Predicting outcomes for ET kidneys from brain-dead donors using optional nephropathology

Below you will find a summary of the “2-stage scores with optional nephropathology to predict adverse events in brain-dead donor kidneys in Eurotransplant”, published in the April 2024 issue of Nephrology by Ernst et al.

Solid evidence is often lacking when deciding whether to use or discard rare and infrequent brain-dead donor kidneys in Eurotransplant (ET) countries.

Researchers conducted a prospective study to develop and validate two-stage scores reflecting current practice for predicting delayed graft function (DGF) and graft loss at one year (1y-tl) in six Eurotransplant (ET) countries.

They used a training set of 620 cases for DGF and 711 for 1y-tl, with validation sets of n=158 and 162. In step 1, logistic regression models including only clinical predictors were used, and in step 2, risk estimates were updated with nephropathology for relevant risk percentiles.

The results showed that in step 1, a higher risk of DGF was found with longer cold ischemia time, donor and recipient BMI, dialysis age, HLA-DR mismatches, or recipient CMV IgG positivity. The C-statistics were 0.672 and 0.704 for the training and validation set, respectively. The accuracy of DGF prediction improved by 18% and 36% with nephropathology including multiple glomeruli and Banff CV (C-statistics: 0.696 and 0.701). For 1-year TL, the risk increased with cold ischemia time, HLA mismatches, and donor age. The C-statistics were 0.700 and 0.769. Banff-CT improved 1-year TL prediction (C-statistics: 0.706 and 0.765), with appropriate calibration, during training but modestly during validation. Discrimination was comparable to established values ​​during validation.

The researchers concluded that the 2-stage scores can be adapted to nephropathology and promise robust outcomes in ET practice. They potentially outperform existing scores through their versatility in donation after cardiac death and pump use, and offer practical advantages to clinicians.

Source: academic.oup.com/ndt/advance-article-abstract/doi/10.1093/ndt/gfae093/7649367