Predictors of upstage diagnosis after endoscopic resection of gastric low-grade dysplasia

Background The optimal management of precursor lesions such as gastric low-grade dysplasia is crucial in order to improve gastric cancer-related mortality. However, there are no universally accepted management guidelines regarding which lesions should be resected or should be monitored by follow-up visits. Patients and methods We retrospectively analyzed data from 1006 gastric low-grade dysplasia lesions that had been resected via endoscopic submucosal dissection. We also evaluated the endoscopic risk factors associated with upstage diagnosis from low-grade dysplasia to high-grade dysplasia or gastric cancer. Results The mean age of our patients was 63.7 +/- 9.1 years and 70.3% of our study population included men. The predominant location and gross type of lesions was the lower third of the stomach (78.6%) and the elevated type (57.8%), respectively. The rates of pathological concordance, upstage, and downstage diagnosis were 85.3, 12.1, and 2.6%, respectively. Multivariate analysis, after adjusting for age and sex, showed that a lesion size >= 10 mm (Odds ratio [OR] 2.231; p = 0.003), erythema (OR 7.315; p < 0.001), nodularity (OR 5.589; p < 0.001), depression (OR 3.024; p = 0.002), and erosion (OR 7.680; p < 0.001) were all factors significantly associated with upstage diagnosis. Furthermore, an increasing number of risk factors was associated with an increasing frequency of upstage diagnosis; if there were no risk factors, then there was no upstage diagnosis. Conclusions This study identified several risk factors that were significantly associated with the upstage diagnosis of gastric low-grade dysplasia: lesion size >= 10 mm and a variety of surface changes (erythema, nodularity, depression, and erosion). Our data indicate that if there is no evidence of these endoscopic risk factors, then regular follow-up may be considered, according to the patient's combined comorbid conditions.