Supplementary Material

No data


Background: Endoscopic submucosal dissection (ESD) can be used as a minimally invasive treatment in the early stages of esophageal squamous cell carcinoma (ESCC). Due to the higher frequency of underlying illnesses, the superiority of ESD to esophagectomy is more prominent in the elderly than in the non-elderly. This study compared the efficacy and safety of ESD for early ESCC between the elderly and non-elderly patients.

Methods: From August 2014 to December 2019, 59 patients treated with ESD for early ESCC were retrospectively analyzed. The elderly group (n = 20) comprised patients aged ≥ 65 years, and the non-elderly group (n = 39) comprised those aged < 65 years. Clinical and pathological results, including R0 resection, curative resection, survival, and complication rates were compared between the groups.

Result: Both groups were male predominant (elderly: 80%, non-elderly: 87.2%), and the tumor was mostly found in the mid thoracic (elderly: 40%, non-elderly: 61.5%). Tumor size, depth of tumor, and lymphovascular invasion were not significantly different between the groups. Tumor occupying ≥ 3/4 circumference of the esophagus was only found in the non-elderly group (elderly: 0, non-elderly: 8; p = 0.036). The R0 resection rate (elderly: 100%, non-elderly: 94.9%, p = 0.544), curative resection rate (elderly: 80%, non-elderly: 79.5%, p = 1), recurrence rate (elderly: 10%, non-elderly: 10.3%, p = 1), follow-up period (elderly: 20.4 ± 16.8 months, non-elderly: 21 ± 17.6 months, p = 0.904), and post ESD stricture (elderly: 5%, non-elderly: 7.7%, p = 1) were similar between the two groups. The elderly had longer hospital stays than the non-elderly, but without a significant difference (elderly: 5 ± 1.4 days, non-elderly: 4.3 ± 1.3 days; p = 0.08). Post ESD stricture was correlated with the tumor circumference (p = 0.001). Cox model stratified by age groups, adjusting for other factors, non-curative ESD and concurrent other malignancy were associated with increased 5.5-year mortality (hazard ratio 6.30; 95% CI, 1.03–38.68; hazard ratio 13.78; 95% CI, 1.50–126.48).

Conclusion: ESD was effective and safe for early ESCC in the elderly.