Background.
The association between Kidney Donor Profile Index (KDPI) and 1-y estimated glomerular filtration rate (eGFR) with long-term kidney graft survival is well known. Yet, the association between KDPI and 1-y eGFR remains uncertain considering the several concurrent competing risk factors.
Methods.
This single-center, retrospective cohort study analyzed data from 3059 consecutive deceased donor kidney transplant recipients with a 1-y follow-up from January 2013 to December 2017. The aim was to determine the association between the KDPI strata (0%–35%, 36%–50%, 51%–85%, 86%–100%) and 1-y eGFR estimated by the CKD-EPI equation.
Results.
The incidence of delayed graft function (50.6% versus 59.3% versus 62.7% versus 62.0%;
P
< 0.001) and cytomegalovirus infection (36.7% versus 36.6% versus 43.3% versus 57.8%;
P
< 0.001) increased with increasing KDPI strata but not biopsy-proven acute rejection (9.1% versus 9.8% versus 8.4% versus 9.1%;
P
= 0.736). The median 1-y eGFR decreased with increasing KDPI strata (64.8 versus 53.5 versus 46.9 versus 39.1 mL/min/1.73 m
2
;
P
< 0.001). In the Cox regression, the higher the KDPI was, the lower the probability of a lower 1-y eGFR was. Assuming the 0%–35% strata as the reference, the likelihood of eGFR <50 mL/min/1.73 m
2
was increased by 76.6% (hazard ratio [HR] = 1.767, 95% confidence interval [CI] = 1.406–2.220), 2.24- and 2.87-fold higher for KDPI higher >35%–50% (HR = 2.239, 95% CI = 1.862–2.691), and >51%–85% (HR = 2.871, 95% CI = 2.361–3.491), respectively. Other variables associated with a lower graft function were donor sex (HR male versus female = 0.896, 95% CI = 0.813–0.989) and cold ischemia time (HR for each hour = 1.011, 95% CI = 1.004–1.019). This association was sustained after the Poisson mediation analysis, including delayed graft function, cytomegalovirus, and acute rejection as mediators.
Conclusions.
In this cohort of deceased donor kidney recipients, KDPI, and cold ischemia time were the major independent risk factors associated with lower 1-y kidney function.