Back to 2026 Abstracts
Validation of the Vascular Quality Initiative Cardiac Risk Index for Lower Extremity Bypass in the BEST-CLI Trial Open Cohort
Nathan T. p. Patel, MD1, Gheorghe Doros, PhD MBA
2, Michael S. Conte, MD
3, Phillip P. Goodney, MD, MS
4, Michael B. Strong, MA
5, Alik Farber, MD, MBA
2, Matthew T. Menard, MD
5, Kenneth Rosenfield, MD, MHCDS
6, Daniel J. Bertges, MD
1, Mahommad H. Eslami, MD, MPH, MBA
7.
1University of Vermont, Burlington, VT, USA,
2Boston University, Boston, MA, USA,
3University of California San Francisco, San Fransisco, CA, USA,
4Dartmouth Hitchcock Medical Center, Lebanon, NH, USA,
5Brigham and Women's Hospital, Boston, MA, USA,
6Massachusetts General Hospital, Boston, MA, USA,
7West Virginia University, Charleston, WV, USA.
Objective The Vascular Quality Initiative (VQI) Cardiac Risk Index (CRI) for lower extremity revascularization (LER) is a risk prediction model proposed for the estimation of in-hospital post-operative myocardial infarction (POMI). The objective is to externally validate this model in the open revascularization cohort (BEST) from Best Endovascular versus Best Surgical Therapy in Patients with Critical Limb Ischemia (BEST-CLI) Trial.
Methods First, a modified version of the VQI CRI risk assessment tool for LER was constructed to ensure the predictive variables used in the VQI CRI risk model were available in the BEST-CLI database. The calibration slope was used to assess the degree of agreement between the observed and predicted risks for POMI. Classification performance of the model was assessed using area under the receiver operating curve (AUC).
Results BEST included 857 patients and VQI registry included 14,442 patients. The cohorts had similar POMI rates (BEST 30-day: 2.8% and VQI in-hospital: 2.6%) but BEST had higher rate of angina or prior myocardial infarction (23.8% vs. 7.9%, p<0.001), diabetes mellitus treated with insulin (47% vs. 26.1%, p<0.001), end stage renal disease (9.6% vs. 5.4%, p<0.001), Non-White race (28% vs. 9.2%, p<0.001), and infrapopliteal distal anastomosis (54.5% vs. 39.9%, p<0.001). BEST patients with POMI were similar to those without POMI except for a higher rate of preexisting coronary artery disease with angina or prior MI (42% vs 23%, P= .020). External validation of risk model on BEST demonstrated an observed calibration with intercept of 0.004 and slope of 0.807 (Figure 1). With respect to discrimination the VQI CRI revealed an AUC of 0.718 at the observed rate of POMI.
Conclusion In this first external validation of VQI CRI, the model demonstrated good performance for prediction of 30-day POMI. The model has good agreement between observed and predicted POMI and the measured AUC of 0.718 was similar to the discriminatory power measured in the original VQI CRI model (AUC: 0.74, 95% confidence interval, 0.72-0.76). Further analyses may identify patients factors to improve the risk model performance, better allowing for its widespread use during preoperative discussions with patients and family.
Back to 2026 Abstracts
