Higher stroke risk after carotid endarterectomy and transcarotid artery revascularization is associated with relative surgeon volume ratio

Abstract

Objective:

Adoption of transcarotid artery revascularization (TCAR) by surgeons has been variable, with some still performing traditional carotid endarterectomy (CEA), whereas others have shifted to mostly TCAR. Our goal was to evaluate the association of relative surgeon volume of CEA to TCAR with perioperative outcomes.

Methods:

The Vascular Quality Initiative CEA and carotid artery stent registries were analyzed from 2021 to 2023 for symptomatic and asymptomatic interventions. Surgeons participating in both registries were categorized in the following CEA to CEA+TCAR volume percentage ratios: 0.25 (majority TCAR), 0.26 to 0.50 (more TCAR), 0.51 to 0.75 (more CEA), and 0.76 to 1.00 (majority CEA). Primary outcomes were rates of perioperative ipsilateral stroke, death, cranial nerve injury, and return to the operating room for bleeding.

Results:

There were 50,189 patients who underwent primary carotid revascularization (64.3% CEA and 35.7% TCAR). CEA patients were younger (71.1 vs 73.5 years, P < .001), with more symptomatic cases, less coronary artery disease, diabetes, and lower antiplatelet and statin use (all P < .001). TCAR patients had lower rates of smoking, obesity, and dialysis or renal transplant (all P < .001). Postoperative stroke after CEA was significantly impacted by the operator CEA to TCAR volume ratio (P = .04), with surgeons who perform majority TCAR and more TCAR having higher postoperative ipsilateral stroke (majority TCAR odds ratio [OR]: 2.15, 95% confidence interval [CI]: 1.16–3.96, P = .01; more TCAR OR: 1.42, 95% CI: 1.02–1.96, P = .04), as compared with those who perform majority CEA. Similarly, postoperative stroke after TCAR was significantly impacted by the CEA to TCAR volume ratio (P = .02), with surgeons who perform majority CEA and more CEA having higher stroke (majority CEA OR: 1.51, 95% CI: 1.00–2.27, P = .05; more CEA OR: 1.50, 95% CI: 1.14–2.00, P = .004), as compared with those who perform majority TCAR. There was no association between surgeon ratio and perioperative death, cranial nerve injury, and return to the operating room for bleeding for either procedure.

Conclusions:

The relative surgeon CEA to TCAR ratio is significantly associated with perioperative stroke rate. Surgeons who perform a majority of one procedure have a higher stroke rate in the other. Surgeons offering both operations should maintain a balanced practice and have a low threshold to collaborate as needed.

Rapid adoption of transcarotid artery revascularization (TCAR) has led to a significant increase in the number of annual TCARs and a reduction in carotid endarterectomies (CEAs) performed in the United States.¹ Although CEA has been the traditional standard of care for extracranial internal carotid artery revascularization, TCAR offers an alternative method to carotid stenting for patients with challenging anatomy due to high carotid artery bifurcation, high carotid lesions near the skull base, and prior neck operations or carotid interventions.² Furthermore, TCAR has become widely used in lower-risk patients.³ Perioperative stroke and death rate after TCAR have been shown to be comparable to CEA in the Safety and Efficacy Study for Reverse Flow used During Carotid Artery Stenting Procedure trials (ROADSTER) in high-risk patients and subsequently shown to have similar outcomes in standard risk patients positing the TCAR approach as a valuable alternative for carotid revascularization.^4–6

Surgeon volume has been associated with postoperative outcomes for several index operations in general and vascular surgery.^7–12 In response, some major academic centers have imposed minimum volume standards for both hospitals and surgeons for certain procedures.¹³ This relationship has been rigorously documented for CEA outcomes where high operator volumes are associated with decreased risk of periprocedural death and stroke.^7,9,10 Similarly, studies on surgeon experience and carotid artery stenting have found that more experienced surgeons have shorter intervention times and lower adverse outcomes.^14–16

Given that the proportion of carotid revascularization procedures performed by a surgeon is changing, it is important to understand not only the impact of case volume on outcomes but also the impact of relative surgeon volume ratio of CEA to TCAR. We hypothesize that among surgeons who perform both procedures, those who perform the majority of one type of procedure, such as over 75% of one operation, will have a higher stroke risk in the other and that the magnitude of effect will be largest for CEA for those who perform majority TCAR. Therefore, the aim of this study is to determine whether the relative surgeon volume of CEA to TCAR ratio is associated with postoperative outcomes for each respective procedure. We used the Vascular Quality Initiative (VQI) to obtain a large and generalizable study size for both the carotid artery stent and CEA registries with adequate preoperative, intraoperative, and postoperative collected data.

