Migration of the peritoneal dialysis (PD) catheter from the pelvis to the upper abdomen frequently results in peritoneal dialysis failure and removal of the catheter. Previous studies compared PD catheter survival in various catheter configurations. These studies included single-cuff and double-cuff, straight-end and curled-end catheters, and showed an incidence of catheter migration ranging from 5%-35% depending on the catheter type. Recent studies demonstrated that the double-cuff, Swan-neck, curled-end configuration is associated with a considerably lower incidence of migration. Most of these studies, however, had a small patient sample or no control group, or they compared nonequivalent catheters (for example, Swan-neck, curled-end versus straight, non-curled-end). Over a six-year period, we examined two similar double-cuff, curled-end catheters: Group I catheters had a straight segment between the two cuffs, and Group II catheters had a 60 degrees Swan-neck bend between the two cuffs (Quinton Instrument Co., Bothell, WA, U.S.A.). The two catheters were identical, except for the presence or absence of the Swan-neck bend. All catheters were placed by the closed laparoscopic technique. In the two groups of patients in whom the catheters were implanted, no statistically significant difference was observed in primary disease, age, sex, race, weight, prior abdominal surgery, or duration of dialysis before catheter migration. In group I, 33 of the 219 patients developed catheter migration (15%); in group II, 2 patients of 243 patients developed catheter migration (less than 1%, p = 0.002). In conclusion, the Swan-neck configuration presents an independent factor in preventing PD catheter migration. Review of previous studies and the data from our study, show that double-cuff, curled-end, Swan-neck PD catheters are superior to other catheters in regard to prevention of catheter migration and should be the catheter of choice in PD patients.