Aldose reductase activity can be measured in the neutrophil and it has been proposed that this may be a marker for risk of complications in diabetes. We have studied aldose reductase activity in neutrophil, nerve, and lens in diabetic patients undergoing sural nerve biopsy or cataract extraction. A correlation was demonstrated between lens and neutrophil aldose reductase activity (r = 0.53, p = 0.01) but no correlations were demonstrated between nerve aldose reductase activities and nerve morphometry, nerve function or neutrophil aldose reductase activity. No significant difference was found between neutrophil aldose reductase activities in groups of patients with severe neuropathy, or cataract, or no complications (24 (interquartile range 16-32) vs 24 (16-40) vs 24 (16-40) nmol NADPH min-1 10(8)-cells-1). In a group of 56 Type 1 diabetic patients screened within 6 years of diagnosis, multiple regression analysis failed to show any relationship between neutrophil aldose reductase activity and abnormalities of neurophysiological function. These results suggest that neutrophil aldose reductase activity cannot be used as a marker for the development of cataract or neuropathy in diabetes.