METHODS

Data source.

This retrospective cohort study analyzed the VQI CEA and carotid artery stenting (CAS) registries between the years 2021 and 2023. The VQI is a national collaborative comprising community, academic, and teaching hospitals primarily from North America whose goal is to improve patient outcomes from vascular diseases.¹⁷ To date, the registry has 1000 centers with roughly half of participating physicians from vascular surgery, followed by cardiology and radiology.¹⁸ The years 2021 to 2023 were selected to allow time for the wide adoption of TCAR since its introduction in 2015. A narrow time period was selected to best capture a surgeon’s CEA and TCAR ratio at one point in time.

The CEA and CAS registries were analyzed separately. Both symptomatic and asymptomatic interventions were evaluated. Only cases that were conducted by surgeons participating in both CEA and CAS registries were included. Participants aged 18 years or older who underwent primary carotid artery intervention for atherosclerotic disease were included. We excluded carotid revascularizations for indications other than atherosclerotic disease, those with occlusions, prior ipsilateral carotid interventions, and those who underwent concomitant intracranial interventions or other additional carotid interventions. The Boston University Chobanian and Avedisian School of Medicine institutional review board approved this study and informed consent was waived.

Variables.

Patient demographic, medical, and operative characteristics were evaluated for CEA and TCAR cohorts, respectively. Demographic variables included age, sex, race, ethnicity, current smoking, insurance status (Medicare, Medicaid, and commercial insurance), and obesity, defined as having a body mass index of ≥30 kg/m². Relevant medical and surgical variables included a history of hypertension, coronary artery disease (CAD), congestive heart failure, chronic obstructive pulmonary disease, diabetes mellitus (DM) requiring insulin, dialysis use or renal transplant, and a history of a coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI). Relevant preoperative medications included the use of aspirin, a P2Y12 inhibitor, statin, angiotensin converting enzyme inhibitor or angiotensin receptor blocker, and chronic anticoagulation. Procedural and lesion characteristics included indication for symptomatic or asymptomatic disease, the status of the case, whether elective or urgent, the degree of carotid artery stenosis, and the use of general anesthesia. Symptomatic disease was defined as having a transient ischemic attack or stroke within 6 months of the index operation.

Surgeons were selected if they had overlapping identification in both the CEA and CAS registries. Surgeon CEA and TCAR ratios were calculating using total counts for both procedures between 2021 and 2023. Surgeon CEA to CEA+TCAR ratios were calculated using overall surgeon experience in both procedures from 2021 to 2023. For simplification of categories, these groups were categorized into four even groups to maximize the overall analysis precision as follows: 0 to 0.25 (majority TCAR), 0.26 to 0.50 (more TCAR), 0.51 to 0.75 (more CEA), 0.76 to 1.00 (majority CEA). Surgeon volume is defined as the total number of CEAs and TCARs performed from 2021 to 2023 and was split into terciles to maximize precision without compromising group size by even allocation into the subgroups. Tercile groups were divided as 1 to 6 operations per year, 7 to 23 operations per year, and 24 or more operations per year.

Outcomes.

Primary outcomes evaluated were the odds of sustaining a postoperative ipsilateral stroke, death, cranial nerve injury (CNI), and return to the operating room for bleeding (RTOR) within 30 days of index operation. Outcomes were evaluated for CEA and TCAR separately by operator ratio category. CNI was not assessed for TCAR.

Statistical analysis.

Patient characteristics were described using counts and percentages for categorical variables and means and standard deviations for continuous variables. Unadjusted rates of perioperative outcomes were evaluated as dichotomous outcomes and presented as counts and percentages. Multivariable logistic regressions were conducted for each dichotomous outcome and adjusted for clinical confounders. These include age, sex, race, the presence of symptoms from carotid stenosis, dialysis use or renal transplant, CAD, DM, and preoperative antiplatelet (aspirin or P2Y12 inhibitor), statin use, and surgeon volume. CEA outcomes were referenced to surgeons who perform majority CEA, whereas TCAR outcomes were referenced to surgeons who perform majority TCAR. The associations were expressed as adjusted odds ratios (ORs) with corresponding 95% confidence intervals (CIs). A P value of <.05 was considered statistically significant. Data analyses were performed using SAS Studio 3.81 software (SAS Institute Inc).

