic, with or without synovial biopsy
29807 Arthroscopy, shoulder, surgical; repair of SLAP lesion
29819 Arthroscopy, shoulder, surgical; with removal of loose body or foreign body
29820 Arthroscopy, shoulder, surgical; synovectomy, partial
29821 Arthroscopy, shoulder, surgical; synovectomy, complete
29822 Arthroscopy, shoulder, surgical; debridement, limited
29823 Arthroscopy, shoulder, surgical; debridement, extensive
29824 Arthroscopy, shoulder, surgical; distal claviculectomy including distal articular surface (Mumford procedure)
29825 Arthroscopy, shoulder, surgical; with lysis and resection of adhesions, with or without manipulation
29826 Arthroscopy, shoulder, surgical; decompression of subacromial space with partial acromioplasty, with coracoacromial
ligament (i.e., arch) release, when performed
29827 Arthroscopy, shoulder, surgical; with rotator cuff repair
29828 Arthroscopy, shoulder, surgical; biceps tenodesis
CPT, Current Procedural Technology.
 CVA RISK AND PATIENT POSITIONING DURING SA 3

Table 2. Patient Demographic Data

Lateral (N ¼ 240) Beach Chair (N ¼ 471) P Value

Age, y, mean  SD 34.6  12.2 57.2  13.8 <.001*
BMI 29.5  7.2 30.5  7.1 .07
ASA class, n (%) 2.1  0.7 2.3  0.7 .001*
1 47 (19.6) 50 (10.6) e
2 122 (50.8) 243 (51.6) e
3 68 (28.3) 171 (36.3) e
4 3 (1.3) 7 (1.5) e
Comorbidities
Hx CVA 0 0 e
Hx TIA 0 0 e
Hx DM 23 (9.6) 86 (18.3) .002*
Hx HTN 5 (2.1) 12 (2.5) .3
Hx PVD 0 3 .2
Operative time. min 93.4  43.3 95.2  40.6 .6
Preoperative blocky 228 (95) 452 (96) .6
Ambulatory OR 159 (66.3) 323 (68.6) .5
NOTE. Continuous variables recorded as mean  SD, and categorical variables recorded as number (percentage).
ASA, American Society of Anesthesiologists physical status classification; BMI, body mass index; CVA, cerebrovascular accident; DM, diabetes
mellitus; HTN, hypertension; Hx, history; OR, operating room; PVD, peripheral vascular disease; SD, standard deviation; TIA, transient ischemic
attack.
*Statistical significance.
y
Preoperative regional anesthetic block administered.

Results undergoing scans in the postoperative period. There

A total of 711 patients were included in the analysis, was 1 patient in the beach-chair cohort with docu-
with 471 (66.2%) in the beach chair-cohort and 240 mentation of neurologic sequelae. This patient was
(33.8%) in the lateral cohort. All procedures included presenting with recurrent headache, who had history of
in the study can be found in Table 1. There were some CVA at age 4 years with longstanding residual left-sided
baseline differences in demographics between the weakness, that was first recorded in the chart (using
groups, namely age and American Society of Anesthe- ICD-10) during the postoperative period, but did not
siologists class, with the lateral group being significantly represent a new neurologic sequela. However, this did
younger (P < .001) and lower American Society of not represent a significant difference between the
Anesthesiologists class (P ¼ .001) than the beach-chair groups (P ¼ .4) (Table 3).
group. Mean body mass index, history of CVA, tran- There was a lower percentage of patients who expe-
sient ischemic attack, hypertension, and peripheral rienced hypotension (mean arterial pressure [MAP]
vascular disease were not significantly different <65 mm Hg) in the beach-chair cohort, although this
(Table 2). was not found to be significant (P ¼ .1). The patients in
There were no documented CVAs in either cohort. the lateral cohort also had significantly more frequent
There was no significant difference in the number of episodes of hypotension with an average of 1.7  2.8
postoperative radiologic scans to evaluate for CVA be- compared with 1.2  2.1 in the beach-chair group (P ¼
tween groups (P ¼ .8), with only 2 patients in the .01). There was no significant difference in the average
lateral group and 3 patients in the beach-chair group minimum MAP between groups (P ¼ .4), but there was

