A Novel Amputation-Specific Index for the Prediction of Major Amputation Despite a Patent Infrainguinal Bypass

Document Type

Conference Proceeding

Publication Date

6-1-2024

Publication Title

J Vasc Surg

Abstract

Objectives: While lower extremity limb salvage parallels infrainguinal bypass (IIB) graft patency, some patients who receive an IIB for peripheral artery disease still end up with a major amputation even with a patent bypass graft. This study aims to derive the risk factors associated with major amputation despite bypass patency and develop a weighted index for the prediction of this outcome. Methods: The Vascular Quality Initiative (VQI) IIB module was queried for all patients between 2003 and 2021. Patients with an occluded IIB were excluded. Patients receiving a major amputation with documentation of a patent bypass graft at the time or after the amputation date were included. The outcome of interest was major amputation. Univariate and multivariate analyses were used to identify variables associated with amputation. A risk predictive logistic model was derived using a 70% derivation cohort and validated on the remaining 30%. Finally, the amputation-specific index (ASI) was created and assessed with discrimination and calibration abilities. Results: A total of 21,973 patients (mean age, 66.3 years) were included. Of the full cohort, 1536 (7.0%) patients received an amputation despite a patent IIB. The most important predictors of the primary outcome on multivariate analysis included Black race (OR, 1.58; 95% CI, 1.33-1.87), dialysis (OR, 1.64; 95% CI, 1.26-2.12), non-ambulatory status (OR, 1.64; 95% CI, 1.28-2.12), emergency surgery (OR, 1.58; 95% CI, 1.34-1.86), rest pain (OR, 2.35; 95% CI, 1.80-3.06), tissue loss (OR, 3.92; 95% CI, 3.06-5.03), prosthetic graft (OR, 1.50; 95% CI, 1.29-1.75), infra-popliteal target (OR, 2.44; 95% CI, 2.08-2.86), prior ipsilateral IIB (OR, 1.88; 95% CI, 1.58-2.23), and persistence of ischemic symptoms (OR, 4.46; 95% CI, 3.82-5.20). The final ASI model encompassed the 10 predictors with different weights assigned to each predictor. The ASI performance and calibration testing provided an area under receiver operator curve (AUROC) of 80% and 77% on derivation and validation cohorts, respectively (Fig 1), with a calibration R-squared = 0.99 and proper goodness of fit. To better display the ASI index, values were divided into ranges and grouped into four risk stages, for which the risk of amputation was calculated (Table I). Conclusions: Seven percent of patients who undergo IIB may need an amputation despite a patent graft. While wound, ischemia, and foot infection prove to be important determinants of limb salvage, our risk index had a similar predictive value to the WIfI classification system, and can accurately predict amputation risk in patients planned for IIB based on preoperative and operative characteristics. [Formula presented] [Formula presented]

Volume

79

Issue

6

First Page

e263-e264

Find It @ Sladen

Share

COinS