In a retrospective study we analysed two groups each consisting of 100 consecutive patients of similar age and sex distribution who underwent surgery for carotid disease with an intervening period of 5 years (group A 1980/82, group B 1986/87) between the collectives. Against a background of changing indications, tactics and techniques the aim of the study was to detect any differences between the two groups. Group A had a higher proportion of coronary and peripheral vascular disease. The states of cerebral ischemia I, II and III were distributed equally, but state IV was seen more frequently in group B (p less than 0.05). The number of shunt/without shunt operations in group A was 97/2, in group B 10/84 (p less than 0.005). The external carotid artery was deobliterated in 58/81 cases group A versus group B (p less than 0.005). We closed the artery by direct suture in 8/31 (p less than 0.005), by autologous venous patch in 53/26 (p less than 0.005) and by Dacron patches in 39/41 patients. In group A the operative mortality was zero and in group B 1 patient died; one patient in group B developed sudden occlusion (with TIA) postoperatively. Transient intra-/postoperative neurological deficits occurred in 1/2, permanent in 4/2 patients (n.s.). 54/25 patients have died up to 31/08/91. Coronary heart disease was the main cause of late complications and deaths in group A (p less than 0.025). Statistically, there was no dependence of neurological deficits on group, sex, age or intraoperative management. Only patients with preoperative PRINDS hat a higher postoperative neurological deficit rate than the others.