RESULTS

CEA

Demographics, comorbidities, and procedural details.

There were 40,695 patients in the CEA registry treated between 2021 and 2023. Of these, 32,277 met the inclusion criteria. This cohort had an average age of 71.1 years (±8.9 years), 60.2% were male, and 87.3% were of White race (Table I). The most common comorbidities were hypertension (89.9%), prior CABG or PCI (30.7%), and CAD (25.9%). For preoperative medical therapy, 83.1% were on aspirin, 86.3% were on a statin, and 39.5% were on P2Y12 inhibitors. Of these, 41.7% had symptomatic carotid artery disease, 62.1% had greater than 80% stenosis, and 84.3% of the procedures were elective. The majority of patients who underwent a CEA were operated on by a surgeon who performed majority CEA (71.4%) (Table I). By surgeon volume tercile, there were 927, 955, and 962 surgeons in the highest, middle, and lowest terciles of total procedures, respectively.

Table I.

Demographics by carotid endarterectomy (CEA) vs transcarotid artery revascularization (TCAR) interventions

Characteristic	CEA (N = 32,277)	TCAR (N = 17,912)	P value
Patient demographics
Age, years, mean ± SD	71.1 ± 8.9	73.5 ± 8.7	<.001
Male sex, No. (%)	19,432 (60.2)	11,328 (63.2)	<.001
Race and ethnicity, No. (%)			<.001
White	28,160 (87.3)	15,344 (85.7)
Black	1654 (5.1)	1041 (5.8)
Hispanic	669 (2.1)	583 (3.3)
Other	2433 (7.5)	947 (5.5)
Obese, No. (%)	11,008 (34.2)	5891 (32.9)	.005
Smoking, No. (%)	8059 (25.0)	3862 (21.6)	<.001
Insurance, No. (%)			<.001
Medicare	22,104 (68.5)	13,326 (74.4)
Medicaid	1406 (4.4)	674 (3.8)
Commercial	7682 (23.8)	3359 (18.8)
Medical and surgical history, No. (%)
Hypertension	29,008 (89.9)	16,350 (91.3)	<.001
CAD	8359 (25.9)	8693 (48.6)	<.001
CHF	4019 (12.5)	2920 (16.3)	<.001
COPD	6890 (21.4)	4180 (23.3)	<.001
CABG/PCI	9915 (30.7)	6474 (36.2)	<.001
Diabetes on insulin	4220 (13.1)	2531 (14.1)	<.001
Dialysis/transplant	358 (1.1)	42 (0.2)	<.001
Preoperative aspirin	26,761 (83.1)	16,043 (89.6)	<.001
Preoperative statin	27,788 (86.3)	16,212 (90.6)	<.001
Preoperative P2Y12	12,736 (39.5)	15,974 (89.2)	<.001
Preoperative ACE inhibitor/ARB	17,214 (53.4)	9726 (54.3)	<.05
Preoperative chronic AC	4425 (13.7)	2771 (15.5)	<.001
TIA/stroke 6 months	13,260 (41.7)	6776 (39.0)	<.001
Procedural characteristics, No. (%)
Operator ratios			<.001
Majority TCAR	375 (1.2)	9792 (54.7)
More TCAR	2127 (6.6)	3158 (17.6)
More CEA	6731 (20.9)	3651 (20.4)
Majority CEA	23,044 (71.4)	1311 (7.3)
Elective case	27,148 (84.3)	15,388 (85.9)	<.001
Stenosis 80+	19,590 (62.1)	12,661 (77.7)	<.001
General anesthesia	1277 (4.0)	15,986 (89.3)	<.001

AC, Anticoagulation; ACE, angiotensin converting enzyme; ARB, angiotensin receptor blocker; CABG/PCI, previous coronary artery bypass graft or percutaneous coronary intervention; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; Dialysis/transplant, dialysis use or history of renal transplant; SD, standard deviation; TIA, transient ischemic attack.

Majority TCAR (0.25 CEA), more TCAR (0.26–0.50 CEA), more CEA (0.51–0.75 CEA), and majority CEA (0.76–1.00 CEA).

