Appropriateness of the indication for upper endoscopy: a meta-analysis

Abstract

Background: Application of appropriate indications for upper endoscopy (EGD) should conserve limited endoscopic resources. To perform a systematic review and meta-analysis to assess the efficacy of American Society for Gastrointestinal Endoscopy and European Panel on the Appropriateness of Gastrointestinal Endoscopy guidelines in selecting patients referred for an upper endoscopy relative to the detection of gastro-oesophageal cancer.

Methods: Studies comparing the appropriateness of EGD indication according to American Society for Gastrointestinal Endoscopy or European Panel on the Appropriateness of Gastrointestinal Endoscopy guidelines and the detection of relevant endoscopic findings and cancer were identified by searching the Medline (1982 to September 2008). Pre-defined outputs of the meta-analysis were sensitivity, specificity, and positive and negative likelihood ratios (LR+, LR-).

Results: We included eight cohort studies comprising 13,856 patients; 10,643 EGD indications were categorized as appropriate, and 3010 (22%) as inappropriate. For relevant findings, the adjusted sensitivity, specificity, LR+, and LR- were 85% (95% CI, 84-86%), 28% (95% CI, 27-29%), 1.18 (95% CI, 1.1-1.3) and 0.6 (95% CI, 0.5-0.7), respectively. For cancer detection, the adjusted sensitivity, specificity, LR+, and LR- were 97% (95% CI, 94-98%), 22% (95% CI, 22-23%), 1.2 (95% CI, 1.1-1.4), and 0.2 (95% CI, 0.05-0.9), respectively.

Conclusions: For inappropriate EGD, the very low likelihood of cancer argues against endoscopic referral, whereas the low specificity substantially reduces the predictive value of an appropriate indication for both cancer and relevant endoscopic findings.