Irish men under investigation for PCa. 0187: SINGLE CENTRE EXPERIENCE FROM THE ACCURACY OF CONVENTIONAL 1.5 T MAGNETIC RESONANCE IMAGING FOR THE DIAGNOSIS AND PRE-OPERATIVE STAGING OF PROSTATE CANCER

s / International Journal of Surgery 23 (2015) S15eS134 S121 Conclusion: The novel model outperformed the PCPT risk calculator in all 5 patient cohorts and can be used to improve clinical decision making in Irish men under investigation for PCa. 0187: SINGLE CENTRE EXPERIENCE FROM THE ACCURACY OF CONVENTIONAL 1.5 T MAGNETIC RESONANCE IMAGING FOR THE DIAGNOSIS AND PRE-OPERATIVE STAGING OF PROSTATE CANCER H. Raja, D. Moschonas, N. Haldar, K. Misra, J. Kelleher. Buckinghamshire Hospitals NHS Trust, UK Aim: Not all trusts in the UK have 3T MRI scanners. Are patients with suspected prostate cancer therefore adequately staged with a more conventional 1.5T MRI scanner? The aim of the study was to evaluate the accuracy of standard 1.5T MRI without endorectal coil for the staging of prostate cancer compared against the histopathological results following radical prostatectomy. Methods: Prospective study enrolled 65 patients with biopsy-proven PCa who underwent MRI prior to undergoing radical prostatectomy from December 2012 to June 2014. Results: Median times for patients having an MRI scan after TRUS biopsy were 32.5 days. Preoperatively 38.5% of patients were diagnosed with T3/ T4 disease with MRI. The prevalence of pT3 and above prostate cancer confirmed histologically was 22/65 (33.84%). Postoperative down staging was documented in 12/65 (18.46%) while upstaging in 9/65 (13.85%). The sensitivity of MRI diagnosing T3 disease was 14/22 (63.63%), specificity 29/ 43 (67.44%), accuracy 43/65 (66.15%) Conclusion: Standard MRI detected 63.63% of all patients with T3 prostate cancer, a crucial feature when planning definitive treatment. The promising detection rate of our study, while being operator dependent, offers a cost effective imaging modality for preoperative staging of prostate cancer. 0188: A COMPARISON OF LEARNING CURVES IN ROBOTIC PROSTATECTOMY WITHIN A CHANGING REFERRAL PRACTICE A. Srinivasan , N.C. Shah , D.E. Neal , V.J. Gnanapragasam . Cambridge University Hospitals Trust, UK; University of Cambridge, UK Aim: Increased numbers of high-risk patients with advanced disease are being treated with robotic prostatectomy (RP). We have explored the impact of this on the learning curves and pathological outcomes of three surgeons in a single RP tertiary unit. Methods: Data was sourced from a prospective database and a previous publication from our unit. Three periods corresponding to the first and second 50 cases of three surgeons (A: 2005-2006, B: 2008-2009, C: >2010) were interrogated. One surgeon, (A), had prior extensive open RP experience. Results: Low-risk disease represented 55%, 33% and 25% of referrals in periods A, B and C respectively while high-risk referrals accounted for 10%, 15% and 24% (p1⁄40.0001). Analysing cases 1-50 demonstrated declining numbers of pT2 cases (A:74%, B:60% and C:46%) but an increase in pT3a cases (A: 22%, B:34% and C:48%) and final tumour volumes (A:9%, B:8% and C:22%) (p1⁄40.002). Comparing cases 51-100 revealed identical trends. Positive surgical margin (PSM) rates for cases 1-50 were A:12%, B:20% and C:23% (p1⁄40.12). For cases 51-100 there was again no significant difference in PSM rates. Conclusion: There is an increasing risk profile amongst referred patients which has however had minimal impact on pathological outcomes during temporally different learning curves. 