One hundred and thirteen consecutive patients admitted with a perforated duodenal ulcer over a 5-year period (1978-82) and treated by simple closure have been followed prospectively over a median period of 43 months. Patients were divided into two categories according to their previous history of dyspepsia; group 1 (66 patients) with a chronic history of more than 3 months and group 2 (47 patients) where there was no history of dyspepsia or a history of less than 3 months duration. The follow-up periods were similar (group 1, 44 months; group 2, 43 months). The overall recurrence rate was 42 per cent and to date only 14 per cent of the total group of 113 patients have required a definitive operation. The incidence of ulcer recurrence was higher in group 1 than in group 2 in terms of total recurrence (group 1, 50 per cent; group 2, 32 per cent) and patients requiring further surgery (group 1, n = 14, 21 per cent; group 2, n = 7, 15 per cent). Five of these 21 patients required an emergency procedure for haemorrhage or reperforation (group 1, n = 2; group 2, n = 3). There were no significant differences between the 5-year predictive recurrence rate or the requirement for definitive surgery between the groups. Five of the 27 patients currently on medical treatment have required maintenance treatment while the remaining patients receive active treatment for a mean of 20 per cent of the time since they developed recurrent ulceration. These results support continuation of our 'wait and see' policy following simple closure of perforated duodenal ulcer, even in patients with a history of chronic dyspepsia.