What Is Known/What Is New
What Is Known
- Esophagogastroduodenoscopy often is performed to evaluate for mucosal and anatomical abnormalities before bariatric surgery, such as vertical sleeve gastrectomy.
- Adult studies from United States cohorts found that 1% to 9% had an alteration or delay in sleeve gastrectomy because of abnormalities found on esophagogastroduodenoscopy. In addition, 2 recent meta-analyses found that 7.6% and 7.8% had a delay or required an alteration in surgical management because of upper endoscopy results.
What is New:
- In this cohort of adolescent patients, abnormal esophagogastroduodenoscopy findings were not associated with modification of bariatric surgery or any adverse outcome.
- Gastrointestinal symptoms were the only predictor for esophagogastroduodenoscopy abnormalities.
Rates of obesity are historically high in all age groups, with the most severe forms rising in prevalence (1,2). The incidence of adolescents undergoing bariatric surgery with vertical sleeve gastrectomy is rapidly increasing in the United States (US) (3). Bariatric surgery is an effective weight-loss strategy for adolescent patients with morbid obesity (4–6). A recent meta-analysis found that the 3 most commonly used bariatric techniques (gastric banding, gastric bypass, and sleeve gastrectomy) resulted in sustained mean absolute BMI change and weight loss that persisted up to at least 36 months postoperatively (7). Adolescents undergoing surgery had better glycemic control and cardiovascular risk markers than patients who received only medical therapy (8).
Bariatric interventions are complex, and a multidisciplinary, prebariatric assessment is an essential part of the evaluation and treatment process. Despite this complexity, however, no uniform guidelines regarding assessment of upper GI tract anatomy and histology before bariatric surgery exist. One suggested screening includes a visual evaluation of the stomach via an esophagogastroduodenoscopy (EGD) to assess for underlying pathology that might negatively impact surgery and postoperative outcomes. Studies from adult US cohorts have demonstrated that patients with EGD abnormalities are often asymptomatic, and only 1% to 9% had an alteration or delay in surgery (9–12). As a result, the 2015 American Society of Gastrointestinal Endoscopy (ASGE) guidelines recommended that physicians make decisions about the potential utility of preoperative endoscopy on an individual basis and after thorough preoperative evaluation (13). More recent non-US-based studies, however, have suggested that gastrointestinal pathology in the bariatric population may negatively impact outcome, and more routine preoperative endoscopic evaluation may be warranted (14–16).
Although the results from adult studies are conflicting, no study or guideline exists to help direct an informed decision regarding preoperative endoscopic evaluation among adolescent patients. The prevalence and risk factors of anatomic or mucosal upper gastrointestinal abnormalities that could interfere with bariatric surgery among adolescents is unknown. Therefore, the aim of this study was to determine the relationship between prebariatric EGD and modification in management. More specifically, we aimed to assess number of endoscopies completed, proportion of endoscopies with abnormal findings, and proportion of endoscopies resulting in intervention. A secondary aim was to examine demographic and clinical predictors of an abnormal EGD before bariatric surgery.
We performed a retrospective cohort study of adolescents undergoing evaluation for bariatric surgery candidacy at the Center for Adolescent Bariatric Surgery (CABS) of NewYork Presbyterian Morgan Stanley Children's Hospital and Columbia University Irving Medical Center. The CABS program is a multidisciplinary program including pediatric surgery, endocrinology, nutrition, psychiatry, and gastroenterology with a current case volume of 40 to 50 gastric sleeve procedures per year. The electronic medical records of all patients undergoing a presurgical evaluation were reviewed for a period of 2 years (between October 2014 to September 2016). This study was approved by the Institutional Review Board of Columbia University Irving Medical Center.
Patients were eligible for inclusion if they entered the bariatric surgery program, underwent EGD, followed by sleeve gastrectomy after completing the standard multidisciplinary pre-surgical evaluation. Patients were excluded if they did not undergo an EGD or if they were not considered a candidate for bariatric surgery. EGDs were performed in a pediatric endoscopy suite or pediatric operating room under conscious sedation.
We obtained demographic and anthropometric data in addition to EGD findings, biopsy pathology, gastrointestinal symptoms, and surgical outcomes. An EGD was considered abnormal if either abnormal gross findings or abnormal pathology was reported. No subject underwent a preoperative screening upper GI contrast study in part to eliminate radiation exposure. During the study period, the bariatric surgical intervention at our bariatric center was a sleeve gastrectomy (SG). Of note, as part of the routine sleeve gastrectomy, the esophageal hiatus is dissected to evaluate for widening at the time of surgery. In addition to the preoperative EGD by the gastroenterologist, all patients underwent intraoperative endoscopy by the surgeon following sleeve gastrectomy to evaluate for bleeding, torsion, or staple line leak. Records were reviewed from the first available preoperative visit until a standardized 6-week post-operative visit.
