Racial Disparities in the Outcomes of Bariatric Surgery

Racial disparities in surgical outcomes have been identified in multiple fields including bariatric surgery. Obesity and metabolic co-morbidities often affect racial minorities to a greater extent than Caucasian patients; however, the impact of bariatric surgery is often dampened. The purpose of this review is to evaluate possible racial difference in bariatric surgery outcomes. Obesity is a national epidemic which disproportionately affects racial minorities. Bariatric surgery, while safe and effective, has been shown to result in variable outcomes depending on the race of the patient. Non-Hispanic Black patients often have less weight loss, lower resolution of diabetes, and a higher rate of post-operative complications. Socioeconomic status has been theorized to account for the variance seen among races; however, various studies into socioeconomic factors have yielded mixed results. Based on current evidence there remains a racial difference in some, but not all postoperative surgical outcomes. Despite these findings, bariatric surgery remains safe and effective and patients meeting criteria for bariatric surgery should strongly consider the options. However, further investigations are needed to bridge the racial differences in bariatric outcomes.


Introduction
Obesity is a complex, multifactorial disease that results from increased body adiposity and poses a health risk in multiple body systems. There is well-established evidence between obesity and a variety of diseases including the following: cerebrovascular disease, cardiovascular disease, respiratory disease, malignancies, psychiatric disorders, and all-cause mortality [1]. Nationally, the rate of obesity in adults has increased over the previous two decades by 37%, from 30.5% in 2000 to 41.9% in 2020. Severe obesity has increased from 4.7 to 9.2%, over the same period [2•]. While the prevalence of obesity has increased in almost all racial groups, certain racial minorities are disproportionately affected. In a 2020 report from the Centers of Disease Control, Non-Hispanic Black (NHB) persons had the greatest prevalence of obesity (49.9%), followed by Hispanics (45.6%), non-Hispanic Whites (NHW) (41.4%), and non-Hispanic Asians (16.1%) [2 •]. Similarly, studies have shown that minority groups bare a greater burden of obesity related medical issues [3]. Racial difference in obesity rates are currently being evaluated, but likely represent a multifactorial combination of genetic susceptibility [4], socioeconomic status [5], psychosocial factors [6], and environmental factors [7•].
Metabolic bariatric surgery (MBS) is the most effective therapeutic intervention in the treatment of moderate to severe obesity [8]. Numerous trials are have demonstrated superior short-and long-term weight loss following MBS compared to lifestyle modifications and medical management [9,10]. Furthermore, its benefits extend beyond weight loss leading to improvement and resolution of comorbidities, [11][12][13][14] improvement in quality of life, and decreased overall mortality [15]. As the national prevalence of obesity is expected to grow beyond 50% by 2030 [16], the utilization of is expected to similarly increase. Understanding the factors that impact the clinical outcomes of MBS will be necessary to deliver optimal care. A growing body of evidence has suggested the impact of racial and ethnic differences in bariatric surgery outcomes. This article will review the influence of race and ethnicity on clinical outcomes following MBS.

