Clinical and radiographic risk factors for operative stroke and death in the European carotid surgery trial.

OBJECTIVES: carotid endarterectomy is associated with significant morbidity and mortality. A better understanding of the relationships between baseline characteristics and outcome may help to reduce the risks of surgery. In order to make accurate and unbiased estimates of surgical risk it is important to study cohorts of patients that were established prospectively, where independent physicians assessed outcome, and where the decision to analyse and report the results was not data-dependent. The surgical arm of the European Carotid Surgery Trial (ECST) is such a cohort. METHODS: the 30-day outcome of carotid endarterectomy was analysed in ECST surgery patients in relation to their baseline clinical and angiographic characteristics. The severity of operative strokes was compared with that of strokes that occurred in the medical group. RESULTS: 1729 patients underwent trial surgery. There were 17 deaths (1.0%, 95% CI=0.6-1.6) and 105 non-fatal major strokes (6.1%, 95% CI=5.0-7.3) within 30 days of surgery. The risk of major stroke or death was 7.1% (95% CI=5.9-8.4). The risk of disabling or fatal stroke was 3.0% (95% CI=2.1-3.8). The ratio of disabling to non-disabling operative strokes was similar to that in the medical group. Several baseline characteristics predicted the operative risk of stroke and death in univariate analyses, but only four were independent risk factors in a multiple regression analysis: presentation with cerebral TIA vs ocular ischaemic events only (HR=2.99, 95% CI=1.33-6.69, p=0.008); female sex (HR=2.04, 95% CI=1.37--3.06, p=0.001); systolic hypertension (HR=1.01 per 10 mmHg, 95% CI=1.00-1.02, p=0.03) and peripheral vascular disease (HR=2.17, 95% CI=1.17-2.89, p=0.001). CONCLUSIONS: the operative risk of stroke and death in the ECST was comparable with other prospective studies and trials in which patients were assessed postoperatively by both a physician and a surgeon. Case fatality and disability after operative stroke are similar to strokes that occur on medical treatment only. Several baseline patient characteristics predict surgical risk and it may be possible to use these characteristics to aid patient selection and surgical audit.