KDPI score and donor urinary biomarkers: poor predictors of graft outcome

KDPI (kidney donor profile index) is a numerical measure that combines ten variables about a donor, including clinical parameters and demographics, to express the quality of the donor kidneys relative to other donors.  It has been reported by UNOS in attempt to classify the quality spectrum of cadaveric kidneys in order to use it for better matching with recipients’ characteristics (previously reviewed by Andrew here). 

One of the major limitations of the KDPI score is its poor predictive power with a C-statistics of 0.6 (0.5 would be the flip of a coin) (Reese et al. JASN 2015). The score relies heavily on age (diverse renal function within same age group is common), donor terminal creatinine (may reflect acute kidney injury), and it does not take into account HLA matching or other immunological risk factors. Therefore, KDPI score is not a good predictor of graft outcome.

In attempt to improve this prediction, Reese et al. (JASN 2015) conducted a nice prospective study in deceased kidney donors and respective recipients to assess associations between urinary biomarkers (NGAL, KIM1, IL18, L-FABP) in deceased-donor urine with three outcomes: donor AKI, recipient delayed graft function and recipient's graft function at 6 months post-transplant. Although donor urinary injury biomarkers was strongly associated with donor AKI, biomarkers provided limited value in predicting delayed graft function or early allograft function after transplant. The search for better predictive tools continues...

Figure above from Kidney Transplant iBook