Getting the Time Right: What Matters, What Doesn't, and How Should We Really Be Defining Ischemic Times in DCD Withdrawals?

Abstract

Over 100 DCD heart transplants have now been performed worldwide, with excellent early and mid-term results. However, there is a high degree of variability across units with regard to the definitions and cut-offs used for functional warm ischemic time (WIT). This study aimed to evaluate the various timepoints and definitions in a high volume single-centre cohort to determine their relationship to early graft function. Between July 2014 and September 2019, 34 heart transplants from DCD donors were performed at a single-centre. Twenty-five of the 34 donor files from the local database contained complete end-of-life withdrawal data, and were utilised for analysis. Donors were grouped into those in whom the recipient did (n=7) or did not (n=18) require peri-operative ECMO support, indicative of severe primary graft dysfunction (sPGD). Despite concerns regarding the WIT DCD allografts are exposed to during withdrawal of life support (WLS), an extended WIT according to any of the current or proposed definitions (from systolic < 90 mmHg, systolic <50 mmHg or sats <70% to preservation flush; or systolic <50 mmHg to ex-situ reperfusion) was not associated with sPGD in our cohort. If fact a longer duration from WLS to circulatory arrest (CAr), systolic < 90 mmHg to CAr and sats <70% to CAr was associated with a significantly lower rate of sPGD (p=0.02, p=0.04 and p=0.02, respectively). The only period, if prolonged, associated with an increased rate of sPGD is the time between CAr and delivery of the preservation flush, in essence the asystolic WIT (aWIT) (14.6±3.1 and 11.8±2.4 min for the ΕCΜΟ and non-ΕCΜΟ groups respectively, p=0.01). The definition of WIT in DCD heart transplantation has proved controversial, however with more in-depth analysis of the patterns associated with WLS and performance of the donor allograft it does not appear that a prolonged WIT, as defined by any of the current or proposed definitions, is in fact associated with sPGD. At present the only clear negative predictor is a prolonged aWIT. This ought to provoke renewed discussions regarding these definitions, and an increased focus on the minimisation of aWIT during DCD allograft retrievals.