Differences in Postoperative Weight loss
Contemporary studies evaluating racial disparities in MBS outcomes make a compelling case for a sustained difference. Most studies compare NHBs to NHWs, with fewer studies including Hispanic patients. While Hispanic patients represent a significant minority population, a meta-analysis by Zhao [18••]. Similarly, in a 5-year study of 1695 patients undergoing Roux-en-Y gastric bypass (RYGB) NHW patients, on average, had 1.94% greater total weight loss than NHB patients (p < 0.001) [19]. While few studies fail to demonstrate weight loss differences [20,21], numerous clinical trials and meta-analyses find similar weight loss differences over various evaluation times [22][23][24]. Interestingly, Samaan et al. evaluated patient satisfaction following bariatric surgery. At 2 years, NHB patients had 39% excess weight loss compared to 53.4% in NHW patients (p = 0.026). Despite this, rates of self-satisfaction, satisfaction with outcomes, and willingness to redo the operation were equivalent (all p > 0.5) [25]. Weight loss expectations were similar. A proposed explanation for this finding is a difference in body weight perception. In a 2018 study, 70,000 college students were surveyed to estimate their weight. NHB students were found to be more likely to underestimate their body weight compared to NHW and Hispanic students [26]. Additionally, in a separate study, NHB persons demonstrated fewer attempts to lose weight with greater body size relative to NHW and Hispanics [27]. The differences in weight awareness and acceptance may explain the differences in satisfaction rates.
Evaluation of postoperative outcomes in Hispanic patients is limited due to a paucity of data. Studies that include Hispanic patients are often limited by the low representation of this demographic as a percentage of the overall study population. Available data suggests Hispanic patients demonstrate greater weight loss than NHB patients, and equivalent weight loss to NHW patients across all surgical intervention options and both short-and long-term endpoints. In a retrospective study by Khorgami et al.,3268 patients underwent RYGB or adjustable gastric banding with Hispanic patients representing 47.7% of the study population [28]. At 6 months, 1-year, and 2-year intervals NHB patients had less excess weight loss than Hispanic and NHW patients, with no differences between Hispanic and NHW patients. However, the authors noted NHW patients had significant greater weight loss in patients undergoing adjustable gastric banding and body mass index > 40. Contemporary studies corroborate these results [25,[29][30][31].
The basis for racial disparity in postoperative weight loss remains unclear. Socioeconomic factors have been identified as possible contributors. Previous studies in obesogenic environments found that the spatial distribution of fast-food restaurants was concentrated in largely black communities, regardless of income [32]. Few studies attain the granularity to discern socioeconomic differences, however Wood et al. found socioeconomic status was not a significant factor for worse weight loss outcomes [18••]. Conversely, in a retrospective review of a Veterans Affairs hospital in patients undergoing RYGB, there was significantly lower 10-year weight loss in patients with low income compared to low-mid level income [33]. Additional theories point to biological determinants which may account for the observed differences. Studies have demonstrated racial differences in postprandial levels of incretins, such as glucagon-like peptide 1 and ghrelin [34,35]. Additionally, total and resting energy use has been found to be decreased in NHB patients following weight loss [36].

Diabetes Mellitus
Being overweight or obese is associated with increased incidence of many associated medical problems including the following: diabetes mellitus type 2 (DM2), hypertension, obstructive sleep apnea, and hyperlipidemia. Bariatric surgery has been shown to be an effective and lasting treatment; however, evaluation of racial differences is generally focused on DM2 and hypertension. Contemporary data shows there is limited racial and ethnic differences in co-morbidity resolution. In a retrospective analysis, Valencia et al. examined NHB, NHW, and Hispanic patients over 12 months. At 6 months, all groups had significant improvement in fasting insulin, fasting glucose, and hemoglobin A1c levels. Between 6 and 12 months, contrary to NHW and Hispanic patients, NHB patients did not have continued improved glucose metabolism, and significantly increased HgbA1c from 5.9 to 6.2%, p = 0.032 [37]. This suggests that NHB patients may benefit from additional postoperative support in maintaining appropriate glycemic control. In a retrospective review using the Longitudinal Assessment of Bariatric Surgery (LABS-2) database, Turner et al. evaluated ~ 1700 patients undergoing RYGB with 5 years of follow up examining co-morbidity resolution. On initial analysis, NHW patients were 70% more likely to have resolution of diabetes postoperatively. However, after accounting for severity of disease utilizing baseline C-peptide levels and use of insulin, racial disparity was eliminated [19]. These findings are consistent among other studies [

Hypertension
Various studies show mixed results on racial differences in hypertension resolution. Turner et al., using LABS 2 data, showed that 5 years after RYGB there was no difference between black and white patients (RR 1.04, p = 0.857); however, when compared to other races, white patients had significant greater rates of hypertension resolution (RR 0.67, p < 0.001) [19]. These results remained despite accounting for severity of disease. Similarly, in a retrospective review of the Bariatric Outcomes Longitudinal Database, Nunez et al. evaluated over 1500 adolescents undergoing bariatric surgery and found no difference in hypertension resolution at one year when account for disease severity [31]. Other studies substantiate limited racial differences in hypertension resolution [18 ••, 38].

Obstructive Sleep Apnea and Dyslipidemia
Few studies have evaluated racial differences in postoperative resolution of obstructive sleep apnea (OSA) and dyslipidemia. Wood noted greater rates of sleep apnea resolution in NHB patients compared to NHW patients, 62.6% vs 56.1% (p < 0.005). In contrast, three recent studies all evaluated resolution of obstructive sleep apnea with no racial difference in rates of OSA resolution [19,31,38]. Few studies, examined race on the impact of dyslipidemia resolution; however, 4 recent studies found no difference in resolution between races [18••, 19, 31, 40].

