Incidence of Lower-Extremity Amputation in American Indians

Incidence of Lower-Extremity Amputation in American Indians The Strong Heart Study Helaine E. Resnick , PHD, MPH 1 , Elizabeth A. Carter , MPH 1 , Jay M. Sosenko , MD, MS 2 , Susan J. Henly , PHD, RN 3 , Richard R. Fabsitz , PHD 4 , Frederick K. Ness , MD 5 , Thomas K. Welty , MD 6 , Elisa T. Lee , PHD 7 and Barbara V. Howard , PHD 1 1 MedStar Research Institute, Hyattsville, Maryland 2 University of Miami School of Medicine, Miami, Florida 3 School of Nursing, University of Minnesota, Minneapolis, Minnesota 4 National Heart, Lung, and Blood Institute, Bethesda, Maryland 5 Diabetes Program, Mille Lacs Band of Ojibwe Indians, Mille Lacs, Minnesota 6 Missouri Breaks Research, Timber Lake, South Dakota 7 University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma Address correspondence and reprint requests to Helaine E. Resnick, PhD, MPH, Department of Epidemiology and Statistics, MedStar Research Institute, 6495 New Hampshire Ave., Suite 201, Hyattsville, MD 20783. E-mail: helaine.e.resnick{at}medstar.net Abstract OBJECTIVE —To define incidence and predictors of nontraumatic lower-extremity amputation (LEA) in a diverse cohort of American Indians with diabetes. RESEARCH DESIGN AND METHODS —The Strong Heart Study is a study of cardiovascular disease and its risk factors in 13 American-Indian communities. Data on the presence/absence of amputations were collected at each of three serial examinations (1989–1992, 1993–1995, and 1997–1999) by direct examination of the lower extremity. The logistic regression model was used to quantify the relationship between risk of LEA and potential risk factors, including diabetes duration, HbA 1c , peripheral arterial disease, and renal function. RESULTS —Of the 1,974 individuals with diabetes and without prevalent LEA at baseline, 87 (4.4%) experienced an LEA during 8 years of follow-up, and a total of 157 anatomical sites were amputated among these individuals. Amputation of toes was most common, followed by below-the-knee and above-the-knee amputations. Age-adjusted odds of LEA were higher among individuals with unfavorable combinations of risk factors, such as albuminuria and elevated HbA 1c . Multivariable modeling indicated that male sex, renal dysfunction, high ankle-brachial index, longer duration of diabetes, less than a high school education, increasing systolic blood pressure, and HbA 1c predicted LEA risk. CONCLUSIONS —The 8-year cumulative incidence of LEA in American Indians with diabetes is 4.4%, wit