Background: Postoperative pancreatic fistula (POPF) remains one of the most common causes of morbidity following pancreaticoduodenectomy (PD). This randomized trial examined whether external stent drainage of the pancreatic duct decreases the rate of POPF after PD and subsequent pancreaticojejunostomy (PJ). Methods: Consecutive patients who underwent PD with subsequent construction of a duct‐to‐mucosa PJ were randomized into a stented and a non‐stented group. The primary outcome was the incidence of clinically relevant POPF. Secondary outcomes were morbidity and mortality rates, and hospital stay. Results: Of 114 PD procedures, 93 were suitable for inclusion in the study after informed consent. The rate of clinically relevant POPF was significantly lower in the stented group than in the non‐stented group: three of 47 (6 per cent) versus ten of 46 (22 per cent) (P = 0·040). Among patients with a dilated duct, rates of POPF were similar in both groups. Among patients with a non‐dilated duct, clinically relevant POPF was significantly less common in the stented group than in the non‐stented group: two of 21 (10 per cent) versus eight of 20 (40 per cent) (P = 0·033). No significant differences in morbidity or mortality were observed. Univariable analysis identified body mass index (BMI), pancreatic cancer, pancreatic texture, pancreatic duct size and duct stenting as risk factors related to clinically relevant POPF. Multivariable analysis taking these five factors into account identified high BMI (risk ratio (RR) 11·45; P = 0·008), non‐dilated duct (RR 5·33; P = 0·046) and no stent (RR 10·38; P = 0·004) as significant risk factors. Conclusion: External duct stenting reduced the risk of clinically relevant POPF after PD and subsequent duct‐to‐mucosa PJ. Registration number: UMIN000000952 (http://www.umin.ac.jp/ctr/index‐j.htm). Copyright © 2012 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. To stent or not to stent