Safety of Bariatric Surgery
Bariatric surgery has undergone a substantial evolution since its inception. The vertical banded gastroplasty, adjustable gastric banding, and jejunoileal bypass have been generally abandoned due to intolerable side effects, high rates of reoperation, and poor long-term efficiency [41]. The sleeve gastrectomy and RYGB are currently the most performed procedures with complication rates below 5% [42], rates of major complications below 1% [43], and short-and longterm mortality rates ranging between 0.06 and 0.23% [44]. However, studies evaluating racial differences in postoperative complications and mortality have yielded varied results [44,45].

Mortality
Data on racial disparities in postoperative mortality are mixed. Several studies have noted modestly increased rates of mortality in NHB patients following sleeve gastrectomy. Most of the studies supporting racial disparity are retrospective reviews of the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database, which allows for evaluation of 30-days postoperative outcomes. For example, using the MBSAQIP, Sheka et al. found a significant difference in 30-days mortality in NHB (0.2%) compared to NHW patients (0.1%), (p < 0.001) undergoing LSG [46], there was no difference in patients undergoing RYGB. These findings are supported by other short-term database studies with NHB mortality rates ranging (0.09-0.21%) compared to NHW (0.04-0.2%) [47][48][49]. While the cause of mortality is not consistently identified, the authors cite greater rates of pulmonary embolism, myocardial infarction, and cardiac arrest in NHB patients. Conversely, multiple single and multi-institution studies do not show a significant racial disparity in LSG or RYGB in short-or long-term evaluations [18 ••, 50, 51]. Weiss et al. utilized the California Longitudinal database and evaluated over 129,000 patients undergoing RYGB and found no difference in mortality at 10 years between NHB, NHW, and Hispanic patients [52]. Given the available data it is difficult to determine true racial differences in mortality following MBS, and further investigations are warranted.

Postoperative Complications
Postoperative complications account for a wide range of problems ranging from mild to severe and vary based on short-and long-term time points. Over the previous decade there are varied reports regarding the observation of racial difference in postoperative complications, but recent data supports that NHB patients have greater postoperative complications compared to NHW patients. Wood et al. noted at 30-days after adjusting for confounders that NHB patient had greater rates of any complication and serious complications compared to NHW patients, 8.8 vs 6.8% and 2.5 and 1.9% (p < 0.05), respectively [18]. These findings were sustained after adjusting for patients' characteristics, surgeon, and hospital. Similar findings were noted in evaluation of the MBSAQIP database. Welsh et al. found that NHB patients had significantly greater rates of Claviden-Dindo (CD) grade 1, 3, and 5 complications, and Hispanic patients had greater risk of CD grade 3 complication compared to NHW patients [49]. Similarly, Acevedo et al. found NHB patients undergoing robotic LSG to have greater rates of postoperative complications, but there was no difference between NHW and Hispanic patients [50].
Conversely, multiple studies fail to identify racial differences. In a recent MBSAQIP study evaluating elderly patients undergoing MBS, Edwards et al. studied over 5000 patients in propensity matched cohorts and found no differences in postoperative leak rats, bleeding, cardiovascular, pulmonary, renal, thromboembolic, or infections complications (all p > 0.05) [53]. Interestingly, Weiss et al. found that at 10 years, NHB patients had a reduced hazard ratio of 0.8 relative to NHW patients, and no difference in complications between Hispanic patients and NHW patients [52]. Other studies have found similar results.

Length of Stay and Readmission
Unlike other postoperative complications there appears to be a consistent racial disparity in postoperative length of stay, and rates of readmission. Numerous studies have found NHB patients tend to have longer hospitalizations for both LSG and RYGB [18••, 46, 47, 51], with an average range of 1.8-5 days, compared to NHW patients 1.

