Background There are no clear clinical criteria for the management of gastric lesions diagnosed as adenomas (Vienna classification category 3) by pre-treatment biopsy. In the present study, we examined the feasibility of magnifying endoscopy with narrow-band imaging (ME-NBI) in discriminating early gastric cancers (Vienna classification category 4 or 5) from adenomas in lesions diagnosed as adenomas by pre-treatment biopsy. Methods This was a single-center cross-sectional retrospective study at a tertiary referral center. One hundred thirty-seven consecutive cases of gastric lesions diagnosed as adenomas in pre-treatment forceps biopsy were examined with conventional non-magnifying endoscopy under white light, non-magnifying chromoendoscopy, and ME-NBI. We investigated the association between the final pathological diagnoses (carcinoma or adenoma) and the following factors: lesion size (mm), color (red or white), macroscopic type (depressed or others), presence of ulceration, and positive ME-NBI finding. The presence of an irregular microvascular pattern or an irregular microsurface pattern with a demarcation line between the lesion and the surrounding area was regarded as a positive ME-NBI finding. Results Lesion size was significantly larger in carcinomas than adenomas (P = 0.005). Depressed lesion (P = 0.011), red color (P \ 0.001), and positive ME-NBI finding (P \ 0.001) were significant predictive factors for carcinoma. Multivariate logistic regression confirmed that red color (odds ratio [OR] 3.04, 95% confidence interval [CI] 1.26-7.34, P = 0.14) and a positive ME-NBI finding (OR 13.68,, P \ 0.001) were independent predictive factors for carcinomas. A positive ME-NBI finding was the strongest predictive factor. Conclusions ME-NBI is useful in planning the management of lesions diagnosed as adenomas by pre-treatment forceps biopsy.