Stroke outcomes

Postoperative stroke after CEA was observed in 12 patients (3.2%) in the majority TCAR, 42 (2.0%) in the more TCAR, 99 (1.5%) in the more CEA, and 347 (1.5%) in the majority CEA group (P = .045) (Table II). Postoperative death, CNI injury, and RTOR observations were similar and low across the operator ratio group. In the multivariable analysis, there was a statistically significant association between surgeon CEA to TCAR ratio and stroke (P = .04) (Table III). Surgeons who performed majority and more TCAR had higher postoperative stroke than those who performed majority CEA (majority TCAR OR: 2.15, 95% CI: 1.16–3.96, P = .01; more TCAR: 1.41 95% CI: 1.02–1.96, P = .04). Overall operator volume for CEA was independent of the surgeon volume ratio and inversely associated with stroke (OR: 2.05, 95% CI: 1.42–2.95, P < .001 for 1–6 cases; OR: 1.44, 95% CI: 1.17–1.77, P =.001 for 7–23 cases), as compared with operator volume 24+ cases.

Table II.

Carotid endarterectomy (CEA) outcomes by the relative surgeon CEA to transcarotid artery revascularization (TCAR) ratio count and percent

Outcome	Majority TCAR (N = 375), No. (%)	More TCAR (N = 2127), No. (%)	More CEA (N = 6731), No. (%)	Majority CEA (N = 23,044), No. (%)	P value
Stroke	12 (3.2)	42 (2.0)	99 (1.5)	347 (1.5)	.045
Death	3 (0.8)	14 (0.7)	46 (0.7)	152 (0.7)	.99
CNI	9 (2.4)	39 (1.8)	162 (2.4)	513 (2.2)	.46
RTOR	5 (1.3)	12 (0.6)	65 (1.0)	212 (0.9)	.24

CNI, Cranial nerve injury; RTOR, return to the odds ratio for bleeding.

Bolded numbers indicate statistical significance (P value <.05).

Majority TCAR (0.25 CEA), more TCAR (0.26–0.50 CEA), more CEA (0.51–0.75 CEA), and majority CEA (0.76–1.00 CEA).

Table III.

Adjusted multivariable analysis for stroke after carotid endarterectomy (CEA) by the relative surgeon CEA to transcarotid artery revascularization (TCAR) ratio

Variable	OR	CI	P value
Surgeon ratio			.04
Majority TCAR	2.15	1.16–3.96	.01
More TCAR	1.41	1.02–1.96	.04
More CEA	1.03	0.82–1.29	.81
Majority CEAa	Ref.	Ref.	Ref.
Volume 1–6 vs 24+	2.05	1.42–2.95	<.001
Volume 7–23 vs 24+	1.44	1.17–1.77	.001

CI, 95% Confidence interval; CNI, cranial nerve injury; OR, odds ratio. Bolded numbers indicate statistical significance (P value <.05). Majority TCAR (0.25 CEA), more TCAR (0.26–0.50 CEA), more CEA (0.51–0.75 CEA), and majority CEA (0.76–1.00 CEA).

Volume 1–6 TCAR cases per year represents lowest volume tercile. Volume 7–23 TCAR cases per year represents medium tercile. Volume 24+ TCAR cases per year represents highest volume tercile.

^aReference group for CEA is majority CEA. Italic font represents the overall model.

TCAR

Demographics, comorbidities, and procedural details.

There were 24,051 patients in the CAS group who were treated using TCAR. Of these, 17,912 met our inclusion criteria. The TCAR cohort had an average age of 73.5 years (±8.7 years), 63.2% were male, and 85.5% were of White race (Table I). The most common comorbidities were hypertension (91.3%), CAD (48.6%), CABG/PCI (36.2%), chronic obstructive pulmonary disease (23.3%), and DM (14.1%). For preoperative medical therapy, 89.6% were on aspirin, 90.6% were on a statin, and 89.2% were on a P2Y12 inhibitor. Of these, 39% had symptomatic carotid artery disease, 77.7% had greater than 80% stenosis, and 85.9% of the procedures were elective. Most patients who underwent a TCAR were operated on by a surgeon who performed majority TCAR (54.7%), followed by surgeons who performed more CEAs (20.4%), and more TCAR (17.6%) (Table I).

Stroke outcomes.

Postoperative stroke after TCAR was observed in 164 (1.7%) patients in the majority TCAR, 50 (1.6%) in the more TCAR, 82 (2.2%) in the more CEA, and 31 (2.4%) in the majority CEA group (P = .054) (Table IV). In the multivariable analysis, there was a statistically significant association between surgeon CEA to TCAR ratio and stroke (P = .02) (Table V). Surgeons who performed majority and more CEA had higher postoperative stroke than those who performed majority TCAR (majority CEA OR: 1.51, 95% CI: 1.00–2.27, P = .05; more CEA OR: 1.50, 95% CI: 1.14–2.00, P = .004). Again, surgeon volume had an inverse relationship with stroke where operators who performed less than 24+ cases a year had higher postoperative stroke rates (OR: 1.70, 95% CI: 1.05–2.74, P = .03 for 1–6 cases per year; OR: 1.43, 95% CI: 1.10–1.84, P = .006 for 7–23 cases per year).

