Aims of the study: Organ donors on intensive care may develop hyperglycaemia, which is managed with insulin. It is unclear to what extent hyperglycaemia is caused by reversible insulin resistance (perhaps by catecholamines, inflammation and steroids) and how much is generated by beta-cell death.. We hypothesised that donor insulin use (DIU) is a marker of pancreatic beta-cell death and aimed to assess relationships of DIU to subsequent pancreas and islet transplant outcome and function.
Methods: Data on organ donors from the UK Transplant Registry was linked with regional data from our local solid organ pancreas transplant (SPT) programme (2010-2015) and national data from the UK Islet Transplant Consortium to determine associations between DIU and graft function and graft survival (2008-2016). In SPT recipients, the outcome non-technical graft failure (transplant pancreatitis) within 3 months was assessed from histology reports and in a sub-group of SPT recipients, we determined the relationship between DIU and c-peptide secretion within 72 hours post-transplant. In first-time islet transplant recipients (IT), graft failure was defined as a 3-month stimulated c-peptide <50 pmol/L.
Results: In 168 SPT recipients,graft loss due to non-technical failure was more commonly associated with DIU (proportion failing: with vs. without insulin: 6/72 [8.3%] vs 1/96 [1.0%]  p=0.020) and in the sub-group with c-peptide data (n=46), graft function 72-hours post-SPT was lower in transplants from donors requiring insulin (mean (SD) random c-peptide; insulin vs no insulin donors: 1431 (1117) vs. 2496 (1702) pmol/L, p<0.001).
In 91 IT recipients, graft function was worse in IT recipients from donors requiring insulin (mean (SD) HbA1C; insulin vs no insulin donors:51 (14) vs. 45 (9) mmol/mol, p=0.04; mean (SD) 90-minute stimulated plasma glucose: 15.9 (6.3) vs. 12.4 (4.0) mmol/l, p=0.006; mean  (SD)  90-minute stimulated c-peptide: 633 (505) vs 488 (461)pmol/L, p=0.121). There was a suggestion that islet graft failure was more commonly associated with DIU but this did not reach statistical significance (proportion failing: insulin vs. no insulin: 5/50 [10.0%] vs 1/41 [2.4%] p=0.17).
Discussion: In pancreas and islet transplant recipients, DIU in intensive care predicts early graft function in IT and SPT recipients and non-technical graft failure in SPT recipients.DIU may be a marker of beta cell death and could improve donor selection leading to better transplant outcomes. Larger studies are needed to confirm these novel findings which could have important clinical implications.