Validation and Comparison of the Prognosis Predicting Ability of Inflammation-Based Scores Following Endovascular Treatment for Peripheral Artery Disease

Abstract

We assessed the prognostic ability of several inflammation-based scores and compared their long-term outcomes in patients with peripheral artery disease (PAD) following endovascular treatment (EVT). We included 278 patients with PAD who underwent EVT and classified them according to their inflammation-based scores (Glasgow prognostic score [GPS], modified GPS [mGPS], platelet to lymphocyte ratio [PLR], prognostic index [PI], and prognostic nutritional index [PNI]). Major adverse cardiovascular events (MACE) at 5 years were examined, and C-statistics in each measure were calculated to compare their MACE predictive ability. During the follow-up period, 96 patients experienced MACE. Kaplan–Meier analysis showed that higher scores of all measures were associated with a higher MACE incidence. Multivariate Cox proportional hazard analysis showed that GPS 2, mGPS 2, PLR 1, and PNI 1, compared with GPS 0, mGPS 0, PLR 0, and PNI 0, were associated with an increased risk of MACE. C-statistics for MACE for PNI (.683) were greater than those for GPS (.635, P = .021), mGPS (.580, P = .019), PLR (.604, P = .024), and PI (.553, P < .001). PNI is associated with MACE risk and has a better prognosis-predicting ability than other inflammation-scoring models for patients with PAD following EVT.