0237: PRESENCE OF DETRUSOR MUSCLE IN BLADDER TUMOUR RESECTION: THE ROLE OF OPERATOR EXPERIENCE S.M. Lee, A. Russell, G. Hellawell. Northwick Park Hospital, UK Aim: Detrusor muscle (DM) in transurethral resection of bladder tumour (TURBT) specimens is a surrogate maker for resection quality. Studies have suggested a correlation with the level of operating surgeon. We audited tumour and surgical factors related to detrusor muscle sampling. Methods: Retrospective review of TURBT cases from 01/2011-01/2014 in a district hospital. Demographics, tumour and surgical characteristics were collected and analysed. Results: 179 patients were included (mean age 71.2±12.2 years, male:female 141:38). Senior surgeons performed 129 (72.1%) resections. All TURBTs were performed on dedicated lists with a senior surgeon supervising. DM was present in 140 (78.2%) samples. Within the cohort, there was no significant difference in recurrence rate when DMwas not sampled (OR 1.315, 95%CI 0.617-2.805, p1⁄40.479). Demographics for junior/senior surgeons significantly differed for ASA Grade, tumourmultiplicity and location. No difference in DM resectionwas identified between junior/senior surgeons. Univariate analysis identified that DM was less likely to be sampled in high grade (G3) tumours (versus G1/G2, OR 0.345,p1⁄40.018*). Conclusion: In contrast to previous studies, DM resection rates were not influenced by surgeon level. Impact of operator experience may be mitigated through the use of dedicated TURBT lists and direct senior supervision, although further investigation is needed. 0239: VALIDATION OF INFLAMMATION-BASED PROGNOSTIC SCORES AND HAEMATOLOGICAL PARAMETERS TO PREDICT BLADDER CANCER INVASION S.M. Lee, A. Russell, G. Hellawell. Northwick Park Hospital, UK Aim: There is increasing evidence that inflammation plays a role in cancer development. We aimed to compare the predictive value of inflammationbased prognostic scores and haematological values in differentiating nonmuscle invasive (NMIBC) and muscle invasive (MIBC) bladder cancer. Methods: Bladder cancer cases from 01/2011-12/2013 were analysed retrospectively. Patient/tumour characteristics, prognostic scores (Neutrophil-to-lymphocyte ratio (NLR), derived NLR (dNLR) and plateletto-lymphocyte ratio (PLR)) and haematology results were analysed. Results: 227 patients were included: 176 and 51 in the NMIBC and MIBC groups, respectively. Groups were significantly different (p<0.05) with regards to age, tumour grade, size, NLR, dNLR, PLR, albumin and white cell, neutrophil, lymphocyte and platelet counts. Of the prognostic scores, dNLR (threshold>2.32) had the greatest area under the ROC curve of 0.725 (Sensitivity 56.9%, Specificity 79.4%) as compared to NLR (0.710) and PLR (0.642) and was included in multivariate analysis. Mean dNLR was 1.86±0.85 in NMIBC and 2.95±2.04 in MIBC. Multivariate logistic regression analysis identified tumour grade (OR 19.968, 95%CI 5.008-79.614, p<.0001*) and dNLR (OR 2.260, 95%CI 1.0324.947, p1⁄40.041*) as independent predictors of MIBC. Conclusion: Our comparison indicates that dNLR may provide a simple, cost-effective marker for MIBC that can be performed at time of cystoscopy, thereby assisting in treatment planning. 0259: HOSPITAL ADMISSIONS WITH SEPSIS AFTER TRUS-GUIDED PROSTATE BIOPSY: ITS INCIDENCE AND BACTERIAL CHARACTERISTICS IN BOLTON, UK Z. Zhou, K. Edwards, M. Pantelides. Royal Bolton Hospital, UK Aim: Transrectal ultrasound (TRUS)-guided prostate needle biopsy is a diagnostic procedure for prostate cancer. We aim to investigate incidence and bacterial characteristics of post-biopsy sepsis. Methods: A retrospective analysis of medical records was performed in 1105 patients that underwent a 12-core TRUS-guided prostate biopsy at our hospital over a five-year period from January 2009 to December 2013. Cases in which patients admitted with sepsis within 14 days after the biopsy were investigated. Results: Overall, 20 out of 1105 patients (1.81%) were admitted with sepsis within 7±3.68 days after the biopsy. All patients received ciprofloxacin prophylaxis for 3 days. 15 patients (1.36%) had positive microbial culture. Escherichia coli (E. coli) grew in samples from 9 patients (60%), Enterococcus species in 2 patients (13.3%), non-E. coli coliforms in 5 patients (33.3%).One patient had different bacteria cultured from urine and blood samples. Fluoroquinolone-resistant bacteria were confirmed in 10 patients (62.5%). Resistance to trimethoprim, co-amoxiclav and tazocin was 43.8%,