The Colorectal Disease Poster Prize Abstract 2016

Posted 28 July 2016 in

Congratulations to the winning poster: Variation in liver resection rates in patients presenting with colorectal cancer and synchronous liver metastases.

Abigail Vallance1 ,4, Angela Kuryba1, Kate Walker1, Ian Botterill2, James Hill3, David Jayne2 ,4, Jan vanderMeulen1

1Clinical Effectiveness Unit, Royal College of Surgeons of England, London, UK, 2St James’s University Hospital, Leeds, UK, 3Manchester Royal Infirmary, Manchester, UK, 4University of Leeds, Leeds, UK


In patients with liver metastases (LM) arising from colorectal cancer (CRC), liver resection (LR) is the only modality of care associated with cure. This study investigates: i) factors associated with LR in patients presenting with synchronous CRC and LM, ii) regional variation in LR rates.


Data from the National Bowel Cancer Audit of patients diagnosed with primary CRC between 2010 and 2014 who underwent bowel resection in English NHS hospitals was linked to Hospital Episode Statistics data. Multivariable logistic regression was used to identify determinants for LR. Cox proportional hazard models were used to study survival.


2,248 of the 6,134 (36.6%) patients presenting with CRC and LM who underwent a bowel resection, had a LR. The proportion of patients undergoing LR increased from 33.9% in 2010 to 41.33% in 2014 (p=0.001). Older patients, those with co-morbidities, nodal disease, emergency presentation, or those diagnosed at trusts with no hepatobiliary MDT were less likely to undergo LR. After adjusting for patient and colorectal tumour characteristics, there was significant variation in LR rates at regional (from 32.45 to 48.1%) and trust (from 14.3 to 70.2%) level. With adjustment for patient and colorectal tumour characteristics, mortality was found to be 3.8-fold (95% CI 3.5 to 4.2) higher in those who did not have LR.


There is a large regional variation in the surgical management of these patients in England. An on-site liver resection service is important in determining whether a patient presenting with CRC and synchronous LM undergoes a LR.