A Comprehensive Risk Quantification Score for Deceased Donor Kidneys: The Kidney Donor Risk Index
Rao PS, Schaubel DE, Guidinger MK, et al
Transplantation. 2009;88:231-236
Summary
The authors queried the Organ Procurement and Transplantation Network, Scientific Registry of Transplant Recipients (OPTN/SRTR) database and analyzed retrospectively 92,102 deceased donor kidney transplants performed between January 1 and December 31, 2005. Pediatric recipients, retransplants, multi-organ transplants, ABO-incompatible transplants, and donors with missing data were excluded, leaving 69,440 primary, adult, kidney alone transplants available for analysis.
A Cox multivariable regression model was used to develop a continuous graft failure (death or graft loss) risk score that would capture both donor and transplant characteristics while adjusting for recipient variables. The Cox model was stratified by recipient transplant center, recipient age, and diabetes status. Factors with nonsignificant effects were deleted from the model in sequence.
The proposed Kidney Donor Risk Index (KDRI) includes the following 14 donor and transplant factors, each found to be independently associated with graft failure:
Donor age;
Race;
History of hypertension;
History of diabetes;
Serum creatinine level;
Cerebrovascular cause of death;
Height;
Weight;
Donation after cardiac death;
Hepatitis C virus (HCV) status;
Human leukocyte antigen B and DR mismatch;
Cold ischemia time; and
Double or en bloc transplant.
The reference donor (KDRI = 1.00) had the following characteristics: 40 years of age, non-African American race, serum creatinine 1.0 mg/dL, nondiabetic, nonhypertensive, noncerebrovascular cause of death, height 170 cm, weight ≥ 80 kg, brain-dead donor, HCV negative, 2 HLA-B mismatch, 1 HLA-DR mismatch, and a cold ischemia time of 20 hours. The distribution of calculated KDRI values ranged from a hazard ratio (HR) of 0.5 to 4.2 with a median of 1.05. Covariate-adjusted graft survival rates were calculated by KDRI quintiles and adjusted to a reference 50-year-old nondiabetic recipient.
Transplants of kidneys in the highest KDRI quintile (HR > 1.45) had an adjusted 5-year graft survival of 63%, compared with 82% and 79% in the 2 lowest KDRI quintiles (< 0.79 and ≥ 0.79- 0.96, respectively). Median graft half-life ranged from 7.5 years in the highest quintile to 13.6 years in the lowest quintile. Median graft half-life for the middle quintile (0.96-1.15) was 10.8 years. Considerable overlap existed in the KDRI for the dichotomous classification of expanded criteria donor (ECD) vs non-ECD kidneys.
Through cross-validation, investigators evaluated the discriminatory power of the Cox model and determined that the KDRI is more useful for distinguishing more extreme categories of graft failure risk and of less utility for distinguishing donors in the middle range.
Viewpoint
Due to changing donor demographics, excessive waiting times, and the increasing disparity between organ supply and demand, the use of kidneys from ECDs has become generally accepted and increasingly common. The burgeoning crisis in organ supply challenges the transplant community to reassess thresholds for acceptable risk and to maximize and optimize the use of organs from all consenting donors. To promote the use of ECD kidneys, a previous analysis was performed of "expanded" donors using the SRTR database.
On the basis of this study, a consensus definition of an ECD kidney was developed according to basic donor characteristics that were associated with a relative risk of graft loss > 1.7 (70% greater likelihood of graft loss) when compared with kidneys transplanted from "ideal" donors. ECD was defined as all donors older than 60 years and donors older than 50 years with any 2 of the following criteria: (a) hypertension; (b) cerebrovascular cause of brain death; or (c) pre-retrieval serum creatinine level > 1.5 mg/dL. On the basis of this analysis and definition, the United Network for Organ Sharing (UNOS) introduced a new policy that went into effect on October 31, 2002 that addressed special allocation issues pertaining to ECDs. Not surprisingly, not all ECD kidneys are created equal and a continuum of donor quality exists within the ECD category.
The above study by Rao and colleagues is the most comprehensive attempt to date to characterize the spectrum of deceased donor kidney quality using standard variables to more accurately predict outcomes compared to the ECD vs non-ECD classification. Such data can be used to enhance allocation efficiency and increase sophistication among providers as well as candidates to estimate as well as predetermine the maximum degree of risk that they are willing to consider for a given kidney offer. In addition, it has been proposed that some type of continuous KDRI will replace the current ECD/non-ECD classification in the next generation of the kidney allocation system whenever it is finalized and implemented.
Abstract