AIM: To investigate whether or not bariatric surgery weight outcomes vary by ethnicity in a large, nationally representative sample of adolescents.

METHODS: The Bariatric Outcomes Longitudinal Database was used for analysis and contains data on surgeries performed on adolescents from 2004 to 2010 from 423 surgeons at 360 facilities across the United States Adolescents (n = 827) between 11 and 19 years old who underwent either gastric bypass or adjustable gastric banding surgery were included in the analysis. Outcome measures included changes in anthropometric measurements [weight (kg) and body mass index] from baseline to 3 (n = 739), 6 (n = 512), and 12 (n = 247) mo after surgery.

RESULTS: A year after patients underwent either gastric bypass (51%) or adjustable gastric banding (49%) surgery, mean estimated weight loss for all ethnic groups differed by a maximum of only 1.5 kg, being 34.3 kg (95%CI: 30.0-38.5 kg) for Hispanics, 33.8 kg (95%CI: 27.3-40.3 kg) for non-Hispanic blacks, and 32.8 kg (95%CI: 30.9-34.7 kg) for non-Hispanic whites. No overall pairwise group comparisons were significant, indicating that no ethnic group had better weight loss outcomes than did another.

CONCLUSION: Bariatric surgery substantially reduces the weight of severely obese adolescents at 1 year post-procedure with little variation by ethnicity and/or gender. These results suggest that bariatric surgery is a safe and reasonable treatment for all severely obese adolescents with the appropriate indications.

The Centers for Disease Control and Prevention (CDC) recently reported that the prevalence of childhood obesity in the United States has stabilized over the past few years, although it remains high among all age, sex, and ethnic groups[1]. An estimated 12% of all United States children between 2 and 19 years old have a body mass index (BMI) ≥ 97^th percentile for age and sex. Moreover, the prevalence of obesity differs significantly among racial and ethnic groups. In 2009-2010, 21% of Hispanic and 24% of non-Hispanic black children and adolescents were obese, whereas 14% of non-Hispanic white children and adolescents were obese.

Childhood obesity has several health-related consequences that, until recently, were documented only in adulthood, including hypertension, insulin resistance, glucose intolerance, and dyslipidemia[2]. In turn, these conditions are risk factors for type 2 diabetes and cardiovascular disease in both childhood and adulthood[3-5]. Weight-loss surgery among both adults and adolescents has become increasingly recognized as effective treatment for severe obesity and several of its related comorbidities[6-10]. Current studies suggest that neither pharmacologic nor dietary treatment can maintain weight loss in obese adolescents as effectively as can weight-loss surgery[11,12].

Several studies have shown that among adults, weight loss outcomes after bariatric surgery differ by ethnicity with some studies reporting better outcomes among non-Hispanic whites than among non-Hispanic blacks[13-16]. Although the literature on outcomes of bariatric surgery in adolescents has increased exponentially, many studies are small and usually from single institutional series[8]. As a result, patients are not geographically or ethnically diverse, despite the fact that obesity disproportionately affects ethnic minorities[1].

Accordingly, we analyzed data from the Bariatric Outcomes Longitudinal Database (BOLD), a large national database that tracks outcomes in patients from a wide geographical area who have undergone bariatric surgery. We report here the results of our analysis, which indicate that weight loss in a large cohort of severely obese adolescents 1 year after bariatric surgery did not differ by ethnicity.

Of the 890 eligible adolescents, 75% were females (mean age, 18.5 years 68% were white non-Hispanic, 15% were Hispanic, 11% were non-Hispanic black, and 6% were “other”’ (Table 1). Gastric bypass surgery was performed in 51% of the patients, and gastric band surgery in 49%. At baseline, males were significantly heavier than females. Those undergoing gastric bypass surgery were significantly heavier than those undergoing adjustable gastric band surgery. Non-Hispanic whites were significantly lighter than their ethnic group counterparts (Table 1). Because of these significant baseline differences in sex and surgery type, all subsequent analyses are adjusted for these variables.

All anthropometric measures significantly decreased in non-Hispanic whites and Hispanics throughout the 1 year follow-up period (Table 2). Among both non-Hispanic whites and Hispanics, mean BMI percentile decreased to just below non-obese levels 1 year after surgery (94.67^th percentiles for both groups, respectively).

When the sample was stratified by sex the variation among ethnic groups was only 3.7 kg for girls and 10 kg for boys (Tables 3 and 4). Further analysis revealed no significant differences between pairwise ethnic group comparisons for all anthropometric measures (Table 5).

