OC-097 An externally validated nomogram to predict the risk of bowel dysfunction following an anterior resection

Abstract

Introduction Almost half of the patients undergoing an anterior resection for rectal cancer report quality of life impairment due to bowel dysfunction. The Low Anterior Resection Syndrome (LARS) score is an internationally validated patient reported outcome measure (PROM) designed to assess the severity of these symptoms. 1 The aims of this study were to (i) produce a LARS score prediction model based on the risk factors for bowel dysfunction (ii) to validate the model externally and (iii) to incorporate findings into a nomogram for use in clinical practice. Method Patients who were recurrence free and more than one-year post restorative anterior resection (UK, 2001–2011; Denmark, 2001–2007 (median 54 and 56 months since surgery respectively), were invited to complete EORTC QLQ-C30, LARS and Wexner incontinence scores. Demographics, tumour characteristics, pre/post-operative treatment, and surgical procedures were recorded. Advanced linear regression shrinkage techniques (i.e. LASSO) were used independently for both datasets (UK, DK) for an unbiased selection of the most influential factors. Finally a nomogram based on linear regression was generated. Results The model was developed from 463 British patients and validated in 938 Danes. The respective cohorts were similar in gender (females 40% [184/463] versus 43% [402/938] NS) but differed by age (65 [13] versus 64 [13], p = 0.013), pre-operative radiotherapy (31.6% [145/463] versus 20.4% [191/938], p = 0.001), tumour height (median [IQR] 9 cm [4] versus 11 [4], p = 0.001), global quality of life (mean (SD) QLQ C30 Scale, 77 (19) versus 78 (21), p = 0.001) and LARS score (mean (SD) 26.1 (11) versus 24.5 (12), p = 0.012). Tumour height (from the anal verge) and pre-operative radiotherapy were the key variables in the nomogram with several additional factors including age, gender and temporary stoma influencing the predicted LARS score. Conclusion This is the first nomogram to predict patient reported bowel dysfunction following an anterior resection. Despite differences between cohorts, with the UK tending to restore lower tumours and use more pre-operative radiotherapy, we were able to externally validate the model in a Danish population. Colorectal Surgeons and Nurse Specialists may use this predictive tool to help patients understand their risk of bowel dysfunction, it is intended to aid the consent process and highlight patients that may need additional post-operative support. Disclosure of interest None Declared. Reference Juul T, Ahlberg M, Biondo S, et al . International validation of the low anterior resection syndrome (LARS) score. Ann Surg. 2014;259(4):728–34