Temporal trends in safety of carotid endarterectomy in asymptomatic patients: systematic review.

OBJECTIVE: To systematically review temporal changes in perioperative safety of carotid endarterectomy (CEA) in asymptomatic individuals in trial and registry studies. METHODS: The MEDLINE and EMBASE databases were searched using the terms "carotid" and "endarterectomy" and "asymptomatic" from 1947 to August 23, 2014. Articles dealing with 50%-99% stenosis in asymptomatic individuals were included and low-volume studies were excluded. The primary endpoint was 30-day stroke or death and the secondary endpoint was 30-day all-cause mortality. Statistical analysis was performed using random-effects meta-regression for registry data and for trial data graphical interpretation alone was used. RESULTS: Six trials (n = 4,431 procedures) and 47 community registries (n = 204,622 procedures) reported data between 1983 and 2013. Registry data showed a significant decrease in postoperative stroke or death incidence over the period 1991-2010, equivalent to a 6% average proportional annual reduction (95% credible interval [CrI] 4%-7%; p < 0.001). Considering postoperative all-cause mortality, registry data showed a significant 5% average proportional annual reduction (95% CrI 3%-9%; p < 0.001). Trial data showed a similar visual trend. CONCLUSIONS: CEA is safer than ever before and high-volume registry results closely mirror the results of trials. New benchmarks for CEA are a stroke or death risk of 1.2% and a mortality risk of 0.4%. This information will prove useful for quality improvement programs, for health care funders, and for those re-examining the long-term benefits of asymptomatic revascularization in future trials.