bowel perforation, and progression-free survival (PFS) were analyzed to determine cost and efficacy and establish a favorable incremental cost-effectiveness ratio (ICER) per life-year saved (LYS). Results: For the 240 patients entered into each arm of the OCEANS trial, and with the baseline estimates of PFS (8.6 months for the GC group and 11 months for the GCB+B group, assuming 21% improvement of new treatment) and bowel perforation, the cost of GC was $4.0 million, compared with $36.5 million for GCB + B. These costs resulted in an ICER of $677,250 per LYS for GCB+B compared with GC. If one were to assume a PFS of 15 months (six additional months of benefit) associated with GCB + B, then this results in an ICER of $253,968. Using a maximum ICER threshold of $100,000 per LYS to deem an intervention as cost effective, the cost of B would have to be decreased by 76% or the PFS extended to 25 months for GCB + B. If no additional risk of intestinal perforation is attributed to the addition of B (2% all arms and 25% of them are fatal perforation), the ICER in GCB+B would continue to exceed $665,229 per LYS. Conclusions: In this exploratory analysis of the OCEANS trial, the addition of bevacizumab to combination chemotherapy for the treatment of recurrent ovarian cancer is associated with significant costs with potential benefits. Further investigations are warranted. doi:10.1016/j.ygyno.2010.12.073 67 Intraperitoneal chemotherapy for recurrent ovarian cancer appears efficacious with high completion rates and low complications M. Skaznik-Wikiel, J. Lesnock, W. McBee, S. Taylor, S. Beriwal, A. Smith, K. Zorn, S. Richard, T. Krivak, R. Edwards Magee–Womens Hospital of the UPMC, Pittsburgh, PA Objective: Three prospective phase III clinical trials have shown a benefit of intraperitoneal (IP) chemotherapy as a front-line therapy in the treatment of ovarian cancer. However, little is known about the use of IP chemotherapy in recurrent ovarian cancer. The purpose of this descriptive study was to determine progression-free survival (PFS), overall survival (OS), completion rates, and frequency of complications in patients with epithelial ovarian cancer (EOC) treated with IP chemotherapy for recurrent disease. A retrospective, single-institution analysis of women who re- ceived IP chemotherapy for recurrent EOC between January 2005 and April 2010 was conducted. Study patients were identified from the tumor registry. PFS and OS were estimated by Kaplan–Meier methods. Results: A total of 56 women who received IP chemotherapy for their first EOC recurrence were identified. Their mean age was 56.7 years (range: 43–79). All patients had previously undergone primary surgical cytoreduction followed by intravenous platinum-based chemotherapy. Fifty-four patients (96.4%) had previously completed at least six cycles of IV chemotherapy. All patients were considered platinum sensitive. Among the patients, 80.4% were initially diag- nosed with advanced-stage disease (stage IIIA–IV). All patients underwent secondary cytoreduction at the time of IP port placement; 65.3% were considered optimally cytoreduced (< 1 cm residual disease) at the end of the secondary debulking surgery. Forty-four patients (78.6%) were able to successfully complete at least six cycles of IP chemotherapy. Reasons for noncompletion were: disease progression (five patients), allergic reaction (three patients), elevated creatinine (one patient), pain (one patient), severe nausea and vomiting (one patient), death (one patient), and patient refusal (one patient). Six patients (10.7%) developed port complications: pain around port site (two patients), port malfunction (one patient), pain and port malfunction (two patients), port erosion into small bowel (one patient). Median PFS since the initiation of IP chemotherapy was 12 months (95% CI: 7.5–16.4) and median OS was 51 months (95% CI: 40.8–61.1). Conclusions: Intraperitoneal chemotherapy is an efficacious option for patients with recurrent epithelial ovarian carcinoma, with high completion rates, low frequency of complications and meaningful extension in survival. Further assessment in randomized controlled trials is warranted. doi:10.1016/j.ygyno.2010.12.074 68 Clinical practice guidelines decrease unnecessary Pap tests in survivors of gynecologic malignancies L. Meyer 1 , K. Schmeler 1 , J. Wallbillich 2 , D. Urbauer 1 , P. Soliman 1 , M. Frumovitz 1 , C. Burke 1 , D. Bodurka 1 , C. Levenback 1 1 University of Texas M.D. Anderson Cancer Center, Houston, TX, 2 University of Texas Medical School, Houston, TX Objective: Widespread use of the Pap test has had minimal impact on the early diagnosis of recurrent disease in survivors of gynecologic cancers. The use of clinical practice guidelines (CPGs) is a well- described approach to reduce practice variation, control cost, and improve quality of health care. The objective of this study was to demonstrate the effectiveness of performance improvement strate- gies to alter practice patterns with the aim of decreasing the number of unindicated Pap tests performed for surveillance in gynecologic cancer patients by 50%. A multidisciplinary operations team oversees outpatient care provided by 13 gynecologic oncologists and 20 midlevel providers who perform 13,400 patient visits annually at a single institution. The team identified Pap test utilization as a target for improvement, used a fishbone diagram to identify specific barriers (see figure), approved the new CPGs, and constructed a communication plan and project map. A convenience sample of patient visits to the practice during weeks two and four of August 2009 and August 2010 was used to represent practice patterns before and after the implementation of the CPG. For analysis, the percentage of total and unindicated (based on newly adopted CPGs) Pap tests performed out of all visits during each two-week period were calculated. Fisher's exact test was used for comparison. We estimated 80% power to detect a difference as low as 10.5%. Results: Four hundred sixty-six patient visits occurred in the 2009 study period. One hundred two of the 466 (21.88%) patients had Pap tests performed during the two-week interval in 2009. Unindicated Pap tests were performed at 41 of 466 visits (8.80%). Four hundred thirty-eight patient visits occurred in the 2010 study period. Sixty-six of 438 (15.06%) had Pap tests performed. Unindicated Pap tests were performed at 19 of 438 (4.34%) visits, indicating a 50% reduction. The introduction of the CPGs correlated with a significant decrease in the proportion of total Pap tests performed (OR = 0.63, P =0.01), as well as in the proportion of visits on which an unindicated Pap test was performed (OR=0.47, P =0.01). The cost per test is estimated at $170. By decreasing Pap test utilization from 22 to 15% of visits, 938 fewer tests would be performed annually with an estimated savings of $159,535. Conclusions: The quality improvement process can successfully effect meaningful change in complex clinical settings. We now aim to eliminate all unindicated Pap tests. We are using the same tools to S30 ABSTRACTS / Gynecologic Oncology 120 (2011) S2–S133