259Rectosigmoid resection for ovarian cancer cytoreduction whencoupled with additional large bowel resection is associated withincreased risk of anastomotic leakE. Kalogera1, S. Dowdy1, A. Mariani1, A. Giovanni2, J. Bakkum-Gamez1,W. Cliby1. 1Mayo Clinic, Rochester, MN, 2European Institute of Oncology,Milan, Italy.

Objective: Anastomotic leak (AL) is one of the most fearedcomplications after large bowel resection (LBR) for ovarian cancer(OC) cytoreduction. Our aim was to investigate whether multiple LBRis associated with AL after OC cytoreduction and to determine theshort- and long-term outcomes after AL.Methods: AL cases after LBR between January 1994 and June 2011 wereidentified. Caseswerematched1:2with controls (LBR forOCwithout AL)on stage (I; II; IIIA/IIIB; IIIC; IV), date of surgery (+/−4 years), and date ofbirth (+/−5 years). Relevant clinical risk factors, use of protectivestomas, and outcomes (length of stay (LOS), ability to start chemother-apy, time to chemotherapy (TTC), 90-day mortality, and overall survival(OS)) were abstracted. A stratified conditional logistic regression modelwas fit to determine the association between each factor and AL. Astratified Cox model was fit to compare OS between cases and controlswhile censoring the follow-up at the earliest follow-up for the membersin each case–control set.Results: Forty-two AL cases met inclusion criteria. In the case–controlanalysis, we observed that rectosigmoid resection coupled withadditional LBR was associated with increased risk of AL (38% vs. 20%,OR 2.45, 95% CI 1.05–5.72, P=0.039) whereas protective stomas wereassociatedwith decreased risk (0% vs. 9.5%, P=0.042). Therewas a trendtoward a higher proportion of patients with history of abdominal/pelvicsurgery in cases vs. controls (69% vs. 56%, OR 1.78, 95% CI 0.81–3.93,P=0.15). With regard to outcomes, cases had longer LOS (median: 19.5vs. 9.5 days, Pb0.001), were less likely to start chemotherapy (71.4% vs.91.9%, P=0.047), less likely to start chemotherapy within 30 days aftersurgery (12.2% vs. 35.2%, P=0.027) and had longer TTC (median 34.5 vs.47.5 days, P=0.009). Cases tended to have higher 90-day mortality(19.1% vs. 7.1%, P=0.061). Casesweremore likely to have poorer survival(HR=2.50, 95% CI 1.30–4.82, p=0.006).Conclusions: AL after OC cytoreduction significantly delayed che-motherapy and negatively impacted short- and long-term outcomesand overall survival. In addition to preoperatively identified poornutrition, rectosigmoid resection coupled with an additional LBRappears to confer an increased risk of AL after OC cytoreduction.Surgeons should strongly consider performing bowel diversion in thisgroup of patients given the catastrophic short- and long-termconsequences of AL.

doi:10.1016/j.ygyno.2011.12.260

260Evaluation of Society of Gynecologic Oncologists (SGO) qualitysurgical measuresR. Gogoi1, R. Urban2, H. Sun3, B. Goff2. 1Geisinger Medical Center,Wynnewood, PA, 2University of Washington Medical Center, Seattle, WA,3Geisinger Medical System, Wynnewood, PA.

Objective: The Society of Gynecologic Oncologists has recentlydeveloped two measures to assess and improve the surgical care ofpatients with ovarian cancer — (1) description of residual diseasefollowing cytoreduction and (2) adequate surgical staging. Our aimwas to establish baseline surgeon compliance with these measures attwo tertiary medical centers.Methods: A retrospective review of patients with ovarian,fallopian tube or peritoneal cancer undergoing surgery between7/1/2006 and 7/1/2011 for the purposes of staging and/orcytoreduction was performed at the University of Washingtonand Geisinger Medical Center. Operative and pathology reportswere reviewed to obtain information pertaining to stage, histology,residual disease after surgery and the extent of surgical staging.Adequate staging was defined as including washings, omentect-omy and pelvic and paraaortic lymphadenectomy. Operativereports were reviewed to determine whether the extent ofresidual disease was adequately described. Χ2 and Fisher's exactt-test were used as appropriate.Results: Five hundred and thirty-five patients between 2006 and2011 underwent surgery — 91% with ovarian, 6% with fallopian tubeand 3% with peritoneal cancer. 61% of patients had at least stage IIICdisease; 15% had recurrent disease and 16% had neoadjuvant therapy.For patients with stage I–IIIB disease, 74% had full surgical staging,10% did not have full surgical staging but documented the reason forthis in the operative report; 15% did not have full surgical staging, noreason was noted. Twenty-five percent of all operative reports lackeddocumentation of residual disease with 40% documenting no grossresidual disease, 18% with residual disease b1 cm and 18% hadsuboptimal debulking with N1 cm disease remaining. There was astatistically significant increase in appropriate documentation ofamount of residual disease over time (pb0.001).Conclusions: Our study sets benchmarks for evaluation of surgicaloutcomes in centers where gynecologic oncology care is provided.Improved documentation and staging will allow for the implemen-tation of the SGO quality surgical measures across institutions.

doi:10.1016/j.ygyno.2011.12.261

No visibleresidualdisease

Residualdisease≤1 cm

ResidualdiseaseN1 cm

Residual diseasenot documented

p-value

Stage I–IIIB(n=207)

107 (52%) 16 (8%) 19 (9%) 65 (31%) b0.0001

Stage IIIC–IV(n=328)

104 (32%) 79 (24%) 79 (24%) 66 (20%)

Full stagingperformed

Full staging notperformed andclinical reasondocumented

Full stagingnot performed,no reasonsdocumented

Stage I–IIIB(n=125) 93 (74%) 13 (10%) 19 (15%)

Abstracts / Gynecologic Oncology 125 (2012) S3–S167 S109