Abstract 119: Can CREST Procedural Results be Reproduced in General Practice in Post CREST Era?: Analysis of National Surgical Quality Improvement Program (NSQIP) Registry

Abstract

Background: The Carotid revascularization for primary prevention of stroke assessed in CREST trial (year 2010) revealed risk of the composite outcome of stroke, myocardial infarction, or death were almost similar in carotid artery stenting (CAS) and carotid artery endarterectomy (CEA) group. This study analyzed such current clinical practice and outcome (year 2011 to 2013) from multicenter national registry described below. Methods: Cohort of non-high risk patients who underwent targeted CAS or CEA were collected from American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) Registry in post CREST era. This registry prospectively collects around more than 300 variables across the nation pertaining to patient characteristics, co-morbid conditions, operative details, and 30-day postoperative outcomes. Frequency analysis of such outcome was performed to assess the re-validation of CREST trial in current clinical practice. Results: In Post CREST NSQIP Registry we identified 142 (CAS) and 7834 (CEA) patients. In Post CREST era (current clinical practice) the rates of any stroke among CEA patients (1.9%, 95% CI 1.57-2.17, CREST Ref 2.3%) (p=0.25) and CAS patients (2.1%, 95% CI 0.54-5.75, CREST Ref 4.1%) (p=0.25) observed in clinical practice were comparable to those observed in CREST trial. The rates of MI/arrhythmia among CEA patients (1.7%, 95% CI 1.47-2.06, CREST Ref 2.3%) (p=0.22) and CAS patients (2.8%, 95% CI 0.89-6.79, CREST Ref 1.1%) (p=0.13) are comparable from CREST trial. The rates of 1 month mortality among CEA patients (0.6%, 95% CI 0.42-0.76, CREST Ref 0.3%) (p=0.27) and CAS patients (0.7%, 95% CI 0.04-3.47, CREST Ref 0.7%) (p=0.92) were also similar. The rate of cranial nerve injury in CEA patients has declined (2.1%, CI 1.83-2.47, CREST Ref 4.7 %) (p<0.0001).The combined primary endpoint for CEA patients (3.7%, 95% CI 3.32-4.17, CREST Ref 4.5%) (p=0.19) and CAS patients (5.6%, 95% CI 2.62-10.7, CREST Ref 5.2%) (p=0.81) was similar to CREST trial. Conclusions: Our results support reproducibility of CREST procedural outcomes for both CEA and CAS in general practice in post-trial period.