Limitation of Terminal Serum Creatinine as a Kidney Donor Profile Index Variable in Predicting Long-Term Kidney Transplant Outcomes

Background: Donor final serum creatinine (SCr) is a dynamic variable and is 1 of 10 factors used in calculating kidney donor profile index (KDPI). We hypothesize that deceased-donor kidneys (DDKs) with higher SCr were likely accepted for transplantation if procurement biopsy findings were favorable and with long-term outcomes no worse than kidneys with lower final SCr within a KDPI group.

Methods: Using the Organ Procurement and Transplant Network/United Network for Organ Sharing database, we identified DDK transplant recipients from 2000 to 2015 who received induction and calcineurin inhibitor/mycophenolate mofetil maintenance. Patients were divided into 4 KDPI groups: 0-20%, 21%-50%, 51%-85%, and >85%. In each KDPI category, long-term outcomes were compared, with the use of Cox models, between patients who received kidneys with final SCr >2 versus ≤2 mg/dL.

Results: A total of 59,644 patients were divided into KDPI groups 0-20% (SCr >2 mg/dL, n = 478; SCr ≤2 mg/dL, n = 14,769), 21%-50% (SCr >2 mg/dL, n = 1,592; SCr ≤2 mg/dL, n = 17,762), 51%-85% (SCr >2 mg/dL; n = 1,388, SCr ≤2 mg/dL, n = 18,024), and >85% (SCr >2 mg/dL, n = 349; SCr ≤2 mg/dL, n = 5,282). Adjusted overall graft failure risks (hazard ratio [HR] 0.88, P = .04; HR, 0.86, P = .007) and patient death risks (HR, 0.86, P = .04; HR, 0.84, P = .01) for final SCr >2 versus ≤2 mg/dL groups were lower in KDPI categories 21%-50% and 51%-85%, respectively, with similar death-censored graft failure risks.

Discussion: Outcomes of transplanting DDKs with elevated final SCr are no worse than transplanting kidneys with lower final SCr, highlighting the limitation of the single value of final SCr as a variable for calculating KDPI.