Donor diabetes and prolonged cold ischemia time increase the risk of graft failure after liver transplant: Should we need a redefinition of the donor risk index?

Dear Editors, We read with great interest the paper by Brüggenwirth et al. (1) about the importance of cold ischemia time (CIT) and diabetes type II (DM-2) in increasing the risk of graft failure after liver transplantation (LT). The results obtained from the UNOS database are in line with those obtained in a recently published study coming from our center. (2) Our retrospective, single-center analysis was based on data from 1,354 adult LTs performed at the University of Pisa Medical School Hospital: in all the cases, whole sized livers coming from deceased-brain donors were transplanted. Using a propensity score approach, 448 patients receiving a graft younger than 70 years were finally matched with 515 counterparts receiving grafts older than 70 years. Four variables were found to be independently significant as risk factors for graft loss, namely HCV positivity (HR=2.1; p<0.001), donor age (HR=1.0; 95% p<0.001), CIT (HR=1.0; p=0.042), and donor DM-2 status (HR=1.5; p=0.047). It is extremely interesting to underline that two apparently very different databases like a North American and an Italian one consented to obtain similar results, mainly in consideration of their big numerosity (58,226 and 1,354 respectively). In both the contexts, the synergic action of acute (prolonged CIT) and chronic (DM-2) damages eventually ended in promoting post-LT graft failure. Indeed, the fact that advanced donor age is an amplifying factor of chronic damages induced by DM-2 looks to be as a natural consequence. Also Brüggenwirth et al. reported that “the only risk factor that meaningfully altered the HR for DM-2 (and remained statistically significant) in the final models was donor age”.(1) Here comes the problem connected with the vagueness of the definition of donor DM-2. In fact, defining DM-2 status as “use of insulin” or “altered blood sugar levels” is not enough for completely capturing the real pathological changes induced by the disease. For example, using these dichotomous variables completely fails in defining the length and the severity of DM-2: thus, age is probably only a very good surrogate for this purpose. For this reason, identification of pathological markers (3) consenting to pre-operatively define the grafts at high-risk for poor post-LT function should represent a real revolution in the selection and allocation processes. We strongly believe that not age per se, but a combination of acute and chronic damages associated with age, mainly due to metabolic diseases, play a fundamental role in worsening results when using older grafts. Under the light of these experiences, and in agreement with other Authors,(4) a re-evaluation of the Donor Risk Index (DRI) should be considered. In fact, after the publication by Feng et al. in 2006,(5) a larger use of the so called extended criteria donors has been observed, in particular of donors with multiple comorbidities. As a consequence, a new analysis able to update the donor-related risk stratification should be considered. We believe that a new universal DRI should be the future target, including international experiences, and not being limited only to regional databases which may not be able to intercept specific behaviors or needs of a particular region of the world.