Effectiveness of surgical interventions for managing obesity in children and adolescents: A systematic review and meta‐analysis framed using minimal important difference estimates based on GRADE guidance to inform a clinical practice guideline

Abstract

SummaryObjectiveTo summarize the literature on bariatric surgery for managing pediatric obesity, including intervention effects to improve patient‐reported outcome measures (PROMs), cardiometabolic risk factors, anthropometry, and assess adverse events (AEs).MethodsEligible studies were published between January 2012 and January 2022 and included randomized controlled trials (RCTs) and observational (controlled and uncontrolled) studies before and after surgery with a mean age <18 years old. Outcomes and subgroups were selected a priori by stakeholders; estimates of effect for outcomes were presented relative to minimal important differences (MIDs) and GRADE certainty of evidence. We examined data on PROMs, cardiometabolic risk factors, anthropometry, and AEs. Subgroup analyses examined outcomes by follow‐up duration and surgical technique, when possible.ResultsOverall, 63 publications (43 original studies) met our inclusion criteria (n = 6128 participants; 66% female). Studies reported six different surgical techniques that were evaluated using uncontrolled single arm observational (n = 49), controlled observational (n = 13), and RCT (n = 1) designs. Most studies included short‐term follow‐up (<18 months) only. PROMs were measured in 12 (28%) studies. Surgery led to large improvements in health‐related quality of life compared to baseline and control groups, and moderate to very large improvements in cardiometabolic risk factors compared to baseline. Large to very large improvements in BMIz were noted compared to baseline across all follow‐up periods. There was limited evidence of AEs with most reporting mild or non‐specific AEs; serious AEs were uncommon.ConclusionBariatric surgery demonstrated primarily moderate to very large improvements across diverse outcomes with limited evidence of AEs, albeit with low to moderate certainty of evidence.