Endoscopic Decision of Complete Resection Is More Practical Than Pathologic Decision in the Endoscopic Resection for Rectal Carcinoids

Abstract

Background/Aim: Rectal carcinoids are often described as low-grade malignancy and are usually treated by endoscopic resection. However, reports on the outcomes of endoscopic resection for rectal carcinoids remain limited to several single-institution studies. We reviewed the experiences of endoscopic resection for rectal carcinoids in Korea. Methods: This study retrospectively analyzed the medical records of 409 patients with rectal carcinoid at 14 university hospitals in Korea from January 1999 until June 2007 in the aspects of age, sex, tumor size, complete resection, invasion, recurrence, and complications. Residual disease was determined when carcinoid tumor cells were detected histologically at the resection site in 6 months after endoscopic resection. Results: The mean age at diagnosis was 50 years (range 15-90). Male to female ratio was 1.1: 1. The rectal carcinoids were measured 7.66 ± 4.96 mm in maximal diameter. Three hundred fifty-four patients were treated with endoscopic resection, whereas 55 patients underwent operation. In the endoscopic resection group, complete resection rates were 88.9% based on the endoscopic appearance (CR-E) and 59.3% based on the pathologic evaluation (CR-P). The agreement between CR-E and CR-P was low (κ = 0.188). In 83 of 88 patients (94%) whose disease were defined CR-E but were not achieved CR-P, residual diseases were not found on 6 months follow-up endoscopic biopsy. Complications of endoscopic resection occurred in 13 patients, but were managed endoscopically in all cases. The median follow-up period was 26 months (range 6-147). Local recurrence developed in 2 of 210 patients who had complete resection. One patient was treated with a second endoscopic resection, and the other patient underwent operation. No recurrence has been reported until 26 months follow-up after the second endoscopic resection and operation. The 3-year overall survival rate was 100%. There were 14 patients with metastasis at the time of diagnosis. Independent factors predicting metastasis were age (p = 0.0329), tumor size (p < 0.001), central depression (p < 0.001), ulcer (p < 0.001), lymphatic invasion (p < 0.001), and venous invasion (p < 0.001) in multivariate analysis. Conclusions: Endoscopic resection is an effective therapeutic modality for selected cases of rectal carcinoids (≤10 mm). Discrepancies were observed between CR-E and CR-P, but CR-E better predicted complete resection state probably because endoscopic cauterization might have eradicated minimal tumor cells close to the resected margin. Patients with age, tumor size >10 mm, central depression, ulcer, and lymphatic or vascular invasion have higher metastatic potential.