Table IV.

Transcarotid artery revascularization (TCAR) outcomes by the relative surgeon carotid endarterectomy (CEA) to TCAR ratio count and percentage

Outcome	Majority TCAR (N = 9792), No. (%)	More TCAR (N = 3158), No. (%)	More CEA (N = 3651), No. (%)	Majority CEA (N = 1311), No. (%)	P value
Stroke	164 (1.7)	50 (1.6)	82 (2.2)	31 (2.4)	.054
Death	82 (0.8)	20 (0.6)	34 (0.9)	9 (0.7%)	.51
RTOR	38 (0.4)	8 (0.3)	20 (0.5)	8 (0.6)	.17

CNI, Cranial nerve injury; RTOR, return to the odds ratio for bleeding.

Majority TCAR (0.25 CEA), more TCAR (0.26–0.50 CEA), more CEA (0.51–0.75 CEA), and majority CEA (0.76–1.00 CEA).

Table V.

Adjusted multivariable analysis for stroke after transcarotid artery revascularization (TCAR) by the relative surgeon carotid endarterectomy (CEA) to TCAR ratio

	OR	CI	P value
Surgeon ratio			.02
Majority TCARa	Ref.	Ref.	Ref.
More TCAR	1.02	0.74–1.42	.89
More CEA	1.50	1.14–2.00	.004
Majority CEA	1.51	1.00–2.27	.05
Volume 1–6 vs 24+	1.70	1.05–2.74	.03
Volume 7–23 vs 24+	1.43	1.11–1.84	.006

CI, 95% Confidence interval; CNI, cranial nerve injury; OR, odds ratio; RTOR, return to the operating room for bleeding.

Bolded numbers indicate statistical significance (P value <.05).

Sensitivity analysis adjusted for surgeon volume.

Majority TCAR (0.25 CEA), more TCAR (0.26–0.50 CEA), more CEA (0.51–0.75 CEA), and majority CEA (0.76–1.00 CEA).

Volume 1–6 TCAR cases per year represents lowest volume tercile. Volume 7–23 TCAR cases per year represents medium tercile. Volume 24+ TCAR cases per year represents highest volume tercile.

^aReference group for TCAR is majority TCAR. Italic font represents the overall model.

CEA and TCAR

Other outcomes.

There was no significant association between surgeon ratio and postoperative death, RTOR, and CNI for CEA (Table VI). There was no significant association between surgeon ratio and postoperative death and RTOR for TCAR (Table VI).

Table VI.

Adjusted multivariable analysis for other outcomes for carotid endarterectomy (CEA) and transcarotid artery revascularization (TCAR) by the relative surgeon CEA to TCAR ratio

Outcomes by surgeon ratio	CEA		TCAR
	OR (95% CI)	P value	OR (95% CI)	P value
Death		.96		.61
Majority TCARa	1.31 (0.41–4.13)	.65	Ref.	–
More TCAR	1.09 (0.63–1.90)	.76	0.76 (0.46–1.25)	.28
More CEA	1.03 (0.73–1.44)	.87	1.09 (0.72–1.64)	.7
Majority CEAb	Ref.	–	0.87 (0.43–1.74)	.69
RTOR		.24		.31
Majority TCARa	1.47 (0.60–3.61)	.4	Ref.	–
More TCAR	0.61 (0.34–1.10)	.1	0.67 (0.31–1.44)	.31
More CEA	1.06 (0.80–1.41)	.68	1.35 (0.77–2.35)	.3
Majority CEAb	Ref.	–	1.41 (0.63–3.17)	.41
CNI		.47		N/A
Majority TCARa	1.13 (0.58–2.20)	.72	–	–
More TCAR	0.81 (0.58–1.13)	.22	–	–
More CEA	1.07 (0.89–1.28)	.48	–	–
Majority CEAb	Ref.	–	–	–

CI, 95% Confidence interval; CNI, cranial nerve injury; N/A, not applicable; OR, odds ratio, RTOR, return to the operating room for bleeding.

Majority TCAR (0.25 CEA), more TCAR (0.26–0.50 CEA), more CEA (0.51–0.75 CEA), and majority CEA (0.76–1.00 CEA).