Table 3. Cohort Data of CVA and Blood Pressure

Lateral (N ¼ 240) Beach Chair (N ¼ 471) P Value

CVA 0 0 e
Postoperative imaging 2 3 .8
Neuro sequelae 0 1 .5
Patients with MAP <65 mm Hg 105 (43.8%) 188 (40%) .1
MAP recordings <65 mm Hg per case 1.7  2.8 (Var ¼ 7.9) 1.2  2.1 (Var ¼ 4.5) .01*
Minimum MAP, mean  SD 66.2  10.3 67.0  10.3 .4
Received vasopressor 106 (44.2) 294 (62.4) <.001*
CVA, cerebrovascular accident; MAP, mean arterial pressure; SD, standard deviation; Var, variance.
*Statistical significance.
4 C. N. O’NEILL ET AL.

a significantly greater percentage of patients who population undergoing arthroscopic shoulder surgery is
received blood pressure support in the beach-chair generally quite healthy, we wanted to be able to
group (P > .001). generalize our results to a wider population.
There were no significant differences in the anesthetic Surgeries performed in the beach-chair position allow
technique between groups, with 96% of patients in the for numerous benefits, such as better positioning, visi-
beach-chair and 95% of patients in the laterally posi- bility, ease of set up or repositioning, and converting to
tioned group receiving a preoperative interscalene open surgery. Numerous studies have shown decreased
block. There was also no significant differences in the cerebral perfusion in this position and concerns remain
operative time between groups, with an average time of regarding long-term neurologic sequelae.
93.4  43.3 minutes (lateral) and 95.2  40.6 minutes
(beach-chair) (P ¼ .6). Limitations
Limitations of the study include the reliance on both
Discussion CPT and ICD-10 coding for accurate data mining.
The most important finding of this study was that no However, there is well documented literature that
strokes were observed in 711 patients undergoing supports the use of these codes and their level of ac-
shoulder surgery in the beach-chair position, despite curacy for sufficiently robust research.17,18 It is also
frequent occurrences of hypotension and administra- notable that the baseline demographic data differs be-
tion of vasoactive medications. By Hanley’s “rule of tween our 2 groups, which may have an effect on the
three,” which gives the upper limit of the 95% confi- differences see; however, this is not surprising, as
dence interval for the probability of adverse event that generally younger, healthier people have capsulor-
has not yet occurred, we can calculate that the rhaphy procedures compared with other arthroscopic
maximum risk ¼ 3/n (when n >30), giving us 0.64% in shoulder surgery such as rotator cuff repair. Also, the
beach-chair cohort.7 Eypasch et al.8 remind us that absence of stroke in this patient cohort could be due to
surgeons should keep this rule in mind when compli- sample size, where a larger study would likely docu-
cation rates of zero are reported and “when they have ment a non-zero stroke rate. The results here may not
not (yet) experienced a disastrous complication from a be generalizable to a lower-volume community setting,
procedure.” particularly if anesthesiologists are inexperienced in
Although there was no significant difference in min- treating patients in the beach-chair position
imum MAP measurement, there were more patients
with MAPs less than 65 mm Hg in the beach-chair Conclusions
group, and a significantly greater percentage of pa- There were no significant differences identified in the
tient required vasopressors; however, this did not incidence of CVA between patients undergoing
manifest clinically as CVA. Intraoperative hypotension arthroscopic shoulder surgery in the beach-chair and
is a relatively common finding, with Murphy et al. lateral positions.
citing the incidence as 47% to 51% in the 8,396 pa-
tients they reviewed from 2 large-scale studies they Disclosure
reviewed. Some authors have posited that cerebral The authors (C.N.O., K.M., A.B., J.S., B.T., A.R.V.)
blood flow is more important that blood pressure as a declare that they have no known competing financial
marker for cranial perfusion.3 Decreased cerebral oxy- interests or personal relationships that could have
gen perfusion has been measured via near-infrared appeared to influence the work reported in this paper.
spectroscopy,9 continuous-wave near-infrared spec- Full ICMJE author disclosure forms are available for
troscopy,2,5,10-12 cerebral tissue oxygen saturation,4 this article online, as supplementary material.
electroencephalography,13 and middle cerebral artery
flow via Doppler.12 Isolated case reports in which pa- References
tients suffered cerebral infarcts with residual neurologic 1. Pohl A, Cullen DJ. Cerebral ischemia during shoulder
deficits are rare but are an important consideration.1,14 surgery in the upright position: A case series. J Clin Anesth
Although cerebral oxygen desaturation occurs at rates 2005;17:463-469.
of up to 76% to 80%,2,9 numerous studies found that 2. Moerman AT, De Hert SG, Jacobs TF, De Wilde LF,
decreased cerebral oxygen perfusion in the beach-chair Wouters PF. Cerebral oxygen desaturation during beach
chair position. Eur J Anaesthesiol 2012;29:82-87.
position did not significantly correlate with any post-
3. Yadeau JT, Casciano M, Liu SS, et al. Stroke, regional
operative clinical findings.3-6,9,11-13,15,16
anesthesia in the sitting position, and hypotension: a re-
Some previous studies have focused solely on view of 4169 ambulatory surgery patients. Reg Anesth Pain
ambulatory surgical patients, but we hoped to include Med 2011;36:430-435.
patients who had a full spectrum of medical comor- 4. Koh JL, Levin SD, Chehab EL, Murphy GS. Neer Award
bidities, including cases both from our ambulatory and 2012: cerebral oxygenation in the beach chair position: A
main hospital operating rooms. Although the patient prospective study on the effect of general anesthesia
 CVA RISK AND PATIENT POSITIONING DURING SA 5

