ScholarWorks@성균관대학교

초록

Background and Aims Gastric indefinite for dysplasia (IND) is an indeterminate lesion with atypia that is insufficient for a diagnosis of dysplasia but not benign. The management of gastric IND is currently based on low-quality evidence. This study aimed to provide evidence through a systematic review and meta-analysis of studies investigating the management of lesions classified as gastric IND on initial endoscopic biopsy. Methods We searched MEDLINE, Embase, Cumulative Index to Nursing and Allied Health Literature, the Cochrane Library, clinical trial registries, and conference proceedings. We screened and selected studies on the management of gastric IND. A meta-analysis was conducted using a random-effects model. Results Sixteen studies comprising 2226 lesions were selected from 5568 records. The pooled estimate of gastric cancer diagnosis was 66% (95% CI, 54%-75%) in the intervention group and 20% (95% CI, 16%-24%) in the rebiopsy group. Among individuals aged ≥60 years, the gastric cancer diagnosis rate was 75% (95% CI, 62%-85%) in the intervention group. Four studies reported the risks of the factors using odds ratios. Lesion size (10-mm cutoff), surface depression, nodularity, redness, spontaneous bleeding, and multiple lesions were reported as significant. Three studies assessed pathologic reliability. Conclusions The meta-analysis results showed a considerably high rate of gastric cancer diagnosis among patients with gastric IND, particularly in the intervention group and in the older age subgroups. We recommend that guideline societies recognize gastric IND as an independent lesion and work toward a consensus on the cytopathologic features that define it.