Background: To perform a prospective, randomized comparison of laparoscopic adjustable gastric banding (LAGB) and laparoscopic Roux-en-Y gastric bypass (LRYGB).
Methods: LAGB, using the pars flaccida technique, and standard LRYGB were performed. From January 2000 to November 2000, 51 patients (mean age 34.0 +/- 8.9 years, range 20-49) were randomly allocated to undergo either LAGB (n = 27, 5 men and 22 women, mean age 33.3 years, mean weight 120 kg, mean body mass index [BMI] 43.4 kg/m(2); percentage of excess weight loss 83.8%) or LRYGB (n = 24, 4 men and 20 women, mean age 34.7, mean weight 120 kg, mean BMI 43.8 kg/m(2), percentage of excess weight loss 83.3). Data on the operative time, complications, reoperations with hospital stay, weight, BMI, percentage of excess weight loss, and co-morbidities were collected yearly. Failure was considered a BMI of >35 at 5 years postoperatively. The data were analyzed using Student's t test and Fisher's exact test, with P <.05 considered significant.
Results: The mean operative time was 60 +/- 20 minutes for the LAGB group and 220 +/- 100 minutes for the LRYGB group (P <.001). One patient in the LAGB group was lost to follow-up. No patient died. Conversion to laparotomy was performed in 1 (4.2%) of 24 LRYGB patients because of a posterior leak of the gastrojejunal anastomosis. Reoperations were required in 4 (15.2%) of 26 LAGB patients, 2 because of gastric pouch dilation and 2 because of unsatisfactory weight loss. One of these patients required conversion to biliopancreatic diversion; the remaining 3 patients were on the waiting list for LRYGB. Reoperations were required in 3 (12.5%) of the 24 LRYGB patients, and each was because of a potentially lethal complication. No LAGB patient required reoperation because of an early complication. Of the 27 LAGB patients, 3 had hypertension and 1 had sleep apnea. Of the 24 LRYGB patients, 2 had hyperlipemia, 1 had hypertension, and 1 had type 2 diabetes. Five years after surgery, the diabetes, sleep apnea, and hyperlipemia had resolved. At the 5-year (range 60-66 months) follow-up visit, the LRYGB patients had significantly lower weight and BMI and a greater percentage of excess weight loss than did the LAGB patients. Weight loss failure (BMI >35 kg/m(2) at 5 yr) was observed in 9 (34.6%) of 26 LAGB patients and in 1 (4.2%) of 24 LRYGB patients (P <.001). Of the 26 patients in the LAGB group and 24 in the LRYGB group, 3 (11.5%) and 15 (62.5%) had a BMI of <30 kg/m(2), respectively (P <.001).
Conclusion: The results of our study have shown that LRYGB results in better weight loss and a reduced number of failures compared with LAGB, despite the significantly longer operative time and life-threatening complications.