Author + information
- Kunihiro Matsushita,
- Shoshana H. Ballew,
- Corey Kalbaugh,
- Bethany Warren,
- Laura Loehr,
- Anna Kucharska-Newton,
- Ron Hoogeveen,
- Christie Ballantyne,
- Gerardo Heiss,
- Aaron Folsom,
- Alan Hirsch,
- Josef Coresh and
- Elizabeth Selvin
Background: The ability of non-traditional markers for predicting ischemic leg amputation in diabetes is unknown.
Methods: In 3321 ARIC participants with diabetes (45-73 y during 1987-98), we identified traditional factors contributing to ischemic leg amputation (ICD code of leg amputation [84.1x] with code of leg artery disease [eg, 440.2]) with follow-up through 2013. We explored prediction improvements with non-traditional markers (coagulation factors, C-reactive protein, albuminuria, glycemic markers, retinal findings, and cardiac markers) assessed at different visits 1-4.
Results: There were 127 incident ischemic amputation cases. Of traditional factors, sex, race, kidney function, blood pressure, adiposity, and prevalent coronary disease had significant associations with amputation, constructing the base model (c-statistic 0.71-0.77 at each visit as baseline). Of non-traditional markers, only glycemic markers (HbA1c, glycated albumin, fructosamine, and 1-5AG) and high-sensitivity troponin T (hsTnT) (assessed at visit 2) improved c-statistic by >0.1 (0.84-0.85 from 0.71), with additional improvements by combining any glycemic marker with hsTnT (c-statistic 0.88-0.89) (Figure). These models with traditional factors, a glycemic marker, and hsTnT were validated using visit 4 data (c-statistic 0.84-0.88).
Conclusions: Our models with traditional factors, a glycemic marker, and hsTnT can be used to identify diabetic patients at particularly high risk of ischemic amputation.
Moderated Poster Contributions
Vascular Medicine Moderated Poster Theater, Poster Hall, Hall C
Sunday, March 19, 2017, 10:00 a.m.-10:10 a.m.
Session Title: Improving Risk Assessment in Atherosclerosis
Abstract Category: 40. Vascular Medicine: Non Coronary Arterial Disease
Presentation Number: 1310M-05
- 2017 American College of Cardiology Foundation