The primary outcome for this research was the number of abnormal endoscopies, which were defined as endoscopies with either gross or histologic abnormalities that lead to a modification in management (medical, surgical, or cancellation of surgery). Secondary outcomes included identifying demographic or clinical factors predictive of an abnormal EGD and/or modification in management. Furthermore, we assessed the frequency of adverse events of EGDs and postoperative complications. EGD adverse events were included in these analyses to assess the benefits and risks of the procedure. We included immediate postoperative complications to determine if there were any means to better predict and thereby prevent any complications for preoperative EGDs.
After assessing for normality, the demographics were analyzed as continuous variables via t tests or the Mann-Whitney U test wherever indicated. After stratification of the primary outcome into groups based on interventions, categorical variables were analyzed with bivariate statistics, chi-square and Fisher exact tests. Odds ratios were calculated with logistic regression analyses to assess if demographic and clinical factors were associated with or were a potential predictor for interventions. Statistical analyses were based on a 2-sided tail model with P < 0.05 and were conducted with IBM SPSS Statistics version 20.0 (IBM, Armonk, NY).
Of 134 patients presenting for evaluation, 94 (70%) underwent a preoperative EGD and were eligible for inclusion (Fig. 1). The median age was 16 years (range: 9–23 years), 73% were girls, 30% non-Hispanic white, 13% non-Hispanic black, 48% Hispanic, and 9% other (Table 1).
Of the remaining 40 patients who were excluded because they did not undergo an EGD, 23 proceeded to surgery and 17 did not proceed with either. Among the patients who did not undergo an EGD before proceeding to SG, there were no visual/anatomical abnormalities reported during the intraoperative EGD and there were no surgical complications. Supplemental Table 3 (Supplemental Digital Content, http://links.lww.com/MPG/B637), describes the abnormal pathology reports for patients who only underwent an intraoperative EGD.
Of the 94 patients included in the study, 51 (54%) had a normal EGD and 43 (46%) had an abnormal EGD. Of the patients with an abnormal EGD, 7 had a possible anatomical abnormality based on gross endoscopy, including 4 with a possible hiatal hernia (none visualized during sleeve gastrectomy) and 3 with a donut-shaped appearance to the gastric cardia because of a prior gastric band (this did not require any further intervention). The remaining 36 patients had mild mucosal abnormalities visualized at the time of endoscopy and/or during pathologic examination of biopsy specimens (Fig. 1).
Among those with EGD abnormalities, 21 (22%) received an intervention but none required a modification or cancellation of their bariatric surgery. The most common post-EGD interventions were PPI administration (n = 10) and Helicobacter pylori eradication (n = 11). One patient underwent a polypectomy during the EGD and also received therapy for an H pylori infection. In addition, no patients had an EGD result that required a bariatric procedure modification and there were no patients whose EGD results were considered a contraindication to surgery. EGD findings did not result in delay of SG.
Predictors of Abnormal Esophagogastroduodenoscopies
Patients with abnormalities identified on EGD had similar demographic characteristics as patients without EGD abnormalities including age, sex, BMI, weight, height, race, ethnicity, and insurance status (private vs Medicaid) (Table 1). Whenever clinical characteristics were compared between patients with and without an abnormal EGD, 35% of patients with an abnormal EGD had gastrointestinal symptoms whereas only 10% of patients with normal EGDs had symptoms (P < 0.01). Logistic regression analysis demonstrated that patients who reported GI symptoms were almost 5 times more likely to have a positive EGD [Odds Ratio (OR) 4.9 [95% confidence interval (CI) 1.6–15.0; P < 0.001]. Other clinical factors including prior gastric band surgery, history of GI comorbidities, and a prior use of GI medications did not vary significantly between the groups.
Predictors for Interventions
Although patients with GI symptoms were 5 times more likely to have an abnormal EGD, they were not significantly more likely to require an intervention (OR 2.3,95% CI, 0.78–6.85; P = 0.13). Only 2 patients with H pylori had symptoms, and only 5/10 patients who received a PPI post-EGD had symptoms. No other demographic or clinical characteristics predicted the likelihood of an intervention.