Changes in BMI varied somewhat by preoperative BMI category. For the entire sample, the highest BMI category lost the most BMI units when compared to lowest BMI category (17.8 kg/m²vs 9.9 kg/m², respectively, P < 0.001). Only Hispanics showed no difference in weight loss between the three groups (11.7 kg/m²) (Table 6). Rate of change over the year for the three BMI categories did not vary much by ethnicity or sex, but they did by surgery type: in all three categories, BMI was reduced more among those who underwent gastric bypass surgeries than among those who had adjustable band surgeries (P < 0.001) (data not shown on tables).

Details about aggregate post-operative complications are published elsewhere[9]. Briefly, of the 120 complications reported (13.5%), 38 were with the gastrointestinal system and 26 were with nutritional deficiencies. Complication rates did not differ by ethnicity. Similarly, when readmissions (n = 55) were analyzed by ethnic group, no significant differences were found. (Data not reported). One death was reported from cardiac arrest 5 mo after (gastric bypass) surgery.

In a national sample of ethnically diverse, severely obese adolescents, 1 year after bariatric surgery, weight and other anthropometric measures had decreased by the same amount in all racial/ethnic groups and in both sexes. Mean estimated weight loss for all three groups at 1 year differed by a maximum of only 1.5 kg. Surgery resulted in an average weight loss of more than 30 kg per person, a loss that far exceeds those reported in non-surgical weight-management programs[21].

Among adults, ethnicity predicts better weight loss outcomes after bariatric surgery, with some studies reporting better outcomes among NHWs than among NHBs. There is little information about outcomes in Hispanics. Harvin et al[13] found that 2 years after undergoing bariatric surgery, NHW adults had lost significantly more weight than did their ethnic group counterparts. Similarly, Buffington et al[15] found that NHB adults had less-successful weight outcomes after gastric bypass than did NHW adults. Among adults, NHB women lost considerably less weight after surgery than did NHW women[13-16]. Additionally, in the Netherlands, Admiraal et al[22] found that African, South Asian, Turkish and Moroccan patients lost less weight at 1-year post-gastric bypass surgery versus their ethnic Dutch counterparts.

Conversely, most studies of adolescent bariatric surgery are case series from a single-institution, resulting in samples that are generally small and relatively homogeneous, which does not permit robust comparisons among ethnic groups[8]. Although race and ethnicity are independently associated with cardiometabolic disease risk[23,24], we found that weight loss was similar among all three ethnic groups and varied more among boys than among girls. However, no overall pairwise group comparisons were significant, indicating that no ethnic group had better weight loss outcomes than did another, unlike comparisons of adults, as described above. Parental influence over post-operative adherence to quality nutrition and physical activity recommendations may partially explain the lack of ethnic group differences.

Although the childhood obesity epidemic continues unabated in most developed countries, non-surgical approaches to the long-term (1 year or more) management and decrease of overweight in childhood have had limited success[21]. Despite standardized indications for bariatric surgery in adolescents[25], obese children are not simply younger versions of obese adults; they are still developing and growing, both physically and psychologically. Extreme obesity should be treated sooner rather than later[8], particularly in adolescents, who may have not yet developed full-blown, related comorbidities, such as diabetes or heart disease. Our analysis found that type of surgery significantly influences weight loss, however and thus must be a consideration for adolescents considering this alternative; gastric bypass surgery resulted in significantly more weight loss at 1-year versus adjustable gastric band surgery. Therefore, gastric bypass may be a viable option for those who are in need of more weight loss to resolve co-morbidities that have already developed. However, the optimal age in adolescence for bariatric surgery is as yet undetermined[24]. Our findings, and those of others[26,27] indicate that bariatric surgery before adulthood can substantially reduce weight[10,25] and resolve comorbidities[25]. Moreover, earlier treatment of obesity may prevent later costs. For example, children and adolescents with a primary or secondary diagnosis of overweight, obesity, or severe obesity require longer hospital stays than do children without these diagnoses[28].

Even as the emerging data on bariatric surgery-including those from randomized-controlled trials[26,29] -continue to show important long-term weight loss and improvement in most obesity-related comorbidities, many pediatric specialists still hesitate to refer patients for surgery[30]. A survey of several hundred pediatricians in the United States showed that, although they believed pediatric obesity to be a major problem, less than half would be somewhat or very likely to refer a severely obese adolescent for surgery[30]. Yet, the medical consequences of childhood obesity continue and suggest that overt disease beginning in early adulthood may become chronic[31].

We thank all the surgeons and centers of excellence that have contributed their data to Bariatric Outcomes Longitudinal Database (BOLD). We also thank ASMBS for its continued efforts to support BOLD. Finally, we are grateful to all the bariatric patients who agreed to share their data for the current analyses.