BackgroundEndoscopic submucosal dissection (ESD) is a preferred method for early esophageal cancer, yet its application to esophageal adenocarcinoma (EAC), especially in the Eastern population with its relative rarity, lacks sufficient literature. This study evaluates ESD's long-term outcomes for EAC, focusing on non-curative resections and diagnostic accuracy. MethodsA retrospective study (2012–2022) included 68 patients undergoing ESD for early EAC at Jiangsu Province Hospital. Primary outcomes encompassed ESD efficacy, en bloc resection, R0 resection, curative resection rates, and follow-up. Secondary outcomes involved non-curative ESD, T1a/T1b stage comparison, and diagnostic consistency. ResultsPostoperative staging revealed T1a (n=53) and T1b (n=15) tumors. En bloc resection rate was 97.1%, R0 resection rate was 79.4%, and non-curative rate was 30.9%. T1a had significantly higher R0 rate and curative resection rate. Among non-curative ESDs, 33.3% underwent esophagectomy, 42.9% had surveillance endoscopies, 19.1% repeated curative ESD, and 4.7% were lost to follow-up. Average follow-up was 63.76±28.47 months. Six cases had recurrence, three had residual lesions, and six deaths occurred, unrelated to ESD. No significant difference in survival or recurrence rates between curative and non-curative ESD groups was observed. ESD led to a histologic diagnosis change in 70.6% of cases, all upstaged. ConclusionsESD is effective for EAC, with higher curative rates for T1a than T1b. Non-curative ESD cases may benefit from conservative approaches. Long-term follow-up underscores poor consistency between residual lesions and positive margins. ESD serves as a valuable diagnostic staging tool, particularly for T1b patients, considering the low accuracy of EUS and preoperative biopsy.