Development of an individualized scoring system to predict mid-term survival after carotid endarterectomy

Carotid endarterectomy (CEA) is a prophylactic surgery focused in preventing stroke in the mid-long term. The purpose of this study was to analyze mid-term mortality in patients undergoing CEA, identify predictors of 3-year mortality and design a score to estimate individual risk of mortality in this population.

A retrospective single-center study including consecutive patients undergoing CEA between 1997-2010. Demographic data and comorbidities, postoperative results and patient follow-up data were registered and evaluated. Kaplan Meier analysis was used to analyze survival. After multivariable COX regression analysis, a score based on the calculated Hazards Ratios (HR) was designed. The sum of all points performed the individual score for each patient for estimating 3-years mortality. Population was stratified into four groups according to percentiles of score obtained: Group A (-7 to 4 points), Group B (5-8 points), Group C (9-10 points), Group D (score greater than 11 points).

A total of 453 patients with a mean follow-up of 53.4 months were included in the study. Overall 3-year survival was 88.4%. On the univariate analysis the variables associated with significant increasing in 3-year mortality were: female gender (OR 2.32), diabetes mellitus (OR 2.28), COPD (OR 2.98), ischemic heart disease (OR 2.29), critical carotid stenosis >90% (OR 2.16) and antiplatelet therapy as a protective factor (OR 0,23). Factors associated with mortality in multivariate analysis were age (HR 1.14 P=0.001), diabetes mellitus (HR 1.62, P=0.031), COPD (HR 1.88 P=0.022), ischemic heart disease (HR 1.59 P=0.05), critical stenosis >90% (HR 1.70 P=0.015) and antiplatelet therapy as a protective factor (HR 0.23 P=0.027). The scoring system includes the following items: female gender (+2 points), age (50-69 years +7 points, 70-79 years +12 points, >80 years +15 points), diabetes (+4 points), COPD (+5 points), ischemic heart disease (+4 points), carotid stenosis> 90% (+4 points). Antiplatelet (-7 points). The score range from -7 to 26 points. The 3-year mortality range was 5.6% (group A) versus 25.5% (group D). The incidence of stroke at 3-year folllow-up was not correlated with the score (99%, 100%, 97% and 94.5%, respectively groups A-D, P=0.11) CONCLUSIONS: The score developed based on the risk factors of mortality allows individualized risk prediction of 3-year mortality in patients with carotid stenosis. This represents a useful and practical tool for decision-making in the indication of the CEA, allowing surgeons to identify high-risk patients who would benefit from medical treatment due to their limited life expectancy, mainly in asymptomatic patients.