Background: Half the patients who develop a diabetic foot ulcer will die within 5 years, most commonly from heart attacks and stroke. However, we do not know whether ulcer severity or common mediators of vascular disease, such as statin use, are associated with mortality. The aim of this study is to identify predictors of mortality for patients with an incident foot ulcer.
Methods: Using US Department of Veterans Affairs healthcare datasets, we constructed a national, retrospective, rolling cohort of Veterans with type 2 diabetes who developed an incident diabetic foot ulcer between January 1, 2006 and September 1, 2010. Patients were followed until death or the end of the study period, January 1, 2011. Ulcers were characterized as early stage, osteomyelitis, or gangrene at the time of presentation. Comorbidities, hemoglobin A1C, cholesterol, systolic blood pressure, statin use, healthcare utilization were determined in the 12-month period preceding ulceration. Kaplan Meier survival curves, stratified by ulcer severity were constructed. Stepwise Cox proportional hazard modeling identified independent predictors of death.
Results: Of the 66,323 Veterans included in the cohort, 33,554 died during the study period (50.59%). Compared to early stage ulcers, gangrene was associated with an increased risk of mortality (HR 1.70, 95% CI 1.63 1.78, Figure). The magnitude of this effect was greater than estimates for coronary artery disease (HR 1.15, 95% CI 1.13 1.16), peripheral arterial disease (HR 1.09, 95% CI 1.07 1.11), and stroke (HR 1.14, 95% CI 1.12 1.17). 71% of the cohort was prescribed a statin at baseline, which was protective (HR 0.89, 95% CI 0.87 0.90). Systolic blood pressure and hemoglobin A1C were best modeled with U-shaped curves to reflect either stringent or lax control being associated with an increased hazard of death.
Conclusion: Diabetic foot ulcers are common and portend a high risk of death. Gangrene is a strong, independent predictor of subsequent mortality, out-performing diagnosed vascular disease. This may reflect previously unrecognized or underestimated vascular disease. Statin use in this population may be particularly beneficial, and initiating statins should be further explored as a means of lowering associated mortality.
B. Bartle, None
J. Cooper, None
J. Kang, None
M. Smith, None
M. W. Sohn, None
C. Crnich, None
See more of: Poster Abstract Session