BACKGROUND: The value of prophylactic gastroenterostomy (usually combined with a biliary bypass) in patients with unresectable cancer of the pancreatic head is controversial. METHODS: A systematic review of retrospective and prospective studies, and a meta-analysis of prospective studies, on the use of prophylactic gastroenterostomy for unresectable pancreatic cancer were performed. RESULTS: Analysis of retrospective studies did not reveal any advantage or disadvantage of prophylactic gastroenterostomy. Three prospective studies comparing prophylactic gastroenterostomy plus biliodigestive anastomosis with no bypass or a biliodigestive anastomosis alone were identified (altogether 218 patients). For patients who had prophylactic gastroenterostomy, the chance of gastric outlet obstruction during follow-up was significantly lower (odds ratio (OR) 0.06 (95 per cent confidence interval (c.i.) 0.02 to 0.21); P < 0.001). The rates of postoperative delayed gastric emptying were similar in both groups (OR 1.93 (95 per cent c.i. 0.57 to 6.53); P = 0.290), as were morbidity and mortality. The estimated duration of hospital stay after prophylactic gastroenterostomy was 3 days longer than for patients without bypass (weighted mean difference 3.1 (95 per cent c.i. 0.7 to 5.5); P = 0.010). CONCLUSION: Prophylactic gastroenterostomy should be performed during surgical exploration of patients with unresectable pancreatic head tumours because it reduces the incidence of long-term gastroduodenal obstruction without impairing short-term outcome.