Allocation of the Highest Quality Kidneys and Transplant Outcomes Under the New Kidney Allocation System.

ABSTRACT

RATIONALE & OBJECTIVE: Increased access to transplantation for highly sensitized candidates following implementation of the kidney allocation system (KAS) has been mostly due to higher use of organs with a lower kidney donor profile index (KDPI; a quality metric for donated kidneys), although changes in allocation of these organs was not intended. It is unclear whether clinical outcomes have changed in association with these changes. We investigated the use of kidneys with low and high KDPI scores over time and whether KDPI score affects patient and graft survival differently across varying levels of allosensitization. STUDY DESIGN: Observational cohort study. SETTING & PARTICIPANTS: Adult (aged ≥18 years) recipients of a deceased donor kidney transplant between October 1, 2009, and September 30, 2017 (Organ Procurement and Transplantation Network/United Network for Organ Sharing database; n = 84,451). PREDICTORS Calculated panel-reactive antibody (cPRA) level (0%, 1%-79%, 80%-89%, 90%-98%, and 99%-100%) and KDPI score (≤20%, 21%-85%, and >85%). OUTCOMES: Death, graft loss. ANALYTICAL APPROACH: Time to event. RESULTS: Allocation of kidneys with KDPI scores ≤ 20% and KDPI scores of 21% to 85% to recipients with cPRA levels ≥ 99% increased 4-fold following implementation of the KAS with little change in allocation of kidneys with KDPI scores > 85%. Patient survival and graft loss were strongly associated with KDPI score, whereas the association with cPRA level was minimal. There was no evidence of a differential effect of KDPI scores across the range of cPRA levels on patient survival (P for interaction=0.06-0.9) or graft loss (P for interaction=0.5-0.9). Patient survival at 5 years among the 5 cPRA groups ranged from 87.2% to 89.8% for recipients of kidneys with KDPI scores ≤ 20% (P=0.2), 82.8% to 85.5% for KDPI scores of 21% to 85% (P=0.04), and 70.2% to 79.2% for KDPI scores > 85% (P=0.2). Cumulative incidence of graft loss by cPRA level ranged from 7.7% to 10.6% for recipients of kidneys with KDPI scores ≤ 20% (P=0.2), 11.8% to 15.0% for KDPI scores of 21% to 85% (P < 0.001), and 19.8% to 29.7% for KDPI scores > 85% (P = 0.4). LIMITATIONS: Lack of data for crossmatches, donor-specific antibodies, and immunomodulation. CONCLUSIONS: Highly sensitized recipients received kidneys with lower KDPI scores following implementation of the KAS, reducing access to these kidneys by less-sensitized candidates. KDPI score has a stronger association with patient survival and graft loss than cPRA level. The association of KDPI score with these outcomes was not modified by the recipient's level of sensitization. The impact of the redistribution of kidneys with low KDPI scores on outcomes among less-sensitized recipients needs further evaluation.