A case-control study was carried out to analyse leucocyte counts in relation to proteinuria and smoking in type 1 (insulin-dependent) diabetes. The subjects were 180 smoking (87 women, age 32 +/- 11 years, diabetes duration 14 +/- 6 years) and 188 nonsmoking (88 women, age 32 +/- 11 years, diabetes duration, 14 +/- 6 years) type 1 diabetic patients. Leucocyte counts were higher in smokers than in nonsmokers, both in women (7.6 +/- 2.3 vs 6.8 +/- 1.9 x 10(9)/l, P < 0.01) and in men (8.1 +/- 2.7 vs 6.4 +/- 2.0 x 10(9)/l, P < 0.0001). Leucocyte counts correlated with the number of cigarettes smoked per day (r = 0.32, P < 0.0001), but were unrelated to glycosylated haemoglobin levels. Among nonsmokers, leucocyte counts were comparable between patients with no mal proteinuria (6.7 +/- 2.2 x 10(9)/l, n = 106), microproteinuria (6.3 +/- 1.7 x 10(9)/l, n = 66) and macroproteinuria (6.8 +/- 1.1 x 10(9)/l, n = 16). Among smokers, patients with macroproteinuria or microproteinuria had higher leucocyte counts than those with normal proteinuria [8.9 + 3.0 x 10(9)/l, (n = 36) vs 8.4 + 2.7 x 10(9)/l (n = 61) vs 7.0 + 1.9 x 10(9)/l (n = 83); P < 0.0001)], a finding which was not due to differences in the number of cigarettes smoked per day. It is concluded that in these type 1 diabetic patients leucocyte counts were higher in smokers than in nonsmokers. Among nonsmokers leucocyte counts were comparable between patients with normal proteinuria and increased proteinuria, whereas among smokers leucocyte counts were higher in patients with increased proteinuria. It appears it would be worthwhile examining in prospective studies whether leucocyte counts can contribute to identifying those type 1 diabetic patients with a particularly high risk of cardiovascular complications.