EP374 - Pelvic Exenteration for Benign Pelvic Fistula Disease - Rethinking a Classically Oncologic Operation

Purpose/Background: Pelvic Exenteration (PelvEx) is typically performed for advanced primary and recurrent pelvic malignancies and historically associated with poor outcomes. Less commonly, PelvEx is performed for complications following prior pelvic radiation or surgery. Historically, these fistulas represent the most difficult to repair due to poor wound healing, superinfection and altered anatomy. Using a multi-disciplinary approach, we have developed an algorithmic approach to definitively repair these complex fistulas. We present the largest series of PelvEx for benign pelvic fistulas.

Methods/Interventions: All patients who underwent PelvEx at a tertiary referral center since 2012 were identified using operative note language processing and crossmatched with an existing database. Patients undergoing PelvEx for malignancy were excluded. Two chart reviewers extracted data. Discrepancies were resolved by a third reviewer. Operative variables, complications and short- and long-term outcomes were assessed. Primary outcome was presence of fistula symptoms or urinary tract infection(UTI) at follow-up.

Results/Outcomes: Of 132 PelvEx, 25(19%) were for benign fistulas - median age 68 (range 13-79) and 4 (16%) females. Most common fistulas were rectourethral (15,60%) and involving multiple organs (6,24%). All had previous radiotherapy and 23 (92%) prior pelvic surgery. Most common prior malignancies included prostate (17,68%), rectum (5,20%), and cervix (2,8%).

Median operating time was 502 minutes (IQR 443-555), estimated blood loss 600mL (IQR 500-1000), and 14 (56%) required transfusion. All operations included bladder and rectal resection. 17(68%) included anus resection, 7 (28%) bone, and 9 (36%) required flaps. Colorectal and urology performed all operations. Plastics and orthopedics assisted in 13 (52%) and 7 (28%) respectively.

Seven (28%) patients required intensive care post-operatively. Median length of stay was 10 days (IQR 8-16). Four (16%) had no complications, 12 (48%) were Clavien-Dindo grade I/II, 8 (32%) were III, and one (4%) was IV. Three (12%) underwent unplanned re-operation within 30-days while 8 (32%) had surgery at any point during follow-up. Six (24%) were readmitted within 30-days. There were no postoperative deaths.

Twenty (80%) patients had complete resolution of fistula symptoms/UTI. Of those without resolution, one had recurrent UTIs, one had continued osteomyelitis leading to re-operation without fistula resolution, two had pelvic fluid collections requiring long-term antibiotics, and one had chronic perianal wound drainage. Median follow-up was 159 weeks (IQR 21-259).

Conclusion/Discussion: PelvEx for radiation-induced fistulas is a highly complex operation with historically high morbidity and mortality rates. In this series we found a multi-disciplinary team following an algorithmic approach can provide a safe and definitive repair of these pelvic fistulas from radiation and surgical complications. These operations are associated with high but acceptable complication rates with no mortalities. Further work should be centered on long-term outcomes and quality of life.

Table 1
Operative and post-operative outcomes for patients undergoing pelvic exenteration for benign fistula disease (n=25).