The UK National Pancreas Allocation Scheme

A.J. Hudson, S.V. Fuggle, P.J. Friend, C.J.E. Watson
On behalf of the NHSBT UK Pancreas Advisory Group, Bristol, U.K.

Background:There are 8 pancreas transplant units in the UK and historically deceased donor pancreases have been allocated by the nearest centre to patients on their local waiting list. However with the introduction of a national pancreatic islet transplant scheme and a national organ retrieval service, which dissociated organ retrieval from implantation, there was a need to revise the traditional centre-based allocation system. Since 1 December 2010 all deceased donor pancreases (DBD and DCD) have been allocated to named patients listed nationally for either a vascularised pancreas or pancreatic islet transplant.

Methods: Using the UK Transplant Registry and UNOS registry data, factors affecting outcome of pancreas transplantation and a patient’s relative chance of transplant were identified. Applying simulation modelling techniques, these factors were used to develop a computer-based points scoring system to prioritise patients for a donor pancreas. This scoring system gives priority to patients with prolonged waiting times, higher levels of sensitisation (based on a standardised calculated reaction frequency), dialysis requirements, listed at a centre close to the donor hospital (to assist in minimising cold ischaemia times), that are not poorly HLA mismatched and, to a lesser extent, to patients of a similar age to the donor. Donor body mass index (BMI kg/m²) is also used to help decide whether a pancreas should be preferentially allocated to a patient listed for a vascularised pancreas (low BMI donors) or a pancreatic islet graft (higher BMI donors).

Conclusion: On 1 December 2010 the UK introduced a new patient based National Pancreas Allocation Scheme. The scheme allocates all pancreases from both DBD and DCD donors and prioritises patients listed for vascularised pancreas or pancreatic islet transplantation using a single computer-based matching algorithm.