The Effectiveness of Bariatric Surgery in 607 Predominantly Hispanic and African-American Patients up to 5 Years Post-Operatively

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SAT 660-676-Clinical Obesity Treatment
Clinical
Saturday, June 15, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SAT-668
Tirissa J Reid*1, Saqib Saeed2, Shiranda McCoy2, Amrita Persaud2, Ruchi Bhutani-Kapoor2, Monqidh Al-Sawwaf2 and Leaque Ahmed1
1Columbia University at Harlem Hospital, New York, NY, 2Harlem Hospital, New York, NY
Background:  Bariatric surgery is currently the most effective long-term treatment for morbid obesity (1). Prior studies suggest that weight loss after bariatric surgery may be less pronounced in African-American (AA) (2,3) and Hispanic populations, compared with non-Hispanic Caucasians (4).  We sought to evaluate the effectiveness of bariatric surgery in our predominantly white Hispanic and AA population.

Setting and Methods:  This is a retrospective study of weight changes up to 5 years post-op in 607 predominantly AA and white Hispanic patients who underwent bariatric surgery from 2004-2011 in a New York City hospital.  Height and weight were collected from medical records.  Patients were included if they had height and weight data pre- and post-op at any of the yearly time points.  The primary outcomes were % excess weight loss (% EWL) assessed annually, and the % of patients meeting Reinhold’s criterion for bariatric success at 2 years, having ≥ 50% EWL.  Data are presented as mean ± SE.

Results: 607 patients met inclusion criteria; 92.8% females; mean age was 40.8 ± 0.5 years (range 18.3 – 69.8 yrs). White Hispanics comprised 80.6% of patients and AAs were 17.5%. The baseline BMI was 47.1 ± 0.3 kg/m2 (range 35.1 – 75.4 kg/m2).  Baseline excess weight was significantly higher in males vs. females, 76.8 vs. 56.7kg (p<0.0001), and also higher in AAs vs. Hispanics, 69.6 vs. 55.6 kg (p<0.0001). Gastric bypass was performed in 80.4%, sleeve gastrectomy (SG) in 16.6% and gastric banding in 3.0% of patients.  At 1 yr post-op, there was a mean of 70.9 ± 0.9% EWL [n = 580], 72.0 ± 1.3% EWL at 2 yrs [n = 378], 70.0 ± 1.5% EWL at 3 yrs [n=239], 69.0 ± 2.5% EWL at 4 yrs [n=164], and 64.0 ± 2.2% EWL at 5 yrs [n=98].  Percent EWL was significantly lower in AAs compared with white Hispanics at 1 year post-op, 63.7 ± 2.5% vs. 71.5 ± 0.9% (p=0.0007), respectively.  There was no significant difference in %EWL based on race/ethnicity at years 2-5 post-op.  Percent EWL did not differ significantly between males and females at years 1-5.  There were significant differences in %EWL by type of surgery at 1 yr post-op: RYGB- 72.7 ± 0.9% EWL (n=463, p<0.0001 vs. LAGB & SG); SG- 62.2 ± 2.2% EWL (n=100, p=0.01 vs. LAGB); and LAGB- 44.4 ± 7.0 %EWL (n=18). At 2 yrs post-op, 84.4% of all patients had weight loss success by Reinhold’s criterion of >50% EWL, with more females than males reaching this endpoint, 84.9% (n=298/351) vs. 80.8% (n=21/26), and no difference based on race/ethnicity.

Conclusion:  These data show that bariatric surgery performed in this large cohort of AA and white Hispanic patients resulted in weight loss outcomes within the range of those reported in predominantly non-Hispanic Caucasian cohorts, 64% EWL at 5 years in our group overall vs. 50-74% EWL at 5 yrs in studies of non-Hispanic Caucasians (5,6). Limitations include its retrospective nature, few male patients and loss to follow-up over time.  Prospective studies should be done to confirm these findings.

(1) Fisher BL and Schauer P,  Am J Surg 2002; 184:9S-16S. (2) Admiraal WM et al., Diabetes Care 2012 Sep; 35(9): 1951-8. (3) Madan A et al., Obesity Surgery 2007; 17: 460-464. (4) Nijamkin MP et al., J Acad Nutr Diet 2012 Mar; 112(3): 382-90. (5) Buchwald H et al., JAMA 2004 Oct; 292(14): 1724-1737. (6) Obeid A et al., Surg Endosc 2012; 26:3515-3520.

Nothing to Disclose: TJR, SS, SM, AP, RB, MA, LA

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm