Prognostic value of the clinical examination of the diabetic foot ulcer.
OBJECTIVE: To determine the value of the history, physical examination, and magnetic resonance imaging (MRI) in predicting successful primary healing of a foot ulcer in a diabetic patient. DESIGN: Prospective cohort study. SETTING: Durham (NC) Veterans Affairs Medical Center. PATIENTS: Sixty-four consecutive diabetic patients with 78 dermal ulcers through the full thickness of the skin and at or distal to the malleoli of the ankle. MEASUREMENTS AND MAIN RESULTS: A structured clinical history and physical examination were performed by two examiners, a physician participating in the study and the referring physician. Fifty of these patients with 63 ulcers underwent MRI. Patients were followed prospectively for 6 months after enrollment to ascertain healing of the ulcer, amputation, and death. During the 6-month follow-up period, 8 (13%) of the patients died. Seventeen (22%) of the ulcers were amputated, 17 (22%) of the ulcers failed to heal, and 36 (47%) healed primarily. Univariate predictors of healing at 6 months included age less than 65 years, diagnosis of diabetes within the last 15 years, painless ulcer, palpable ankle pulse, anklebrachial index greater than 0.5, and the physician's assessment of the overall likelihood of osteomyelitis. In a multivariable logistic regression model, predictors of healing included the presence of an audible pulse on Doppler examination (p = .01) and a painless ulcer (p = .04). The diagnosis of osteomyelitis on MRI did not predict healing in these patients. CONCLUSIONS: Foot ulcers in patients with diabetes frequently have poor outcomes; fewer than half the patients in this study healed their ulcers within 6 months. The vascular components of the clinical examination are the best predictors of healing in patients with a diabetic foot ulcer.
Edelman, D; Hough, DM; Glazebrook, KN; Oddone, EZ
Volume / Issue
Start / End Page
Pubmed Central ID
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)