Introduction: White patients undergo bariatric surgery more frequently and with fewer complications compared to non-white patients. Yet, racial disparities in referral pattern for preoperative cardiovascular evaluation (CVE) are unexplored. We hypothesized that racial disparities exist in preoperative CVE in white vs. non-white patients undergoing bariatric surgery. Methods: We identified consecutive patients undergoing bariatric surgery at our institution from 2014-2018. The association of white vs. non-white race with referral for CVE, cardiac testing, surgical wait time, and postoperative weight loss was determined in a multivariate model adjusting for baseline demographic comorbidities and income. Additional adjustment for insurance provider was made for surgical wait time. Results: Of the 797 patients undergoing bariatric surgery during the study period, 86% (n=682) were non-white. White vs. non-white patients had similar age, gender distribution, and prevalence of hypertension, hyperlipidemia, Type 2 diabetes, and heart failure. A similar proportion of whites vs. non-whites were referred to cardiology for pre-operative evaluation (65% vs. 68%, p=0.52), while whites were more likely to undergo further preoperative cardiac testing after adjustment (68% vs. 55%, p=0.04; adjusted, OR 0.6, 95%CI 0.35-1.0, p=0.05). Despite less cardiac testing, non-white patients referred to cardiology waited longer for surgery (9.9 months vs. 12.1 months, p=0.008), including after adjustment (HR 0.7, 95%CI 0.53-0.91, p= 0.008. Weight loss at 6 months was higher in whites (12.9 kg/m 2 vs. 11.95 kg/m 2 , p=0.03), but equivalent at 1 year (whites 14.9 kg/m 2 vs. non-whites 14.3 kg/m 2 , p=0.33). Conclusions: In a bariatric surgery population, white vs. non-white patients were referred for CVE in similar proportion. However, white patients underwent more subsequent cardiac testing pre-operatively when compared to non-whites yet had a shorter wait time for surgery. Early weight loss was greater in whites, but equivalent between groups at 1 year.
Read full abstract