analysis, only patients responding with a 1 or 2 for all of the items within a subscale were classified as satisfied for the subscale. Results: A total of 47 reconstructive procedures (29 immediate and 18 delayed) were done between April 2007 and Dec 2010. The study popula- tion consisted of 30 LD and 17 Sub-pectoral implant reconstructions dur- ing this period. 28 out of 47 (60%) patient satisfaction questionnaires were returned. Conclusion: Most patients were satisfied with their breast reconstruc- tions and would recommend this procedure to others. However it is impor- tant to assess the reasons for the need to wear a prosthetic shell in 20% of patients. P40. Intra-operative Ultrasound is an Effective Technique for Excision of Early Breast Cancer Jennifer Pollard, Anu Shrotri, Sameer Pathak, Lee Martin Aintree University Hospitals, Liverpool, UK Introduction: Radio-guided occult lesion localisation (ROLL) has been used to excise impalpable breast lesions instead of wire-guided local- isation. We hypothesise that Intra-operative Ultrasound Marking (IUM) is as effective as ROLL for excision of ultrasound-detectable lesions, more cost-effective and more acceptable to the patient. Methods: Retrospective analysis was performed on consecutive pa- tients undergoing Intra-operative Ultrasound Marked WLE over a 3-year period. Baseline demographics, tumour size, margin clearance, histology and further procedures performed were analysed. Results: Sixty-six IUM WLEs were performed from June 2008- No- vember 2011 for non-palpable lesions and we had 100% success rate for excision. The mean patient age was 58 years (range 27-81 years). The mean radiological tumour size was 13.5mm (range 4.5-28.4mm), mean in- vasive tumour size was 18.7mm (range 3-77mm) and mean specimen weight 43.9 grams. 63 patients (95.3%) had clear radial margins (mean ra- dial clearance 3.35mm, range 1.1-7mm). Three patients (4.7%) had posi- tive margins for invasive malignancy (1 lobular, 2 IDC). These results compare favourably with the ROLL data within the literature. Conclusions: Our results demonstrate no malignant lesions were missed; therefore this procedure can be safely used to excise early, impal- pable screen-detected invasive breast cancers. It improves the peri-opera- tive patient journey; increases patient safety and convenience, whilst making savings in the cost of nuclear medicine services. Furthermore, it can be performed in hospitals without a nuclear medicine department. Our recommendations are that breast surgeons will need to undergo train- ing in ultra-sound scanning, which is already the norm amongst their vas- cular and hepato-biliary colleagues. P41. The National Mastectomy and Reconstruction Audit as a driver for local service improvement. Improving peri-operative pain control in breast surgery James Harvey, Sally Hallam, Adam Critchley, Chris Caddy Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK Introduction: The National Mastectomy and Breast Reconstruction Audit (NMBRA) provides comparative data on local and national practice and outcomes. Unacceptably high numbers of patients reported severe pain following breast surgery (6.2%, 16.5% and 20.1% following mastectomy, immediate and delayed reconstruction respectively). The report recommended further work to identify reconstructive procedures with a higher incidence of pain. Our study aimed to identify procedures causing high levels of post-operative pain and to implement service improvement if required. Methods: A prospective audit of all patients undergoing in-patient breast surgery was performed over nine weeks. A multi-disciplinary group including breast and plastic surgeons and the pain team oversaw a local au- dit of pain control and devised a strategy for service improvement. Results of the audit were presented locally to surgical and anaesthetic departments. New guidelines for the management of peri-operative pain were imple- mented. Outcomes were re-audited in September 2011 and February 2012. Results: Levels of severe pain were similar to that demonstrated in the national audit. Procedures causing significantly more post-operative pain included breast reduction and augmentation surgery. Recommendations in- cluded that 95% of patients should be prescribed regular post-operative an- algesia and for 45% of patient to receive a local anaesthetic block. This has led to a significant increase in the use of local anaesthetic blocks (p<0.05) and a decrease in severe post-operative pain. Conclusions: The NMBRA should be used to drive local service im- provement. This model of national data driving local change has led to im- proved outcomes for patients undergoing major breast surgery. P42. Touch imprint cytology in axillary sentinel lymph node biopsy - a series of 1522 cases over 5 years Salli Owen, Simon Pilgrim, David Thurtle, Gurdeep Mannu, Raman Vinayagam, Amy Burger, Simon Pain Norfolk & Norwich University Hospital, Norwich, UK Introduction: Sentinel lymph node biopsy (SLNB) using radioisotope +/- blue dye has been carried out for five years in our unit as an axillary staging procedure for breast cancer. We carry out intra-operative touch im- print cytology (TIC) using the single longitudinal bisection technique fol- lowed by full histological analysis of the sentinel lymph nodes. This study reviews data collected over a 5 year period to determine the sensitivity and specificity of TIC in practice and reports overall rates of SLNB positivity and negativity. Methods: A prospectively-collected SLNB database was cross-refer- enced with computerised operating and histological records over a sixty- one month period between 2005 and 2010. This was then analysed and cor- related with histological results to ascertain the sensitivity and specificity of TIC. Results: SLNB was carried out in 1522 consecutive cases in 1498 pa- tients. Of the 1522 cases intra-operative TIC was performed in 1274 and was positive in 127. Histological examination of the SLNB showed micro- metastases and macrometastases in 337 (22.3%) cases and was negative or containing only isolated tumour cells in 1172 (77.6%) cases. Overall sen- sitivity for TIC was 43.1% and specificity was 99.89%. Conclusion: This analysis shows that TIC is a useful adjunct to SLNB and has a very high specificity, maintained over a long period of time. It remains a safe and effective technique for intra-operative identification of positive sentinel lymph nodes, reducing the need for two stage opera- tions in our patients. P43. The Use of Total Ductal Excision in the Diagnosis of Breast Carcinoma Hannah Kranenburg, Jane Aitken, Adam Stearns, Benedict McCann, Eamonn Coveney, Balendra Kumar West Suffolk Hospital Breast Unit, Bury St. Edmunds, UK Introduction: Nipple discharge is the presenting symptom in 3-9% of patients attending breast clinic (Hunt et al, 2007.) While discharge is largely due to benign processes, the association of pathological nipple discharge with breast carcinoma is thought to be between 10-20% (Dillon et al, 2006.) Consequently, patients with suspicious discharge, in whom no abnor- mality is detected through triple assessment, often undergo a total ductal 1 Would do it again 100% 2 Outcome met expectations 86% 3 Recommend to others 89% 4 Satisfaction about types of reconstructions offered 100% 5 Carrying out normal daily activities 93% 6 Shell required 20% 430 ABSTRACTS