There has been controversy regarding the relationship between Helicobacter pylori and perforated peptic ulcer, which is known to have a high recurrence rate if only simple patch repair is performed. The aim of this study was to evaluate the association between H. pylori infection and intake of non-steroidal anti-inflammatory drugs (NSAIDs) in patients with perforated duodenal ulcers. Of the 73 patients recruited over a 16-month period, 51 (70 per cent) had evidence of H. pylori infection by intraoperative gastroscopy and antral biopsies. The infection rate rose to 80 per cent if NSAID users were excluded. The H. pylori-infected group was significantly younger (mean 47.6 versus 62.5 years), with a male preponderance (49 of 51 versus 14 of 22 patients), and had significantly less NSAID consumption (three of 51 versus ten of 22) and more prolonged dyspepsia (40 of 51 versus ten of 22), compared with H. pylori-negative patients. H. pylori infection probably plays an important role in the causation of non-NSAID-induced duodenal ulcer perforation. Whether eradication of the bacteria can alleviate the strong ulcer diathesis in this subgroup of patients is unknown.