Evaluation of the BASIL Survival Prediction Model in Patients Undergoing Infrapopliteal Interventions for Critical Limb Ischemia.

ABSTRACT

BACKGROUND: Survival prediction models are clinical tools that help professionals make the best treatment decisions. In the treatment of critical limb ischemia, several scoring methods have emerged; however, many are limited because they are not reproducible in different populations. This study analyze the Bypass versus Angioplasty in Severe Ischemia of the Leg survival prediction model (BASIL SPM), exclusively in patients who underwent infrapopliteal (InfraPo) revascularization for the treatment of critical limb ischemia. METHODS: In this retrospective cohort, patients who underwent InfraPo interventions during a 4-year period (2009-2013) were consecutively included, according to the primary intervention (endovascular or open repair), irrespective of combined treatment in the femoropopliteal and InfraPo segments. Performance of the BASIL SPM was determined using the area under the curve of the receiver-operating characteristic curve. Additionally, secondary patency, limb salvage, and overall survival were analyzed by the Kaplan-Meier method. RESULTS: The mean follow-up duration was 32.7 ± 23.8 months. In the total patient group (n = 134), the mean age was 72 ± 8.6 years (range, 46-91 years), and the main associated comorbidities were hypertension (85.8%) and diabetes mellitus (76.8%). Regarding differences between the groups, patients in the endovascular group (n = 100) were older (73 vs. 69.5; P = 0.033) and more frequently stratified as high risk (61% vs. 38.2%; P = 0.030) than the open repair group (n = 34). The area under the curve (95% confidence interval [CI]) using the BASIL SPM at 6, 12, and 24 months was 0.499 (95% CI 0.344-0.657), 0.508 (95% CI 0.353-0.629), and 0.549 (95% CI 0.420-0.678), respectively. In the total patient group, the 36-month secondary patency, limb salvage, and overall survival were 31.5%, 81.1%, and 65.9%, respectively. CONCLUSIONS: The BASIL SPM was a poor predictor of life expectancy in this patient cohort.