A model for predicting the time of early symptomatic restenosis after infrapopliteal angioplasty in patients with critical limb ischemia.

The objective of this study is to develop a model for predicting the time of early symptomatic (delayed or nonhealing wound) restenosis after infrapopliteal angioplasty in patients with critical limb ischemia (CLI). This is a single-center retrospective cohort study evaluating 60 de novo infrapopliteal lesions of 38 limbs in 35 patients with CLI, who underwent successful endovascular treatment (EVT) from October 2016 to December 2018 and follow-up angiography within 3 months from the procedure. Outcome measures were binary restenosis at follow-up angiography and clinical outcome at 3 months. Patient/limb/lesion characteristics were compared between the restenosis and non-restenosis groups. Angiographic restenosis predictors were assessed to develop a model for predicting the time of restenosis using multinomial logistic regression. The restenosis rate at follow-up angiography (median time, 41 days [IQR 27-58 days]) was 38% (23/60). After adjustment for covariables, longer period between EVT and follow-up angiography and lower C-reactive protein (CRP) were the predictors of angiographic restenosis. We developed a model for predicting the time of early symptomatic restenosis with a probability of 70%: "Days = 200 - 2.1 age - 13 CTO + 3.3 CRP" (R = 0.81, RMSE = 0.27), e.g., 80 years old, CTO (+), CRP 4.4 mg/dl: 32.2 days. The predictive model including age, CTO, and CRP might allow estimation of the period for the angiographic restenosis development.