518 EFFECT OF PRE-OPERATIVE HEMATOLOGICAL DISORDERS ON CANCER-SPECIFIC SURVIVAL RATE AFTER RADICAL CYSTECTOMY FOR BLADDER CANCER

e212

THE JOURNAL OF UROLOGY姞

CONCLUSIONS: Patients with advanced clinical stage experience greater BL during cystectomy, which may impact time under anesthesia and transfusions. Patient performance, pre-op labs, and operative factors also influence transfusions. While BL does not appear to affect prognosis after radical cystectomy, this analysis indicates that transfusions may impact oncological outcomes. Source of Funding: None

517 INITIAL EXPERIENCE ASSESSING PRE-OPERATIVE FRAILTY WITH PATIENTS UNDERGOING RADICAL CYSTECTOMY FOR UROTHELIAL CANCER Kenneth Ogan, Maxwell D. Taylor, Caroline G. Tai, Lindsey A. Herrel, Brian Cross, Viraj A. Master, Bradley C. Carthon, Daniel Canter*, Atlanta, GA INTRODUCTION AND OBJECTIVES: Radical cystectomy for the treatment of bladder cancer is associated with significant postoperative morbidity and mortality, yet pre-operative risk stratification is relatively subjective. The decision to operate or not operate on a patient relies on very little objective/quantifiable data. Recently, the concept of frailty as a surgical risk assessment tool to predict postoperative outcomes has been introduced and is beginning to be examined clinically. In this initial study, we prospectively evaluated bladder cancer patients’ pre-operative frailty and attempted to correlate these presurgical measures with post-operative outcomes. METHODS: Patients undergoing radical cystectomy were voluntarily enrolled in this study after IRB approval was obtained. Preoperative frailty data was collected on the following five domains: shrinking, weakness, exhaustion, low physical activity, and walking speed. Patients were classified as either non-frail, intermediately frail, or frail, according to previously published objective standards. 30-day postoperative outcomes were recorded. RESULTS: During our initial 4-month study period, we collected data on 23 patients undergoing radical cystectomy. On average, data collection required only 4-6 minutes. Of this small cohort, 17 (73.9%) patients were white, and13 (56.5%) patients were males. Mean age was 65.83 years (range⫽43-84). Based on the five frailty domains, there were 12 (52.2%) non-frail, 10 (43.5%) intermediately frail, and 1 (4.3%) frail patients, respectively. Overall, there were 17 total complications: 12 (70.6%) Grade I/II, 3 (17.7%) Grade IIIa, 1 (5.9%) Grade IIIb, and 1 (5.9%) Grade V. There was a statistically significant relationship between increasing frailty and decreasing serum albumin (p⫽0.04). However, there was not a statistically significant association between increasing frailty and patients’ post-operative length of stay (p⫽0.2) and incidence of post-operative complications (p⫽0.4). CONCLUSIONS: Our initial experience with assessing preoperative frailty in patients undergoing radical cystectomy demonstrates that is easy and efficient to do in a busy clinical setting. Although our initial data does not show a correlation with post-operative outcomes, this is most likely due to our initial small sample size. Nevertheless, there is a statistically significant inverse relationship between frailty and serum albumin, suggesting that with a larger cohort frailty may predict post-operative outcomes. Further prospective evaluation is warranted and ongoing.

Vol. 189, No. 4S, Supplement, Sunday, May 5, 2013

eters for survival in bladder cancer (BCa) patients treated with radical cystectomy (RC), alterations of pre-operative hematological parameters have not been sufficiently addressed. This study was undertaken to investigate prognostic indicators including clinico-pathological and pre-operative hematological factors, and to develop a novel prognostic factors-based risk stratification model for cancer-specific survival (CSS) in BCa patients treated with RC. METHODS: We performed a retrospective analysis of 249 consecutive BCa patients treated with RC without neoadjuvant therapy at our institution. Prognostic value of the pre-operative hematological laboratory parameters along with clinico-pathological parameters were evaluated by univariate and multivariate Cox proportional hazard model analyses. A new scoring model was developed to predict CSS after RC using the regression coefficients of the multivariate model. RESULTS: Median follow up period was 72 months (range 47-91). One-, 3-, and 5-year CSS rates were 89.1%, 69.3%, and 62.7%, respectively. In multivariate analysis, pre-operative values of low hemoglobin (Hb) (⬍ 10g /dL) and high C-reacted protein (CRP) (⬎ 0.5 mg / dL), pathological T stage (ⱖ pT3a), lympho-vascular invasion (LVI) and positive surgical margin (PSM) status were independent factors for predicting poor prognosis. The score was calculated as 2 (if PSM) ⫹ 2 (Hb ⬍ 10g /dL) ⫹ 1 (if ⱖ pT3a) ⫹ 1 (if LVI) ⫹ 1 (CRP ⬎ 0.5 mg / dL) and 0 (if otherwise). The 4-year CSS in patients with a score in the ranges 0-1 (low-risk), 2-3 (intermediate-risk) and 4-7 (high-risk) was 83.4%, 48.8% and 5.9%, respectively. All differences in CSS between subgroups were P ⬍ 0.0001 (Figure). CONCLUSIONS: This is the first report to demonstrate the significance of pre-operative Hb as an independent predictor for CSS and is also the largest contemporary series to date demonstrating that two types of pre-operative hematological disorders, assessed by Hb and CRP, are independent predictors for CSS in BCa patients treated with RC. Our risk stratification model may provide physicians with useful prognostic information for identifying patients who may be candidates for multimodal treatments.

Source of Funding: None

518 EFFECT OF PRE-OPERATIVE HEMATOLOGICAL DISORDERS ON CANCER-SPECIFIC SURVIVAL RATE AFTER RADICAL CYSTECTOMY FOR BLADDER CANCER Takehiro Sejima*, Hideto Iwamoto, Toshihiko Masago, Shuichi Morizane, Akihisa Yao, Tadahiro Isoyama, Yonago, Japan; Tsutomu Koumi, Matsue, Japan; Atsushi Takenaka, Yonago, Japan INTRODUCTION AND OBJECTIVES: Despite the wealth of evidence on the oncological significance of clinico-pathological param-

Source of Funding: None