Prospective Validation of Prediction Model for Kidney Discard

Area of Science:

• Nephrology
• Transplant Surgery
• Public Health Policy

Background:

• Thousands of kidneys are discarded annually, representing a significant loss of transplantable organs.
• The Probability of Delay or Discard (PODD) model was developed to identify marginal kidneys at risk of discard or delayed allocation.
• Prospective validation of the PODD model was needed due to policy changes like the Kidney Allocation System (KAS).

Purpose of the Study:

• To prospectively validate the PODD model in predicting kidney discard within the KAS era.
• To assess the accuracy of the PODD model in identifying high-risk kidneys.
• To examine patterns of utilization for high-risk kidneys among transplant centers.

Main Methods:

• Prospective validation of the PODD model using Scientific Registry of Transplant Recipients data (2015-2018).
• Calculation of C-statistic to evaluate predictive accuracy for kidney discard.
• Assessment of kidney utilization clustering using Gini coefficients for high-PODD kidneys.

Main Results:

• The PODD model demonstrated accurate prediction of kidney discard under KAS (C-statistic = 0.87).
• Utilization of high-PODD kidneys was highly clustered among a few centers (Gini = 0.84), with over 60% of centers never transplanting them.
• A significant portion of high-PODD kidney offers (35.0%) were made to centers that never accepted such kidneys.

Conclusions:

• The PODD model is a validated tool for predicting kidney discard in the KAS era.
• Targeted allocation of high-PODD kidneys to accepting centers can potentially minimize discard.
• Optimizing allocation strategies for marginal kidneys is crucial for improving transplant rates.