Dame, Montreal, QC, Canada, e University of Alberta, Edmonton, AB, Canada, f Princess Margaret Cancer Centre, Univeristy Health Network, Toronto, ON, Canada, g Princess Margaret Cancer Centre, University of Toronto, Toronto, ON, Canada Objective: To compare the immediate operating room), inpatient, and overall costs between three surgical modalities—laparotomy, laparoscopy, and robotics—among patients with endometrial cancer. The differences in costs between the three surgical modalities were also compared among patients with a BMI b40 and ≥40. Method: A multicenter prospective observational study examined outcomes of women, with early-stage endometrial cancer, treated surgically. Costs were captured from a hospital perspective (reported in Canadian dollars). Resource use was collected for operating room costs including operating room time and equipment costs (surgical materials, medications, trochars, disposable equipment, and nursing time) and for inpatient costs (nursing time, imaging, bloodwork, transfusions, medications, disposable materials). All comparisons of continuous variables across cohorts were analyzed using an independent-samples Kruskal-Wallis test (data were tested for normality assumptions using a Kolmogorov–Smirnov test). Results: The study sample included 520 patients, of whom 513, 488, and 482 had sufficient data to be included in the operating room, inpatient, and overall cost analyses, respectively (Table 1). The operating room costs were lowest for laparotomy and then laparoscopy and highest for robotics (median = $4,266.38 vs $5,209.27 vs $7,422.22, respectively, P b 0.001). The inpatient costs were higher for laparotomy and laparoscopy compared to robotics ($5,881.61 vs $1,782.60 vs $1,726.31, respectively, P b 0.001). The overall costs were highest for laparotomy, lowest for laparoscopy, with robotics in between ($10,298.20 vs $,7426.04 vs $9,162.20, respectively, P b 0.001). For women with BMI ≥40, total costs were not statistically different among laparotomy, laparoscopy, and robotics ($10,291.50 vs $8,412.63 vs $,9002.48, P = 0.185). Conclusion: Among all patients, the overall cost was highest for laparotomy and lowest for laparoscopy with robotics between the two groups. Among patients with a BMI ≥40, there was no difference in cost between the 3 surgical modalities. doi:10.1016/j.ygyno.2019.04.516 2330 - Poster Session Predictors of 30-, 60-, and 90-day hospital readmissions after ovarian cancer cytoreductive surgery: A nationwide readmissions database study M.B. Wilbur a , B.J. Borah b , J.P. Moriarty b , R. Rojas b , S.C. Dowdy b , R.L. Stone a , A.K. Sinno c , S. Wethington d , A.N. Fader a . a Johns Hopkins Hospital, Baltimore, MD, USA, b Mayo Clinic, Rochester, MN, USA, c David Geffen School of Medicine at UCLA, Los Angeles, CA, USA, d Johns Hopkins School of Medicine, Baltimore, MD, USA Objective: Patients with ovarian cancer (OC) have multiple comorbid conditions, undergo radical procedures, and are at high risk of readmission after cytoreductive surgery (CRS). Prior studies are limited in the ability to capture readmissions outside of the index hospital. Our study aim was to evaluate predictors of readmission for OC patients undergoing CRS on a national scale. Method: The Nationwide Readmissions Database from the Healthcare Cost and Utilization Project was queried to identify all OC patients undergoing CRS during the period 2010–2015. CRS procedures were categorized using ICD-9 codes as simple pelvic (SP, removal of reproductive organs, lymph nodes, omentectomy), extensive pelvic (EP, addition of bowel or bladder resection), or extensive upper abdominal (EUA, addition of diaphragm stripping, splenectomy, pancreatectomy, and/or liver resection). Thirty-, 60-, and 90-day readmissions were identified. Multivariate logistic regressions were used to identify independent predictors for readmission. Results: A total of 61,833 patients (weighted) were identified, and 20.6% experienced a 30-, 60- or 90-day hospital readmission. Most readmissions (55%) occurred during the first 30 days. Teaching hospitals were more likely to perform EP or EUA CRS (P b 0.001), and extent of CRS was adversely correlated with 30-, 60, and 90-day readmission rates (P b 0.001). On multivariate analysis, predictors of 30-day readmission included length of hospital stay (OR = 1.04), Elixhauser comorbid conditions N2 (OR = 1.39), and Medicaid (OR = 1.30) or Medicare (1.26) payer status (ref: private insurance), whereas SP CRS (ref) was associated with a lower risk of readmission compared to EUA CRS (OR = 0.62). These trends were almost identical for 60- and 90-day readmissions. Conclusion: Thirty-, 60-, and 90-day hospital readmission rates after OC CRS were high, especially for patients undergoing radical procedures. The true readmission rate after OC CRS is 20.6%, capturing index and nonindex hospital readmissions and both short- and long-term outcomes. Teaching hospitals performed significantly more radical CRS and had more readmissions than nonteaching hospitals, indicating that readmis- sion may not be an appropriate marker for quality improvement without proper risk adjustment. doi:10.1016/j.ygyno.2019.04.517 2331 - Poster Session Brain metastases in patients with gynecologic cancers: National trends in incidence and management A.I. Tergas a , L. Chen a , C.M. St. Clair a , J.Y. Hou a , F. Khoury Collado a , A. Neugut b , D.L. Hershman a , J.D. Wright a,b,c . a New York-Presbyterian/ Table 1 Surgical, inpatient, and overall costs compared by surgical group among all patients, patients with BMIb40, and patients with BMI ≥40 CostCAN$ Laparotomy Laparoscopy Median (95.05% Lower & Upper CL) n Median (95.05% Lower & Upper CL) n All patients Surgical 4266.38(4110.43, 4487.63) 102 5209.27(5023.79, 5467.13) 167 Inpatient 5881.61(4897.79, 6147.61) 100 1782.60(1695.47, 1916.00) 153 Overall 10298.20(9538.78, 10910.79) 99 7426.04(7057.05, 7863.12) 151 BMIb40 Surgical 4388.28(4093.22, 4737.40) 77 5199.31(4973.59, 5461.00) 152 Inpatient 5850.55(4897.79, 6164.39) 75 1774.55(1682.11, 1858.75) 143 Overall 10241.31(9376.00, 10942.44) 75 7249.73(6891.89, 7780.25) 141 BMI≥40 Surgical 4197.02(3677.86, 4487.63) 22 5524.63(4775.38, 6816.41) 13 Inpatient 5584.28(4444.31, 8389.55) 22 3042.07(1783.35, 3109.85) 10 Overall 10291.50(8536.98,15062.97) 22 8412.63(7891.04, 10798.15) 10 CostCAN$ Robotics p-value Median (95.05% Lower & Upper CL) n All patients Surgical 7422.22(7225.16, 7623.37) 244 b0.001 Inpatient 1726.31(1689.93, 1790.67) 235 b0.001 Overall 9162.20(8982.98, 9505.68) 232 b0.001 BMIb40 Surgical 7500.39(7312.14, 7696.10) 177 b0.001 Inpatient 1696.00(1641.53, 1753.67) 172 b0.001 Overall 9181.23(9025.07, 9664.97) 169 b0.001 BMI≥40 Surgical 7225.16(6814.52, 7642.22) 67 b0.001 Inpatient 1794.51(1742.09, 1923.35) 65 b0.001 Overall 9002.48(8597.72, 9762.23) 64 b0.001 0090-8258/$ – see front matter Abstracts / Gynecologic Oncology 154 (2019) 2–288 223