Does total mesorectal excision require a learning curve? Analysis from the database of a single surgeon's experience.

Abstract

BACKGROUND: The procedure of total mesorectal excision (TME) is the gold standard in the treatment of rectal cancer. However, quality control of TME is still under debate. The present study was conducted to determine whether TME requires a learning curve to allow the surgeon to grasp the necessary technical expertise.

METHODS: We performed a retrospective review of patients with rectal cancer who underwent TME with curative intention between August 1998 and December 2003; 195 consecutive patients were enrolled. From the first patient of the cohort, the first 50 patients were categorized into group 1, the next 50 into group 2, the next 50 into group 3, and the final 45 patients into group 4. Local recurrence rates were compared between the four groups.

RESULTS: No significant difference in clinicopathological features was observed between the four groups, except for age, operative time, and grade of mesorectum. The local recurrence (LR) rate decreased from 22.3% in the inadequate TME group (G1) to 9.1% in the adequate TME group (G2-4) (p=0.035). In multivariate analysis, regional lymph node metastasis, mesorectal grade (incomplete or nearly complete), and early period of learning curve were independent predictors of local recurrence.

CONCLUSIONS: Our results suggest that a learning curve is necessary for the development of technical expertise in the performance of TME for treatment of rectal cancer.