Back to 2016 Annual Symposium Abstacts

Introduction: Since the 2004 approval by the FDA of carotid stenting (CAS), there have been two seminal publications about CAS reimbursement (CMS guidelines [CMSG]; 2008) and clinical outcomes (CREST; 2010). We explored the association between these publications and national trends in CAS utilization among high-risk symptomatic patients.
Methods: The most recent datasets of the Nationwide Inpatient Sample (NIS) was queried for patients undergoing carotid revascularization. Sample was limited to include only patients with symptomatic carotid stenosis and those defined as “high-risk” if they had a Charlson Comorbidity Score of ≥ 3.0. Utilization proportions of CAS were calculated quarterly (Q) from 2005-2011 for NIS. Three time intervals related to CMSG and CREST publication were selected 2005-2008(P1), 2008-2010 (P2), and after 2010 (P3). Logistic regression with piecewise linear trend for time was used to estimate different trends in CAS utilization for overall samples and for neurologically asymptomatic and symptomatic cases. Mutlivariate logistic regression was employed to compare odds of post-operative mortality and stroke between these two procedures at different time intervals independent of confounding variables.
Results: CAS utilization constituted 20.5% of carotid revascularization procedures among high-risk symptomatic patients with a significant increase from 18.6 to 24.4% during the study period; p<0.001. There was an initial increase during P1 in the rate of CAS. Compared to P1, CAS utilization significantly decreased during P2 by a 3.3% decline in the odds ratio (OR) of CAS per quarter (OR: 0.967, 95% confidence interval (CI): 0.943-0.993, p= 0.002). After CREST (P3), CAS utilization continued to increase significantly from the pre- to the post- publication time interval. The odds of in-hospital mortality (OR: 2.56, 9% CI: 1.17-5.62, p=0.019) and postoperative stroke (OR: 1.53, 95% CI: 1.09-3.68, p=0.024) were independently and significantly higher for CAS patients in the overall sample.
Conclusions: CMS guidelines allowing payment for symptomatic high risk patients was associated with a decreased utilization of CAS whereas CREST publication was associated with a significant increase in the CAS utilization among high-risk symptomatic patients.