Development and validation of a score to predict life expectancy after carotid endarterectomy in asymptomatic patients

OBJECTIVE: Recent improvement of best medical treatment for carotid stenosis has sparked a debate on the role of surgery-identification of patients who may benefit from carotid endarterectomy (CEA) is crucial to avoid overtreatment. An expected 5-year postoperative survival is one of the main selection criteria. The aim of this study was the development of a score for predicting survival of asymptomatic patients after CEA. METHODS: Our score was derived from a retrospective analysis of 648 consecutive asymptomatic patients from a single hospital. External validation of the score was then performed on a second cohort of 334 asymptomatic patients from two different hospitals in the same area. Factors associated with reduced postoperative survival within the derivation cohort (DC) were identified and tested for statistical significance. Each selected factor was assigned a score proportional to its β coefficient: 1 point for chronic obstructive pulmonary disease, diabetes mellitus, coronary artery disease, and lack of statin treatment; 4 points for age 70 to 79 years and creatinine concentration ≥1.5 mg/dL; 8 points for age ≥80 years and dialysis. The DC was divided into four groups based on individual scores: group 1, 0 to 3 points; group 2, 4 to 7 points; group 3, 8 to 11 points; and group 4, ≥12 points. Group-specific survival curves were calculated. The validation cohort (VC) was stratified according to the score. Survival of each of the four risk groups within the VC was compared with its analogue from the DC. RESULTS: Median follow-up of the DC and VC was, respectively, 56 and 65 months. Intercohort comparison of 5-year survival was 84.7% ± 1.7% vs 85.2% ± 2% (P = .41). Group-specific 5-year survival within the DC was 97% ± 1.5% (group 1), 88.4% ± 2.2% (group 2), 69.6% ± 4.7% (group 3), and 48.1% ± 13.5% (group 4; P < .0001). Five-year survival within the VC was 95.5% ± 2% (group 1), 89.5% ± 2.7% (group 2), 65% ± 6.1% (group 3), and 44.8% ± 14.1% (group 4; P < .0001). Intercohort comparison of group-specific survival curves showed close similarity throughout the groups. CONCLUSIONS: Our score is a simple clinical tool that allows a quick and reliable prediction of survival in asymptomatic patients who are candidates for CEA. This selective approach is crucial to avoid unnecessary surgery on patients who are less likely to survive long enough to experience the benefits of this preventive procedure.