Adverse Events/Complications of Esophagogastroduodenoscopies
During the 94 EGD procedures, no serious adverse events were observed. Seven of the EGDs had difficulties stated on the procedural report (see Supplemental Table 1, Supplemental Digital Content, http://links.lww.com/MPG/B637, which lists minor adverse events associated secondary to EGDs). The most reported events were patient discomfort during the procedure. These events occurred at similar rate between patients with abnormal and normal EGD findings with an OR of 0.41 (95% CI, 0.08–2.22; P = 0.30).
Of the 83 patients who underwent a sleeve gastrectomy there were 5 (6%) complications related to surgery (see Supplemental Table 2, Supplemental Digital Content, http://links.lww.com/MPG/B637, which lists surgical complications). In addition, 37 patients reported subjective postoperative symptoms at outpatient visits up to 6 weeks postoperatively, which predominately included GI symptoms. Neither complications or postoperative symptoms were more common in the abnormal EGD group OR 0.72 (95% CI, 0.11–4.54; P = 0.73) and normal EGD group OR 0.66 (95% CI, 0.27–1.61; P = 0.360), respectively.
In this cohort of adolescents undergoing evaluation for sleeve gastrectomy, 46% had an abnormal EGD by either gross findings and/or histopathologic examination. Sixteen percent of the abnormal studies were found to have a suspected anatomical abnormality, for which no intervention was required. The remaining 84% had mild mucosal abnormalities only, for which almost half (48%) required an intervention before SG. Interventions were primarily limited to proton pump inhibitor administration or H pylori eradication. Patients with prior GI symptoms were 5 times more likely to have an abnormal EGD result; however, this did not predict the need for an intervention.
There are no current guidelines regarding preoperative screening in children before sleeve gastrectomy; current practice is based on adult studies which have demonstrated discrepant findings. The recent ASMBS Pediatric Metabolic and Bariatric Surgery Guidelines, 2018 make no mention of indications for preoperative endoscopy (17). For this reason, some adolescent bariatric centers have elected to perform an EGD on all patients as part of the preoperative evaluation for sleeve gastrectomy. Only half of the patients in our sample, however, had an abnormal endoscopy and the majority of abnormalities identified were mild and requiring routine interventions that did not correlate with modification or complication rate in sleeve gastrectomy.
As our results demonstrate that no gastric sleeve procedures required modification or cancellation secondary to EGD findings, it may be that adolescents might benefit less from routine preoperative EGDs than adults. Adult studies from US cohorts found that 1 to 9% had an alteration or delay in SG because of abnormalities found on EGD (9–12). In addition, 2 recent meta-analyses found that 7.6% and 7.8% had a delay or required an alteration in surgical management because of upper endoscopy results (18,19). Adults are at higher risk of significant upper gastrointestinal disease that may impact management, such as esophageal and gastric cancer, Barrett's esophagus and hiatal hernia, all of which are rarely seen in young adults (20).
To our knowledge, no prior studies have investigated use of preoperative EGDs among adolescent patients. One prior study evaluated the utility of preoperative imaging in adolescents undergoing the bariatric banding procedure and found that imaging did not affect the surgical plan or outcome (21). In contrast, this present study investigated the utility of EGD in the youngest published cohort of patients undergoing evaluation for sleeve gastrectomy. Another strength is that while some adult cohorts were larger in prior studies, these investigations often included patients who were evaluated for different types of bariatric surgeries, making intervention data difficult to interpret. The present study's cohort was more homogenous, as we only included patients who were evaluated for sleeve gastrectomy and the procedure was performed by a surgical team with the same surgeon (J.L.Z.).
This study has several limitations. The main limitation is its retrospective design, which could have led to the inadvertent exclusion of procedures and subjects. This risk was minimized as all patients visiting the multidisciplinary bariatric surgery clinic were screened for a consecutive time period of 2 years and were all operated on by the same surgeon. Although the majority of patients in our population had a preoperative EGD, studies were performed more selectively towards the end of the second year of the data-gathering period based on the clinical impression of the surgeon. This change reflects a shift in perception regarding the utility of upper endoscopy for bariatric screening. As a result, some patients with EGD abnormalities could have been missed. No patients, however, with complications because of an anatomic abnormality or mucosal abnormalities found at the time of the SG were identified in the total patient sample during the review. This shift in practice could have also hypothetically lead to a selection bias to perform EGDs in patients with more significant gastrointestinal disease, but the number of interventions was still low and no patients had alterations or a delay of their surgical management. Lastly, the study included a relatively small sample size, which may have in turn increased the likelihood of a Type 2 error. Our study, however, had a similar power compared with many adult bariatric studies.
