BACKGROUND: Level I evidence has shown that patch angioplasty, when utilized during conventional carotid endarterectomy (CEA), results in lower rates of restenosis. However, it remains unclear how this information has affected practice patterns and outcomes in “real-world” vascular surgery settings.
METHODS: Within the Vascular Study Group of Northern New England (VSGNNE), we studied 2,934 patients undergoing 3,118 first-time CEA between January, 1st 2003 and December 31st, 2007. Rates of restenosis (defined as on duplex at 1 year follow up) were calculated using life-table analysis, and Cox proportional-hazards models were used to identify multivariate predictors of restenosis.
RESULTS: Overall, 2,724 (87%) of patients underwent conventional CEA, and 394 (13%) underwent eversion CEA. Of patients undergoing conventional CEA, the proportion performed with patching increased over time (87% in 2003, 95% in 2008, p<0.001). Concomitantly, the use of eversion CEA declined significantly (17% of all cases in 2003 to 8% of all cases in 2007, p<0.001). Overall, restenosis occurred in 313 patients (10%), but most commonly was non-critical in nature (7.6% had a 50-80% restenosis, 1.8% had an 80-99% restenosis, and 0.5% had an occlusion). In univariate analyses, we found significant differences in critical (80-99%) restenosis by procedure type (1.1 % in conventional CEA, 6.6% in eversion CEA, p<0.002), the year of procedure (3% in 2003, 0.6% in 2008, p<0.03), and use of patching (2.9% no patch, 1.0% with patch, p<0.008). Among patients undergoing conventional endarterectomy, multivariate analysis demonstrated that surgery by a hospital with a high restenosis rate in prior years (HR 2.6, 95% CI 1.4-4.9), absence of patching (HR 2.7, 95% CI 1.3-5.4), and contralateral 80% ICA stenosis (HR 3.2, 95% CI 1.5-7.0) were independently associated with a higher risk of an 80-99% restenosis at 1 year. Of the 58 patients with 80-99% restenosis, re-intervention was performed within the first year in 13 patients, consisting of 3 re-operations and 10 carotid artery stent procedures. Of the 16 patients with a carotid occlusion within 1 year, one patient suffered a major stroke and one a minor stroke.
CONCLUSIONS: In our region, clinically significant restenosis after CEA fell slightly over time, associated with an increase in patching and a decline in the use of eversion endarterectomy. Other factors, such as the degree of contralateral stenosis, also affected restenosis rates following CEA. Quality improvement efforts aimed toward increasing the use of patch angioplasty may offer an opportunity to limit restenosis following CEA.