The symptomatic effect of pancreatic enzyme substitution therapy was examined in a 4-week double-blind crossover study. Twenty patients, 11 with and 9 without steatorrhoea, were examined. Pancreatic steatorrhoea was reduced from a median of 24 g/day to 10 g/day by the enzyme therapy (P less than 0.01). No significant pain reduction was found in either of the two groups, although there was a tendency to reduction in pain and analgetic consumption in the patients with steatorrhoea when treated with pancreatic enzymes. It is concluded that pancreatic steatorrhoea is still the only indication for pancreatic enzyme therapy.