Introduction: A modified posterior pelvic exenteration (MPE) might be needed to reach an optimal tumoral reduction. The issue of this study is to relate a multicentric experience of this kind of resection.

Materials: Three hundred five patients who needed an MPE were analyzed from 9 French cancer centers. One hundred sixty-eight MPEs were performed during initial surgery (55.1%), 69 during interval surgery (22.6%), 36 after chemotherapy (11.8%), and 32 for recurrences (10.5%).

Results: Three hundred two colorectal anastomoses were realized with a protective stoma in 59 (19.5%) of cases and a stoma closure in 76.5% (51). The rate of functional anastomosis was 96% (290/302). Complications occurred in 26.9% (82/305) of the patients, with a fistula in 25 (8.2%). The reintervention rate was 8.8% (27/305). The median length of hospitalization was 15 days. The absence of a macroscopic residual disease was obtained in 58% (173/303) of cases. A residual disease that was 1 cm or smaller was observed in 73 cases (24%) and 2 cm or smaller observed in 36 (11.9%). Postoperative chemotherapy was started with a median time of 32 days.

Postoperative death occurred in 1 patient (0.33%). The survival rates were 62.7% and 27.6% at 2 and 5 years, respectively. With a multivariate analysis, the 2 significant prognostic factors were residual disease and time of surgery (P < 0.0001).

Conclusions: A rectal invasion should not be an obstacle to reach the aim to obtain a macroscopic minimal residual disease or, if possible, the absence of one. An MPE is useful in those cases to reach optimal cytoreduction, with comparable results whatever the patient's age is. A temporary protective stoma should be considered only exceptionally.