Even though severe obesity is disproportionate by racial and ethnic group in the U.S., there’s limited evidence about whether metabolic and bariatric surgery outcomes differ by race/ethnicity.
In a large, multicenter comparative effectiveness study, Caroline M. Apovian, MD, co-director of the Center for Weight Management and Wellness at Brigham and Women’s Hospital, and colleagues determined the variability of weight loss and hemoglobin A1C (HbA1C) across racial/ethnic groups was small after Roux-en-Y gastric bypass (RYGB) as compared with sleeve gastrectomy (SG). However, safety concerns were greater for Black and Latino/a/x patients after RYGB than after SG, as they report in JAMA Surgery.
Methods
In 2016, the National Patient-Centered Clinical Research Network (PCORnet) funded the PCORnet Bariatric Study (PBS), a retrospective, observational cohort study designed to compare the safety and effectiveness of RYGB and SG in 11 geographically diverse partner networks.
The PBS involved 36,871 adults and adolescents (12–79 years old; mean age, 45; 81% female) who underwent primary RYGB or SG between January 1, 2005, and September 30, 2015, at 41 healthcare systems. Patients self-identified as Black (19%), Latino/a/x (24%), white (53%), or other (2%). Race was unrecorded for 2%.
25 sites agreed to participate in the current analysis. Three cohorts were created:
- All 36,871 patients were analyzed for weight loss at one, three, and five years after surgery
- 8,407 were analyzed for remission of type 2 diabetes (T2D; first postsurgical occurrence of HbA1C <6.5% after at least six months without diabetes medication), T2D relapse (HbA1C level ≥6.5% and/or a prescription for diabetes medication after remission) and change in HbA1C level at one, three and five years after surgery
- 32,853 were analyzed for safety and utilization; the primary outcome of interest was time to the first occurrence of a gastrointestinal operation or intervention across five years of follow-up
Weight Loss
Weight loss was better for RYGB than SG in all years for Black, Latino/a/x, and white patients. At year five, the difference in mean percentage total weight loss between RYGB and SG was (P<0.001 for all comparisons):
- Black: −7.6%
- Latino/a/x: −6.2%
- White: −5.9%
- Other races: −6.0%
T2D Remission, Relapse and Change in HbA1C
Rates of T2D remission were similar with RYGB and SG in all racial/ethnic groups, except that Latino/a/x patients had higher rates with RYGB (HR, 1.19; P=0.002).
All racial/ethnic groups experienced lower rates of T2D relapse with RYGB than with SG:
- Black—HR, 0.73 (P=0.02)
- Latino/a/x—HR, 0.75 (P=0.01)
- White—HR, 0.76 (P=0.006)
- Other races—HR, 0.56 (P=0.04)
For all years studied, differences in mean HbA1C level were significantly larger for RYGB than SG for Black, Latino/a/x, and white patients.
Safety and Utilization
The three largest racial and ethnic groups had a significantly higher risk of gastrointestinal intervention or operation after RYGB than SG. All three groups also had a significantly higher risk of endoscopy with RYGB than SG.
Only for Black and Latino/a/x patients was the risk of hospitalization significantly higher after RYGB than after SG:
- Latino/a/x—HR, 1.37 (P<0.001)
- Black—HR, 1.30 (P<0.001)
- White—HR, 1.05 (P=0.15)
Of particular concern, Latino patients who had RYGB were at higher risk of all-cause mortality and had higher odds of a 30-day major adverse event than those who had SG:
- Mortality—HR, 2.41 (P=0.01)
- Major adverse event—adjusted OR, 1.92 (P<0.001)
Those two risks were not increased for any other racial group.
Takeaway Message
These results suggest that shared decision-making conversations about the choice of bariatric operation should be tailored to the race/ethnicity of the patient.