Abstract. Ha Èma Èla Èinen H, Ro Ènnemaa T, Halonen J-P, Toikka T (Social Insurance Institution, Turku; University of Turku, Finland). Factors predicting lower extremity amputations in patients with type 1 or type 2 diabetes mellitus: a population-based 7-year follow-up study. J Intern Med 1999; 246: 97± 103.Objectives. The aim of the study was to find factors predicting lower extremity amputation in patients with type 1 or type 2 diabetes mellitus through a 7-year follow-up period. Design. Follow-up study. Subjects. Altogether 733 diabetic patients, aged 10±79 years, were drawn from the national drug reimbursement register. Methods. At baseline, the patients underwent a podiatric, circulatory and neurophysiological examination. Seven years later a follow-up study was performed based on clinical and register data. Patient data for those who died during the followup were collected from hospital records and death certificates. All amputations were recorded. The patients with amputation were compared with the other patients and also, in a case-control manner, by taking three nonamputated patients matched by sex, type of diabetes, and age for each patient with amputation.Results. The number of amputations was 25 in the sample. Compared with all patients without amputation, patients with amputation differed in altogether 24 variables concerning diabetes and its complications. Compared with the matched nonamputated patients, the amputated patients had longer duration of diabetes, lower ankle/brachial pressure index (ABI), more often history of retinopathy, nephropathy, and hypertension, more often visual handicap, elevated serum creatinine level, abnormal neurophysiological indices and electrophysiological findings. In the logistic regression analysis, vibration perception threshold, low ABI, history of retinopathy, visual handicap, and male sex were independently associated with lower extremity amputation. Conclusions. Lower extremity amputations were strongly associated with retinopathy, nephropathy, and neuropathy. The presence of any of these complications should lead to intensified actions in order to prevent amputations. As far as arterial circulation is concerned, claudication or absent peripheral pulses were not good predictors of amputation, whereas low ABI, despite its known weaknesses, was a reliable indicator of future amputation.