Physicians' preference for carotid revascularization impacts postoperative stroke and death outcomes.

Objective: Transcarotid artery revascularization (TCAR) is a safe minimally invasive option for patients with carotid artery stenosis who are not appropriate candidates for carotid endarterectomy (CEA). Many physicians have not yet adopted this technique in the management of carotid artery stenosis. The aim of this study is to explore overall outcomes of carotid revascularization based on physicians' practices in the Vascular Quality Initiative (VQI).

Methods: Individual physicians participating in both the carotid artery stenting (CAS) and carotid endarterectomy (CEA) modules in VQI were categorized as performing CEA and TCAR, CEA and transfemoral carotid artery stenting (TFCAS), or all three procedures (CEA, TCAR, and TFCAS). Physicians performing CEA only or TCAR/TFCAS only were excluded. In-hospital and 1-year outcomes were compared between the three groups using univariable and multivariable analysis.

Results: A total of 104,925 carotid revascularization procedures performed by 1433 physicians were included. Most physicians performed CEA and TCAR (n = 714; 49.8%), whereas 35.1% (n = 503) performed all three procedures, and 15.1% (n = 216) performed CEA and TFCAS only. Physicians performing CEA and TFCAS had higher overall stroke/death rates after carotid procedures (2.2%) compared with those performing CEA and TCAR (1.4%) or those performing all three procedures (1.6%; P < .001). They also had higher rates of cranial nerve injuries (3.1% vs 1.9% vs 1.9%; P < .001). After adjusting for baseline characteristics, procedures performed by CEA and TFCAS physicians had significantly higher odds of in-hospital stroke/death compared with those in the CEA and TCAR group (odds ratio, 1.31; 95% confidence interval [CI], 1.03-1.66; P = .03). They also had increased hazard of 1-year stroke/death (hazard ratio, 1.45; 95% CI, 1.1-1.9; P = .01). No significant difference in the adjusted odds of stroke/death was observed between CEA and TCAR performers vs CEA, TCAR, and TFCAS performers (odds ratio, 1.05; 95% CI, 0.92-1.20; P = .44). When adjusting for the type of carotid revascularization technique, difference in outcomes based on surgeon's experience were no longer significant, indicating that differences in outcomes were procedure-specific and attributable to the inferior outcomes associated with TFCAS compared with CEA and TCAR. TCAR case volumes did not impact outcomes of carotid revascularization. On the other hand, a high TFCAS volume among physicians performing all three carotid procedures was associated with higher overall in-hospital and 1-year mortality.

Conclusions: Physicians' preference for carotid artery stenosis management has a bearing on their overall stroke/death rates. Careful patient and procedure selection are the cornerstone to improve carotid revascularization outcomes.