About half of our adolescent cohort undergoing prebariatric EGD had mild abnormalities on EGD, and half of these patients required an intervention before bariatric surgery. These interventions were benign and did not interfere with, delay, or result in an alteration of the bariatric surgical procedure. Although we found that patients with GI symptoms were 5 times more likely to have an abnormal EGD result, no other demographic or clinical characteristics were identified that predicted an abnormal EGD result. Furthermore, no factors predicted the need for an intervention. Although our findings were similar to previous adult US cohorts who also found a low yield for EGDs, future adolescent studies should have a multicenter prospective study design to identify factors associated with clinically relevant interventions and bariatric surgical outcomes. Furthermore, these results suggest that EGD should not be used as a routine procedure, but rather that it may be particularly beneficial for patients who have undergone prior intervention, such as adjustable gastric banding and those who have GI symptoms.
1. Agha M, Agha R. The rising prevalence of obesity: part A: impact on public health. Int J Surg Oncol (N Y)
2. Hales CM, Fryar CD, Carroll MD, et al. Trends in obesity and severe obesity prevalence in US youth and adults by sex and age, 2007-2008 to 2015-2016. JAMA
3. Griggs CL, Perez NP Jr, Goldstone RN, et al. National trends in the use of metabolic and bariatric surgery among pediatric patients with severe obesity. JAMA Pediatr
4. Paulus GF, de Vaan LE, Verdam FJ, et al. Bariatric surgery in morbidly obese adolescents: a systematic review and meta-analysis. Obes Surg
5. Pedroso FE, Gander J, Oh PS, et al. Laparoscopic vertical sleeve gastrectomy significantly improves short term weight loss as compared to laparoscopic adjustable gastric band placement in morbidly obese adolescent patients. J Pediatr Surg
6. Yermilov I, McGory ML, Shekelle PW, et al. Appropriateness criteria for bariatric surgery: beyond the NIH guidelines. Obesity (Silver Spring)
7. Pedroso FE, Angriman F, Endo A, et al. Weight loss after bariatric surgery in obese adolescents: a systematic review and meta-analysis. Surg Obes Relat Dis
8. Inge TH, Laffel LM, Jenkins TM, et al. Comparison of surgical and medical therapy for type 2 diabetes in severely obese adolescents. JAMA Pediatr
9. Gomez V, Bhalla R, Heckman MG, et al. Routine screening endoscopy before bariatric surgery: is it necessary? Bariatr Surg Pract Patient Care
10. Loewen M, Giovanni J, Barba C. Screening endoscopy before bariatric surgery: a series of 448 patients. Surg Obes Relat Dis
11. Mong C, Van Dam J, Morton J, et al. Preoperative endoscopic screening for laparoscopic Roux-en-Y gastric bypass has a low yield for anatomic findings. Obes Surg
12. Zeni TM, Frantzides CT, Mahr C, et al. Value of preoperative upper endoscopy in patients undergoing laparoscopic gastric bypass. Obes Surg
13. Evans JA, Muthusamy VR, Acosta RD, et al. The role of endoscopy in the bariatric surgery patient. Gastrointest Endosc
14. Lee J, Wong SK, Liu SY, et al. Is preoperative upper gastrointestinal endoscopy in obese patients undergoing bariatric surgery mandatory? An Asian Perspective Obes Surg
15. Praveenraj P, Gomes RM, Kumar S, et al. Diagnostic yield and clinical implications of preoperative upper gastrointestinal endoscopy in morbidly obese patients undergoing bariatric surgery. J Laparoendosc Adv Surg Tech A
16. Wiltberger G, Bucher JN, Schmelzle M, et al. Preoperative endoscopy
and its impact on perioperative management in bariatric surgery. Dig Surg
17. Pratt JSA, Browne A, Browne NT, et al. ASMBS pediatric metabolic and bariatric surgery guidelines, 2018. Surg Obes Relat Dis
18. Bennett S, Gostimir M, Shorr R, et al. The role of routine preoperative upper endoscopy in bariatric surgery: a systematic review and meta-analysis. Surg Obes Relat Dis
19. Parikh M, Liu J, Vieira D, et al. Preoperative endoscopy
prior to bariatric surgery: a systematic review and meta-analysis of the literature. Obes Surg
20. Lightdale JR, Acosta R, Shergill AK, et al. American Society for Gastrointestinal Endoscopy. Modifications in endoscopic practice for pediatric patients. Gastrointest Endosc
21. Averill LW, Stevenson KL, Kecskemethy HH, et al. Pre- and postoperative imaging of adolescents undergoing laparoscopic adjustable gastric banding surgery. Pediatr Radiol