The influence of cigarette smoking on the healing of duodenal ulcers remains controversial. The aim of this study was to determine the influence of smoking and other risk factors on short-term healing of duodenal ulcers. Patients (n = 1923) with an endoscopically diagnosed recurrence of a duodenal ulcer of greater-than-or-equal-to 5 mm diameter were treated with ranitidine 300 mg at night. Healing was assessed at 2, 4 and 8 weeks by endoscopy. After 2 weeks, 45.3% of nonsmokers were healed compared with 41.5% of ex-smokers and 36.1% of smokers, the difference between smokers and nonsmokers being significant. There was a weak inverse relationship between healing rates and the number of cigarettes smoked at 2 weeks which did not reach statistical significance. Neither stopping smoking during healing, nor a decrease in the amount of cigarettes smoked improved healing rates at 2 weeks. At 4 weeks, however, healing was significantly more frequent in patients who stopped smoking, compared with those who continued. In a model of a logistic regression analysis including the factors smoking, workload, stress, previous healing and non-steroidal anti-inflammatory drug (NSAID) intake, only smoking was significantly associated with slow healing. These results show that smoking slows healing of duodenal ulcers during treatment with ranitidine. The negative influence of smoking on the ulcer healing process seems to be related to the chronic rather than the acute effects of smoking on the gastroduodenal physiology.