compared with regional anesthesia and sedation. 12. Aguirre JA, Etzensperger F, Brada M, et al. The beach
J Shoulder Elbow Surg 2013;22:1325-1331. chair position for shoulder surgery in intravenous general
5. Laflam A, Joshi B, Brady K, et al. Shoulder surgery in the anesthesia and controlled hypotension: Impact on
beach chair position is associated with diminished cerebral cerebral oxygenation, cerebral blood flow and
autoregulation but no differences in postoperative cogni- neurobehavioral outcome. J Clin Anesth 2019;53:
tion or brain injury biomarker levels compared with su- 40-48.
pine positioning: the anesthesia patient safety foundation 13. Gillespie R, Shishani Y, Streit J, et al. The safety of
beach chair study. Anesth Analg 2015;120:176-185. controlled hypotension for shoulder arthroscopy in the
6. Murphy GS, Greenberg SB, Szokol JW. Safety of beach beach-chair position. J Bone Joint Surg Am 2012;94:
chair position shoulder surgery: A review of the current 1284-1290.
literature. Anesth Analg 2019;129:101-118. 14. Zeidan A, Bluwi M, Elshamaa K. Postoperative brain
7. Hanley JA, Lippman-Hand A. If nothing goes wrong, is stroke after shoulder arthroscopy in the lateral decubitus
everything all right? Interpreting zero numerators. JAMA position. J Stroke Cerebrovasc Dis 2014;23:384-386.
1983;249:1743-1745. 15. Pin-on P, Schroeder D, Munis J. The hemodynamic
8. Eypasch E, Lefering R, Kum CK, Troidl H. Probability of management of 5177 neurosurgical and orthopedic pa-
adverse events that have not yet occurred: A statistical tients who underwent surgery in the sitting or "beach
reminder. BMJ 1995;311:619-620. chair" position without incidence of adverse neurologic
9. Chan JH, Perez H, Lee H, Saltzman M, Marra G. Evalua- events. Anesth Analg 2013;116:1317-1324.
tion of cerebral oxygen perfusion during shoulder 16. Friedman DJ, Parnes NZ, Zimmer Z, Higgins LD,
arthroplasty performed in the semi-beach chair position. Warner JJ. Prevalence of cerebrovascular events during
J Shoulder Elbow Surg 2020;29:79-85. shoulder surgery and association with patient position.
10. Dippmann C, Winge S, Nielsen HB. Severe cerebral Orthopedics 2009;32.
desaturation during shoulder arthroscopy in the beach- 17. Burns EM, Rigby E, Mamidanna R, et al. Systematic re-
chair position. Arthroscopy 2010;26:S148-150. view of discharge coding accuracy. J Public Health Oxf
11. Salazar D, Sears BW, Aghdasi B, et al. Cerebral desatu- 2012;34:138-148.
ration events during shoulder arthroscopy in the beach 18. Singh JA, Ayub S. Accuracy of VA databases for diagnoses
chair position: Patient risk factors and neurocognitive ef- of knee replacement and hip replacement. Osteoarthritis
fects. J Shoulder Elbow Surg 2013;22:1228-1235. Cartilage 2010;18:1639-1642.