7-4.3 days [54•].
Recently, Westerick et al. evaluated the Florida Agency for Healthcare Administrative data for patients undergoing bariatric surgery found that Hispanic had significant greater adjusted odds for prolonged hospitalization (aOR:1.27, p < 0.05) compared to NHW patients [55]. Interestingly, patients with state or federal insurance had lower rates of prolonged hospitalization compared to patients with commercial insurance, suggesting limited influence of socioeconomic status on postoperative outcomes.
A majority of national database, single and multi-institutional studies, and meta-analyses support racial differences in readmissions. In a meta-analysis by Stone et al. readmission rates for NHB patients ranged between 3 and 10.2% compared to 2.3-3.6% in NHW patients, and were observed for patients undergoing both LSG and RYGB [54•]. Similarly, utilizing the MBSAQIP database Edwards et al., NHB patients and Hispanic patients had greater adjusted odds of readmission compared to NHW patients, 1.55 and 1.11 (p < 0.001), respectively [53]. Analysis of causes of readmissions showed higher rates of anastomotic leaks, renal insufficiency, and cardiac problems in NHB patients, while Hispanic patients had greater rates of venous thromboembolism. Some authors have hypothesized that improvements in patient communication may facilitate reduction in readmissions. Wong et al. demonstrated that racial differences between health providers and patients presents a barrier to effective communications and understanding of postoperative instructions in NHB and Hispanic patients, which may contribute to higher rates of readmission [56]. Additionally, Wood et al. found that adverse events and resource utilization decreased with increasing income status, with racial difference being most profound at lower income levels. Previous studies support that some of the variance in postoperative discharge resource utilization may be accounted for by differences in social support, insurance, transport availability, and access to primary care [18••].
Lastly, current literature underscores the significance of the enhanced recovery after surgery (ERAS) protocol to improve and standardize perioperative outcomes. Since the publication of ERAS guidelines in 2016 [57] there remains a lack of large, randomized trial validating these findings. In a recent study, Trotta et al. examined bariatric surgery patients undergoing LSG and RYGB with conventional perioperative care compared to ERAS. They demonstrated reductions is length of hospital stay from 4.7 to 2.1 days (p < 0.05) [58]. Furthermore, studies in patients undergoing colorectal surgery utilizing ERAS demonstrated elimination of racial disparity in length of stay [59 •]. Similarly, a meta-analysis of ERAS use in bariatric surgery found reductions in length of stay and a tendency toward reduced readmissions [60]. More aggressive use of ERAS programs in bariatric surgery may be added to facilitate reduction in racial disparities in length of stay and readmissions.

Rates of Reoperation
Studies evaluating rates of reoperation report mixed results. Using the MBSAQIP database, Hui et al. evaluated racial differences in patients undergoing LSG and RYGB in propensity score matched groups. African American patients undergoing RYGB demonstrated increased rates of reoperation compared to non-African American patients (2.55 vs 2.01%, p < 0.05), but no difference was found in patients undergoing LSG [48]. Interestingly, utilizing the same database, Sheka et al. demonstrated increased rates of reoperation in NHB patients undergoing LSG compared to white patients (1.1 vs 1.0%, p = 0.006), but this was not seen in the RYGB cohort [46]. Moreover, in a recent study utilizing the MBSAQIP database, Edwards et al. showed no difference in rates of reoperation in propensity matched cohorts with previous abdominal organ transplants [53]. Similar studies fail to identify a significant racial disparity in rates of reoperations [18••, 49]. The MBSAQIP database is a powerful source of data with standardized variables, greater than 95% follow up, and more than 600 contributing centers [61]. However, due to the available data it is difficult to attain a deep level of granularity. With current evidence there does not appear to be a racial disparity in rates of reoperation.

Conclusion
Metabolic and bariatric surgery is safe, effective, and its utilization will continue to grow as obesity is recognized as a metabolic and hormonal disease. This highlights the need to optimize postoperative weight loss and co-morbidity management, while mitigating complications and co-morbidity relapses. Racial disparities were observed in overall postoperative complications, length of stay, and readmissions. Strategies to reduce complications such as implementation of an enhanced recovery pathway, and effective communication between patients and providers must be further investigated. Additionally, the overall trend suggests reduced weight loss in Non-Hispanic Black patients compared to other races. Several suggested theories imply differences in weight perception, and biological determinants as possibly causative factors; however, evidence to support these hypotheses is limited. Socioeconomic status is a variable that is often raised to account for unexplained variance, but there are mixed results as to its impact. There was no substantial difference in the resolution of obesity associated medical problems. However, across all races, patients undergoing RYGB had decreased recurrence of metabolic co-morbidities. Overall, while there may be racial differences in the postoperative outcomes, patients of all races that meet criteria will benefit from metabolic bariatric surgery, and those with significant metabolic co-morbidities should consider Roux-en-Y gastric bypass to minimize risk of co-morbidity relapse.
Author contributions PW and KS: jointly wrote and reviewed the main manuscript text.
Funding Open access funding provided by SCELC, Statewide California Electronic Library Consortium. There was no funding received for this project.

Conflict of interest
The authors declare no conflicts of interest with respect to the authorship or publication of this manuscript.

Research Involving Human and Animal Participants
This article does not contain any studies with human or animal subjects performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.