The Kidney Donor Profile Index (KDPI) of Marginal Donors Allocated by Standardiz

[Clark]

http://onlinelibrary.wiley.com/doi/10.1111/ajt.12928/abstract

The Kidney Donor Profile Index (KDPI) of Marginal Donors Allocated by Standardized Pretransplant Donor Biopsy Assessment: Distribution and Association With Graft Outcomes
I. Gandolfini1, C. Buzio1, P. Zanelli2, A. Palmisano1, E. Cremaschi1, A. Vaglio1, G. Piotti1, L. Melfa1, G. La Manna3, G. Feliciangeli3, M. Cappuccilli3, M. P. Scolari3, I. Capelli3, L. Panicali3, O. Baraldi3, S. Stefoni3, A. Buscaroli4, L. Ridolfi5, A. D'Errico6, G. Cappelli7, D. Bonucchi7, E. Rubbiani7, A. Albertazzi7, A. Mehrotra8, P. Cravedi8,† andU. Maggiore1,†,*
DOI: 10.1111/ajt.12928
American Journal of Transplantation
Early View (Online Version of Record published before inclusion in an issue)

Abstract

Pretransplant donor biopsy (PTDB)-based marginal donor allocation systems to single or dual renal transplantation could increase the use of organs with Kidney Donor Profile Index (KDPI) in the highest range (e.g. >80 or >90), whose discard rate approximates 50% in the United States. To test this hypothesis, we retrospectively calculated the KDPI and analyzed the outcomes of 442 marginal kidney transplants (340 single transplants: 278 with a PTDB Remuzzi score <4 [median KDPI: 87; interquartile range (IQR): 78–94] and 62 with a score = 4 [median KDPI: 87; IQR: 76–93]; 102 dual transplants [median KDPI: 93; IQR: 86–96]) and 248 single standard transplant controls (median KDPI: 36; IQR: 18–51). PTDB-based allocation of marginal grafts led to a limited discard rate of 15% for kidneys with KDPI of 80–90 and of 37% for kidneys with a KDPI of 91–100. Although 1-year estimated GFRs were significantly lower in recipients of marginal kidneys (−9.3, −17.9 and −18.8 mL/min, for dual transplants, single kidneys with PTDB score <4 and =4, respectively; p < 0.001), graft survival (median follow-up 3.3 years) was similar between marginal and standard kidney transplants (hazard ratio: 1.20 [95% confidence interval: 0.80–1.79; p = 0.38]). In conclusion, PTDB-based allocation allows the safe transplantation of kidneys with KDPI in the highest range that may otherwise be discarded.

[Clark]

http://onlinelibrary.wiley.com/doi/10.1111/ajt.12930/full

Editorial
KDPI and Donor Selection
A. Gupta1,*, G. Chen2 andB. Kaplan1
DOI: 10.1111/ajt.12930
American Journal of Transplantation
Early View (Online Version of Record published before inclusion in an issue)

Due to shortage of organs, transplant physicians are often confronted with the dilemma of accepting versus rejecting a less than optimal kidney offer. To maximize organ utilization, it has become important to develop tools for identification of kidneys that can be accepted without significantly compromising the allograft or recipient outcomes.

With the above goals in mind, Gandolfini et al [1] analyzed the contribution of donor biopsies in the acceptance or rejection of the so-called marginal kidneys. The majority of the kidneys in this study qualified to be extended criteria donor (ECD) kidneys, in this sense it may be viewed as confirmatory study of a previous study by Remuzzi et al [2]. Similar to the previous study by Remuzzi et al, the use of pretransplant biopsies increased utilization of marginal kidneys in this cohort of patients. The authors further characterized these kidneys with the Kidney Donor Profile Index (KDPI) and used this for further analysis. They found that kidneys with the highest KDPI might have superior outcomes with a lower Remuzzi score than those with a higher score. The results of the current study are far from definitive in the controversy over the utility of pretransplant biopsies and recipient outcome, but offer added evidence that perhaps in marginal/ECD/high KDPI kidneys biopsy data may have some incremental utility. The issue of pretransplant biopsies is far too complex and controversial to be discussed within the limitations of this editorial. Therefore we would like to take this opportunity to concentrate and give thought to the appropriate use of KDPI in kidney transplantation.

KDPI or Kidney Donor Risk Index (KDRI) from which KDPI is derived compiles donor factors that are independently associated with all-cause allograft survival associated with the use of that particular organ [3]. Lower KDPI is associated with longer predicted survival while higher KDPI is associated with a shorter predicted survival for the aggregate population. KDPI was meant to rank order the quality of kidneys as defined by an aggregate population relative risk. It was never meant to be utilized as a discriminatory tool to determine acceptance/rejection of a particular kidney offer. As can be seen by the Organ Procurement and Transplantation Network (OPTN) (Figure 1), the curve is essentially flat between the values of 20 and 80. Hence a difference in KDPI in that range will not offer clinically relevant differences in relative risk of kidney survival and certainly would not be expected to accurately discriminate the outcome of a single kidney. Though the curve steepens (in respect to relative risk) beyond a KDPI of 80, this is still not the same as an individual predictive metric nor should it be utilized as a sole basis to accept or not accept an organ. The average c-statistic for the overall cohort was 0.62 and the R square of the KDPI model is relatively low. In real life scenarios, there are more confounders than the ones accounted for in the calculation of KDPI that may further decrease the discriminatory power. Clinical applicability of a continuous variable has limitations in a binary decision-making process like acceptance or rejection of a kidney (unless there is a cutoff). Thus one must use great caution in utilizing KDPI to predict a specific patient's suitability for that organ or for discriminating kidneys that are suitable for transplantation.

image
Figure 1. Estimated graft survival rates by KDPI. Includes data on primary, solitary, adult kidney transplants from 2003 to 2010 from the Organ Procurement and Transplantation Network (OPTN) data as of March 8, 2013. Donor reference population: all kidney donors recovered in 2012. Source: OPTN.

The United Network for Organ Sharing (UNOS)/OPTN has approved a new allocation policy based on KDPI that will take effect by the end of 2014. The transition in the allocation system from the ECD criteria to KDPI was proposed also to increase utilization of marginal kidneys and decrease the discard rates. As pointed out by the authors, the use of KDPI in the allocation policy has the potential to actually increase the discard rate. Larger KDPI values may make centers discard kidneys that may have been accepted in the absence of a KDPI ranking. This must be avoided at all costs, as the intent of the scoring system was not as a decision tool, but only to better characterize potential donor organs.

Finally, while pretransplant biopsies may offer added information about donor kidneys, they should not be construed the same as the deciding criteria for acceptance of organs. Although the issue of pretransplant biopsies remains controversial, the study by Gandolfini et al [1] offers an opportunity to reflect on the KDPI scoring system and the need to not over interpret what KDPI can offer as a decision tool as opposed to its original purpose as rank order metric for an aggregate population risk.