BACKGROUND Laparoscopic gastric bypass is the 'gold standard' for treatment of morbidly obese patients in many centres. There is debate regarding the optimal length for small bowel limbs. This study aimed to determine whether the proximal or distal approach is better. METHODS Twenty-five patients undergoing primary distal gastric bypass in 2000-2002 were randomly matched for age, sex and preoperative body mass index (BMI) with 25 patients having a primary proximal bypass. All distal operations were performed laparoscopically; one proximal procedure was converted to open surgery. RESULTS Mean operating time was 170 min for proximal and 242 min for distal bypasses (P = 0.004); median hospital stay was similar in the two groups. There were no deaths and the overall complication rate was similar, as was weight loss at 4 years: BMI decreased from 45.9 to 31.7 kg/m2 for the proximal and from 45.8 to 33.1 kg/m2 for the distal approach. Co-morbidities decreased after surgery in both groups; the prevalence of diabetes, arterial hypertension and dyslipidaemia at all time points was similar in the two groups. CONCLUSIONS Proximal and distal laparoscopic gastric bypass operations are feasible and safe, with no differences in weight loss or reduction of co-morbidity in unselected morbidly obese patients.