We performed a systematic review to examine the diagnostic yield (endoscopic and histologic) of esophagogastroduodenoscopy (EGD) for the evaluation of abdominal pain of unclear etiology in children. We also examined the effect of EGD on change in treatment, quality of life, change in abdominal pain, and cost-effectiveness.
All full-length articles published in English during 1966–2005 were included if: (a) participants had abdominal pain without known underlying gastrointestinal disease, (b) participants underwent EGD primarily for the evaluation of abdominal pain, (c) findings of the EGD were reported, (d) participants were under 18 yr, and (e) sample size greater than 50.
Eighteen articles examining 1,871 patients fulfilled the inclusion and exclusion criteria. All were observational and most (13) were prospective. Only three studies were performed in the United States and of those two were prospective. The largest study examined about 400 procedures and 13 studies examined less than 100 procedures. One case of inflammatory bowel disease and 67 duodenal or gastric ulcers were reported, thus diagnostic yield was achieved in 3.6% of cases. The prevalence of nonspecific histological gastrointestinal inflammatory lesions varied between 23% and 93%. Six articles attempted to correlate endoscopic or histologic findings with treatment management decisions. No articles attempted to describe quality of life or cost-effectiveness. None of the studies analyzed the association of alarm symptoms or signs to diagnostic yield.
The diagnostic yield of EGD in children with unclear abdominal pain is low; however, existing studies are inadequate. The effect of EGD on change in treatment, quality of life, improvement of abdominal pain, and cost-effectiveness is unknown. The predictors of significant findings are unclear. Our findings suggest that a large multicenter study examining clinical factors, biopsy reports, and addressing patient outcomes is needed to further clarify the value of EGD in children with abdominal pain.