^aReference group for TCAR is majority TCAR.

^bReference group for CEA is majority CEA.

DISCUSSION

The relative surgeon volume CEA to TCAR ratio is associated with postoperative stroke for both procedures. Surgeons who perform more and majority CEA or TCAR have higher postoperative stroke in the operation performed the least. The largest effect was in patients who underwent CEA by surgeons who perform majority TCAR, although overall rates of stroke are still acceptable among all groups. Surgeons have the lowest rate of stroke for the operation in which they most often perform.

These findings corroborate existing studies that observed an inverse association between operator volume and operative outcomes, particularly for higher stroke rate after CEA and TCAR.^7–12 Although surgeon ratio and surgeon volume are different measures, the more operations performed for each type the lower the stroke rate by both metrics. This phenomenon is likely driven by the strong relationship between postoperative stroke after carotid revascularization and surgeon-specific factors, such as patient selection, surgical skill and experience, and intraoperative decision-making. This study does not find an association between surgeon ratio and mortality, unlike existing surgeon volume studies.^8,9,14–16 It is possible that surgeon ratio may not be associated with mortality in the same way that surgeon volume is and that other factors, such as center volume, may contribute to mortality.

Significant findings between operator ratio and outcomes after adjusting for surgeon volume suggest that these two variables may act via different mechanisms to impact stroke. This implies that patients who undergo a CEA from a surgeon who performs a high volume of carotid revascularization procedures, but performs a low number of CEAs relative to TCAR, may have a higher stroke rate than a medium volume surgeon who performs majority CEA. This nuance in surgeon practice has important implications when considering risk factors for postoperative stroke after each respective procedure.

Our study observed that the CEA stroke rate ranged from 1.5% to 3.2% and that of TCAR ranged from 1.7% to 2.4%, suggesting that overall stroke among all the categories is low for both operations and consistent with the Society for Vascular Surgery recommendations aiming for a perioperative stroke rate less than 3%.¹⁹ However, as the proportions change, there is a significant uptrend in stroke and this should be taken into consideration. Our data suggest that stroke risk is low and comparable for both operations when performed by surgeons who specialize in one type of procedure and remains low for surgeons with a balanced practice. Therefore, surgeons should maintain a balanced surgical experience and skill set with consideration for collaboration in challenging cases. These observations may be particularly more salient for novice surgeons. In our analysis, 71% of patients received a CEA by a surgeon who performs majority CEA and 54% received a TCAR by a surgeon who performs majority TCAR, indicating that most surgeons in the VQI tend to specialize.

There are limitations to this study. As a retrospective cohort study, bias may exist due to the inability to adjust for confounding variables, such as surgeon experience. It is possible that more novice surgeons have higher stroke risk for CEA, irrespective of their CEA to TCAR ratio. Conversely, more experienced surgeons who performed majority TCAR during 2021 to 2023 may have a high degree of experience with CEAs and/or open cases that confer translatable skills for CEA. It is also important to note that the CIs for the “majority TCAR” group for CEA stroke outcomes are wider than those for the other groups given the smaller cohort size and are therefore less precise of an estimate. We were also unable to completely adjust for surgeon-specific patient selection that may have resulted in a healthier TCAR cohort. In addition, we did not evaluate complexity of disease on outcomes. This study did not evaluate transfemoral CAS as it is not as commonly performed, and its outcomes are not comparable to those of CEA and TCAR.²⁰ Follow-up studies should evaluate this relationship by surgeon experience and whether these observations are comparable between different center types.

CONCLUSIONS

This study contributes to our understanding of risk factors for postoperative stroke after carotid artery revascularization procedures and finds that the relative surgeon CEA to TCAR ratio is significantly associated with stroke after CEA and with TCAR. Surgeons who offer both operations should maintain a balanced practice and consider collaboration for challenging cases.

ARTICLE HIGHLIGHTS.

• Type of Research: Multicenter, retrospective cohort study

• Key Findings: Relative surgeon volume ratio of carotid endarterectomy (CEA) to transcarotid artery revascularization (TCAR) procedures is associated with postoperative stroke. Surgeons who perform majority CEA or TCAR have higher postoperative stroke in the operation performed the least. There was no association between surgeon ratio and postoperative death, return to the operating room, or cranial nerve injury for either procedure.

• Take Home Message: Surgeons who offer CEA and TCAR for carotid artery disease should maintain a balanced practice in both operations. Collaboration for challenging cases should be considered.