WH LIM1, D JOHNSON2, G WONG3
1Sir Charles Gairdner Hospital, Perth; 2Princess Alexandra Hospital, Queensland; 3Westmead Hospital, New South Wales
Aim: To examine the association between duration of delayed graft function (DGF) and long-term graft outcome.
Background: Although a short duration of DGF may have no impact on graft outcome, the threshold of DGF associated with an adverse effect on out- come remains poorly defined.
Methods: In Australian and New Zealand, primary deceased donor kidney transplant recipients between 1994-2014 who had experienced DGF requiring dialysis were included. The associations between DGF duration and death censored graft loss (DCGL) and acute rejection at 6 months were examined using adjusted Cox regression analysis.
Results: Of 1718 kidney transplant recipients followed for a median of 5.4 years, 618 (36.0%) required between 3-7 days of dialysis post-transplant (referent), compared to 200 (11.6%), 459 (26.7%) and 441 (25.7%) who had required 1-2 days, 8-14 days and >14 days of dialysis, respectively. DGF duration of 8-14 and >14 days were associated with adjusted hazard ratios (HR) for acute rejection of 1.30 (95%CI 1.01, 1.67) and 1.93 (95%CI 1.52, 2.45), respectively. Compared to DGF duration of 3-7 days, DGF duration of >14 days was associated with adjusted HR for DCGL between 0-1 year of 3.00 (95%CI 1.59, 5.68), 2.14 (95%CI 1.35, 3.41) between 1-5 years, and 0.75 (95%CI 0.42, 1.31) between 5-10 years, independent of rejection. There was no association between DGF of 1-2 or 8-14 days and DCGL.
Conclusions: DGF of more than 14 days strongly increases the risk of DCGL independent of rejection. However, this association was time-dependent, with the excess risk of DCGL confined to within the first 5 years post-transplant. Future research evaluating interventions that aim to reduce DGF duration in susceptible kidney transplant recipients is urgently required.