Open session Open full presentation page
Oral Presentation

Initial kidney graft function and its association with short- and long-term graft outcomes: a paired kidney analysis

Wai Lim
Presenter
Wai Lim (AU)

Department of Renal Medicine and Transplantation · Sir Charles Gairdner Hospital · Perth

Profile

Authors

Ryan Gately1, Armando Teixeira-Pinto2, Dharshana Sabanayagam2, Henry Pleass3, Esther Ooi4, Farzaneh Boroumand2, Shuvo Bakar2, Lin Zhu2, Jiayue Wang2, Helen Pilmore5, William Mulley6, Scott Campbell1, Germaine Wong2, Wai Lim7

Affiliations

1 Princess Alexandra Hospital
2 University of Sydney
3 Westmead Hospital
4 Edith Cowan University
5 Auckland City Hospital
6 Monash Medical Centre
7 Sir Charles Gairdner Hospital

Abstract

Background: Delayed graft function (DGF) is widely used as a surrogate endpoint in kidney transplant trials, yet its validity for predicting long-term graft survival remains uncertain. Prior studies may be confounded by donor-related factors.
Methods: Using the ANZDATA registry (2000–2023), we analyzed 3,256 kidney transplant recipients from 1,628 paired deceased donors with discordant initial graft function. This paired donor kidney design controls for unmeasured donor characteristics. Outcomes included death-censored (DCGL) and all-cause graft loss (ACGL), comparing immediate (IGF), slow (SGF), and delayed graft function (DGF). Cox models with time-varying coefficients were used to account for non-proportional hazards.

Figure 1
Schematic representation of the paired donor kidney design.

Results: Over a median follow-up of 5.5 years (IQR 2.4–9.7), DGF was associated with a higher risk of DCGL and ACGL within the first 6 months compared with IGF (aHR 3.37 [1.43–7.96] and 1.97 [1.09–3.56], respectively), indicating an early but time-limited effect. No associations were observed between 6 and 12 months. Beyond 12 months, a modest association persisted only for ACGL (aHR 1.35 [1.05–1.72]), with no corresponding association for DCGL. No significant differences in graft loss were observed between SGF and DGF or between SGF and IGF.
Figure 2
Adjusted hazard ratios for death-censored and all-cause graft loss by initial graft function status across follow-up intervals.

Conclusions: These findings indicate that graft loss associated with DGF, independent of donor characteristics, is limited to the early post-transplant period. Future trials should therefore be cautious in adopting DGF as a primary surrogate endpoint for medium- and long-term graft survival.