BACKGROUND AND STUDY AIMS: Endoscopic resection (ER) has proven effective and safe for T1 esophageal adenocarcinoma (EAC). However, uncertainty remains concerning risk-benefit return of esophagectomy for submucosal lesions (T1b). Surgical series in past decades have reported significant risk of lymph node metastasis (LNM) in T1b EAC, but these rates may be overestimated due to limitations in histological assessment of surgical specimens. We aimed to test this hypothesis by reassessing histological risk features in surgical specimens from T1b EAC cases with documented LNM. PATIENTS AND METHODS: A retrospective cross-sectional study (1994-2005) was conducted. Patients who underwent direct esophagectomy without prior neoadjuvant therapy for suspected T1b EAC with LNM were included. Additional tissue sections were prepared from archival tumor blocks. A consensus diagnosis on tumor depth, differentiation grade, and lymphovascular invasion (LVI) was established by a panel of experienced pathologists. RESULTS: Specific depth of submucosal invasion (sm1 to sm3) was not specified in 10 of 11 archival case sign-out reports. LVI status was not reported in seven of 11 cases. Following reassessment, one patient was found to have deep tumor invasion into the muscularis propria (T2). The remaining 10 of 11 patients exhibited deep submucosal invasion (sm2-3), with five showing one or more additional risk features (poor differentiation and/or LVI). CONCLUSIONS: Our findings highlight the potential for underestimating tumor depth of invasion and other high-risk features in surgical specimens. Despite the limited cohort size, our study confirmed a consistent high-risk histological profile across all cases. Caution is warranted when extrapolating LNM risk data from historic heterogeneous cross-sectional surgical cohorts to the modern ER era.