Previous Weight Loss As a Predictor of Weight Loss Outcomes After Laparoscopic Adjustable Gastric Banding

Monica Sethi, MD, Melissa Beitner, MBBS, Melissa Magrath, BA, Bradley F Schwack, MD, Marina S Kurian, MD, George A Fielding, MD, Christine J Ren-Fielding, MD. New York University School of Medicine

Introduction: Weight loss after laparoscopic adjustable gastric banding (LAGB) can be influenced by a variety of factors. The primary objective of this study is to investigate whether the maximum amount of previous weight loss with diet and exercise, prior to evaluation for bariatric surgery, is predictive of postoperative weight loss success among primary laparoscopic adjustable gastric banding patients.

Methods: A retrospective cohort study was designed from a prospectively-collected database at a single institution. There were 999 patients who underwent primary LAGB between June 2010 and December 2011 who were considered for this study. Inclusion criteria consisted of age ≥ 18 years, initial body mass index (BMI) ≥35 kg/m2, intake information on the maximum weight loss at any time prior to referral to our bariatric practice, and at least 2 years of postoperative follow-up. Patients with prior bariatric surgery and patients without a recorded maximum previous weight loss were excluded. Data obtained included preoperative demographics, BMI, comorbidities, maximum amount of previous weight loss, and number of previous weight loss attempts, as well as postoperative weight (lbs), BMI, and percent excess weight loss (%EWL), at 12, 18, and 24 months postoperatively. The primary outcomes were mean %EWL, percent that achieve weight loss success (%EWL ≥40), and percent with suboptimal weight loss (%EWL <20) at 2 years post-LAGB. Statistical analysis was performed using SAS, with tests including chi-square, ANOVA, linear and logistic regression.

Results: Of the 999 patients, 462 primary LAGB patients met criteria. Mean previous weight loss was 29.7 lbs (SD=27.6, range=[0,175]). These patients were divided into four previous weight loss groups (0 lbs, 1-20 lbs, 21-50 lbs, >50 lbs) for analysis. Baseline patient characteristics between the four groups demonstrated that patients with the greatest previous weight loss (>50 lbs) were more likely to be male (p=0.01) and had higher baseline weight/BMI (p<0.0001). There were no differences in comorbidity or operative time between the groups. In order to account for the baseline differences between the four previous weight loss groups, an adjusted model was used for analysis, controlling for age, gender, baseline BMI, and band type. In the adjusted model, those with >50 lbs of previous weight loss had 8.8% EWL, 6.7% EWL, and 5.0% EWL more than those with 0, 1-20, and 21-50 lbs of previous weight loss, respectively (p<0.0001) [Table 1]. Patients with >50 lbs of previous weight loss were more likely to achieve weight loss success (>40% EWL) at 2 years post-LAGB (p=0.047), and were less likely to have suboptimal weight loss (<20% EWL) at 2 years post-LAGB (p=0.027).

Conclusion: Previous weight loss is a significant predictor of weight loss after LAGB. With multiple options for weight loss surgery, this study helps elucidate which patients may be more likely to achieve greater weight loss with the LAGB, allowing clinicians to appropriately counsel patients preoperatively.


« Return to SAGES 2015 abstract archive