A mortality prediction model for elderly patients with critical limb ischemia

Abstract Objective To aid physicians in the process of shared decision-making, many predictive models for critical limb ischemia (CLI) have been constructed. However, none of these models is in widespread use. Predicting survival outcomes for a specific individual may be used to guide treatment selection. The aim of this study was to construct a 6-month survival-predicting model representative of elderly patients with CLI undergoing surgical or endovascular treatment. Methods An observational cohort study including all patients with CLI aged ≥65 years who underwent surgical or endovascular treatment of CLI between January 2013 and June 2018 was conducted. The model to predict survival at 6 months was based on a multivariable Cox proportional hazards regression model and a penalized likelihood method. The performance of the model was judged by means of the area under the receiver operating characteristic curve. Results In total, 449 patients were included in the study population. The median age was 76 years (range, 65-97 years), and 52.8% of the population was male. Surgical treatment was performed in 303 patients (67.5%), and 146 underwent endovascular treatment (32.5%). The estimated 30-day survival was 92.7% (standard error [SE], 1.2%); 6-month survival, 80% (SE, 1.9%); and 12-month survival, 71% (SE, 2.1%). Variables with the strongest association with 6-month mortality were age, living in a nursing home, physical impairment, and American Society of Anesthesiologists class. The area under the receiver operating characteristic curve of the 6-month mortality model was 0.81 (95% confidence interval, 0.76-0.85; P  Conclusions A prediction model constructed for 6-month mortality of elderly patients undergoing surgical or endovascular treatment of CLI showed that age, living in a nursing home, physical impairment, and American Society of Anesthesiologists class have the highest association with an increase in mortality. These factors may be used to identify patients at risk for mortality in shared decision-making.