Local recurrence after mesorectal excision for rectal cancer.

AIMS

Controversy still exists about the optimal surgical treatment of rectal cancer. The main purpose of the present study was to compare local recurrence (LR) rates after mesorectal excision (ME) and conventional surgery (CS) technique.

METHODS

All rectal cancer patients from a defined catchment area were included. Outcome after ME in the period 1993-1999 (n=161) was compared with the outcome after CS (n=217) in the period 1983-1992. Partial ME (PME) was the routine in upper, and total ME the routine in mid- and low rectal cancer. The follow-up programmes were identical, and the median observation times very similar (37 and 38 months) in the two periods. Five-year actuarial LR rate and survival were estimated using the Kaplan-Meier method, and adjustment for prognostic factors was performed with Cox regression analysis.

RESULTS

Total LR rate after R0 resection was 7.7% crude and 9% 5 year actuarial in the ME period, as compared with 16.0% crude and 24% actuarial in the CS period (P=0.02). Cox regression analyses confirmed these differences with a hazard ratio of 0.40 for ME vs CS (P=0.02). Isolated LR rate was 2% after ME and 8% after CS. Five-year actuarial total LR rate after rectal resection with curative intent was 11% after ME and 27% after CS (P<0.01). Actuarial total LR rate after PME was 6%, and none of these patients developed isolated LR.

CONCLUSION

Standardization of surgical technique and application of ME resulted in a significant reduction of LRs. LR rate was low after PME, indicating that this procedure is adequate in upper rectal cancer.