Background: Pancreas transplantation offers insulin independence and improved quality of life for those suitable for the procedure, however transplant rates have decreased internationally in recent years.  Improvements in medical care and pre-assessment techniques have resulted in an increase in the number of older people being referred as potential recipients for pancreas transplantation.  We aimed to investigate the outcomes after pancreas transplantation in our older recipient cohort to establish if offering transplantation to this cohort is justified.
Methods: A prospectively-maintained retrospective database was interrogated.  Data relating to donor and recipient variables, post-transplant cardiac events, graft survival and patient survival were recorded.  The cohort was divided into those <55 years at transplant and those >55 years at transplant and compared.
Results: 444 transplants were performed in patients aged 23-54 years, including 333 simultaneous pancreas kidney transplants (SPK) and 111 isolated pancreas transplants (IP).  83 transplants were performed in patients aged 55- 67 years, including 67 SPK and 16 IP.  The cohort included 59 recipients aged 55-59 years, 19 recipients aged 60-64 years and 5 recipients aged 65-67 years.  Older recipients had slightly higher BMI (26.3 vs 25.3kg/m², p=0.025) and received pancreases from older donors (40.5 vs 35.8 years, 0.004) but were otherwise comparable.
There was no difference in death-censored pancreas graft survival or kidney graft survival between the groups for either SPK or IP transplants.  Rates of post-transplant cardiac events and interventions were low in both groups (CVA 3.8% v 2.3%, MI 1.3% v 0.7%, PCI 1.3% v 0.5%).  Patient survival was inferior in the older age group; however early patient survival was equivalent with divergence evident only beyond 2 years post-transplant (1yr 96% v 95%, 3yr 92% v 87%, 5yr 89% v 77%, 10 yr 78% v 36%; p<0.001; fig 1).  In an adjusted multivariate Cox regression model, risk of mortality was not independently associated with recipient BMI or donor characteristics.  However, increasing age (HR 1.05, CI 1.02-1.07; p<0.001), post-transplant MI (HR 7.25, CI 1.75-30.1; p=0.006), pancreas failure (HR 1.91, CI 1.24-2.96; p=0.003) and kidney failure (HR 3.55, CI 2.14-5.89; p<0.001) were independently associated with death.  Older recipients with good pancreas graft function had higher patient survival compared to those who had suffered graft failure (1yr 99% v 78%, 3yr 91% v 70%, 5yr 83% v 50%, 10yr 40% v 33%; p=0.018).
Conclusion: We describe the outcomes of a large series of pancreas transplants in older recipients in the modern era.  Pancreas and kidney graft survival is comparable to outcomes in younger recipients and rates of cerebrovascular and cardiovascular events are low.  Mortality is higher in older patients, with divergence in survival increasing over time and strongly associated with pancreas and kidney graft failure.  This study suggests that with robust cardiac pre-assessment pancreas transplantation is feasible in older recipients, but careful selection of donor organs to optimise graft survival is important to minimise mortality risk.