Optimal cut-off value of preprocedural geriatric nutritional risk index for predicting the clinical outcomes of patients undergoing endovascular revascularization for peripheral artery disease

BackgroundMalnutrition measured by the geriatric nutritional risk index (GNRI) was reported to be associated with poor prognosis for patients with peripheral artery disease (PAD). However, the optimal cut-off value of preprocedural GNRI for critical limb ischemia (CLI) and intermittent claudication (IC) is unknown. We aimed to determine its optimal cut-off value for CLI or IC patients requiring endovascular revascularization.MethodsWe explored data of 2246 patients (CLI: n = 1061, IC: n = 1185) registered in the Tokyo-taMA peripheral vascular intervention research COmraDE (TOMA-CODE) registry, which prospectively enrolled consecutive PAD patients who underwent endovascular revascularization in 34 hospitals in Japan from August 2014 to August 2016. The optimal cut-off values of GNRI were assessed by the survival classification and regression tree (CART) analyses, and the survival curve analyses for major adverse cardiovascular and limb events (MACLEs) were performed for these cut-off values.ResultsIn addition to the first cut-off value of 96.2 in CLI and 85.6 in IC, the survival CART provided an additional cut-off value of 78.2 in CLI and 106.0 in IC for further risk stratification. The survival curve was significantly stratified by the GNRI-based malnutrition status in both CLI [high risk: 47.7% (51/107), moderate: 30.1% (118/392), and low: 10.2% (53/520), log–rank p < 0.001] and IC [high risk: 14.3% (7/49), moderate: 4.5% (29/646), and low: 0.5% (2/407), log–rank p < 0.001]. The multivariate Cox-proportional hazard analysis showed that a higher GNRI was significantly associated with a better outcome in both CLI [hazard ratio (HR) per 1-point increase: 0.97, 95% CI: 0.96–0.98, p < 0.001] and IC (HR: 0.94, 95% CI: 0.91–0.97, p < 0.001).ConclusionsPreprocedural nutritional status significantly stratified future events in patients with PAD. Given that the optimal cut-off value of GNRI in CLI was almost 10-points lower than that of IC, using a disease-specific cut-off value is important for risk stratification.