Impact of flap reconstruction on perineal wound complications following ablative surgery for advanced and recurrent rectal cancers.

ABSTRACT

BACKGROUND: To determine the effect of flap reconstruction on perineal complications in locally advanced rectal cancers (LARC) and locally recurrent rectal cancers (LRRC). Prior studies have suggested that flap reconstruction may decrease wound complications after ablative surgery for rectal cancer but are limited by small sample sizes, heterogeneity of pathologies, and lack of comparison groups. METHODS: A retrospective cohort study (1999-2010) was performed on consecutive patients undergoing abdominoperineal resection (APR) or pelvic exenteration for locally advanced/locally recurrent rectal cancers. Differences in perineal complications between patients treated with and without perineal flap reconstruction were analyzed by using univariable, multivariable, and propensity score regression analyses. RESULTS: Flap reconstruction was performed in 52 of 177 patients (29 %). Patients receiving flap reconstruction had multiple risk factors for perineal morbidity, including longer operative times and more complex procedures. In our final multivariable analyses that were stratified by type of ablative procedure, we found a trend toward lower odds of perineal complications in patients receiving flaps (p = 0.065) compared with primary closure after pelvic exenteration. Although operative time and sacrectomy were significant determinants of perineal morbidity for pelvic exenteration patients, no significant predictors of perineal outcomes were identified for patients undergoing APR. CONCLUSIONS: This study suggests that flap reconstruction may provide some protective effect against perineal complications in patients undergoing pelvic exenteration, although this was not observed for APR. The most important determinants of perineal complications after pelvic exenteration were operative time and sacral resection, but no predictive factors for post-APR perineal outcomes were identified.