Transplanting kidneys without points for HLA-B matching: consequences of the policy change.

In 2003, the US kidney allocation system was changed to eliminate priority for HLA-B similarity. We report outcomes from before and after this change using data from the Scientific Registry of Transplant Recipients (SRTR). Analyses were based on 108 701 solitary deceased donor kidney recipients during the 6 years before and after the policy change. Racial/ethnic distributions of recipients in the two periods were compared (chi-square); graft failures were analyzed using Cox models. In the 6 years before and after the policy change, the overall number of deceased donor transplants rose 23%, with a larger increase for minorities (40%) and a smaller increase for non-Hispanic whites (whites) (8%). The increase in the proportion of transplants for non-whites versus whites was highly significant (p < 0.0001). Two-year graft survival improved for all racial/ethnic groups after implementation of this new policy. Findings confirmed prior SRTR predictions. Following elimination of allocation priority for HLA-B similarity, the deficit in transplantation rates among minorities compared with that for whites was reduced but not eliminated; furthermore, there was no adverse effect on graft survival.

Geographic variation in end-stage renal disease incidence and access to deceased donor kidney transplantation.

The effect of demand for kidney transplantation, measured by end-stage renal disease (ESRD) incidence, on access to transplantation is unknown. Using data from the U.S. Census Bureau, Centers for Medicare & Medicaid Services (CMS) and the Organ Procurement and Transplantation Network/Scientific Registry of Transplant Recipients (OPTN/SRTR) from 2000 to 2008, we performed donation service area (DSA) and patient-level regression analyses to assess the effect of ESRD incidence on access to the kidney waiting list and deceased donor kidney transplantation. In DSAs, ESRD incidence increased with greater density of high ESRD incidence racial groups (African Americans and Native Americans). Wait-list and transplant rates were relatively lower in high ESRD incidence DSAs, but wait-list rates were not drastically affected by ESRD incidence at the patient level. Compared to low ESRD areas, high ESRD areas were associated with lower adjusted transplant rates among all ESRD patients (RR 0.68, 95% CI 0.66-0.70). Patients living in medium and high ESRD areas had lower transplant rates from the waiting list compared to those in low ESRD areas (medium: RR 0.68, 95% CI 0.66-0.69; high: RR 0.63, 95% CI 0.61-0.65). Geographic variation in access to kidney transplant is in part mediated by local ESRD incidence, which has implications for allocation policy development.

Systematic evaluation of pancreas allograft quality, outcomes and geographic variation in utilization.

Pancreas allograft acceptance is markedly more selective than other solid organs. The number of pancreata recovered is insufficient to meet the demand for pancreas transplants (PTx), particularly for patients awaiting simultaneous kidney-pancreas (SPK) transplant. Development of a pancreas donor risk index (PDRI) to identify factors associated with an increased risk of allograft failure in the context of SPK, pancreas after kidney (PAK) or pancreas transplant alone (PTA), and to assess variation in allograft utilization by geography and center volume was undertaken. Retrospective analysis of all PTx performed from 2000 to 2006 (n = 9401) was performed using Cox regression controlling for donor and recipient characteristics. Ten donor variables and one transplant factor (ischemia time) were subsequently combined into the PDRI. Increased PDRI was associated with a significant, graded reduction in 1-year pancreas graft survival. Recipients of PTAs or PAKs whose organs came from donors with an elevated PDRI (1.57-2.11) experienced a lower rate of 1-year graft survival (77%) compared with SPK transplant recipients (88%). Pancreas allograft acceptance varied significantly by region particularly for PAK/PTA transplants (p < 0.0001). This analysis demonstrates the potential value of the PDRI to inform organ acceptance and potentially improve the utilization of higher risk organs in appropriate clinical settings.

Predictability of survival models for waiting list and transplant patients: calculating LYFT.

'Life years from transplant' (LYFT) is the extra years of life that a candidate can expect to achieve with a kidney transplant as compared to never receiving a kidney transplant at all. The LYFT component survival models (patient lifetimes with and without transplant, and graft lifetime) are comparable to or better predictors of long-term survival than are other predictive equations currently in use for organ allocation. Furthermore, these models are progressively more successful at predicting which of two patients will live longer as their medical characteristics (and thus predicted lifetimes) diverge. The C-statistics and the correlations for the three LYFT component equations have been validated using independent, nonoverlapping split-half random samples. Allocation policies based on these survival models could lead to substantial increases in the number of life years gained from the current donor pool.

Calculating life years from transplant (LYFT): methods for kidney and kidney-pancreas candidates.

The Organ Procurement and Transplantation Network (OPTN) Kidney Committee is considering a proposal for a new deceased donor kidney allocation system. Among the components under consideration is a strategy to rank candidates in part by the estimated incremental years of life that are expected to be achieved with a transplant from a specific available deceased donor, computed as the difference in expected median lifespan with that transplant compared with remaining on dialysis. This concept has been termed life years from transplant or LYFT. Median lifespans could be calculated, based on objective medical criteria, for each candidate when a deceased donor kidney becomes available, based on Cox regression models using current candidate and donor medical information. The distribution of the calculated LYFT scores for an average nonexpanded criteria donor kidney is similar across candidate sex, race/ethnicity, insurance status and, with the exception of diabetes, diagnosis. LYFT scores tend to be higher for younger candidates and lower for diabetics receiving a kidney-alone rather than a simultaneous kidney-pancreas transplant. Prioritizing candidates with higher LYFT scores for each available kidney could substantially increase total years of life among both transplant candidates and recipients. LYFT is also a powerful metric for assessing trends in allocation outcomes and for comparing alternative allocation systems.

SRTR program-specific reports on outcomes: a guide for the new reader.

Differences in outcomes indeed exist among transplant programs and organ procurement organizations (OPO). A growing set of tools are available from the Scientific Registry of Transplant Recipients (SRTR) to measure and assess these outcomes in the different phases of the transplant process. These tools are not intended to compare two individual programs, rather to help identify programs whose practices may need further scrutiny, to be either avoided, corrected or emulated. To understand which differences in outcomes might be due to underlying differences in populations served and which might be due to differences in treatment, it is important to compare outcomes to 'risk-adjusted' expected values. Further, it is important to recognize and assess the role that random chance may play in these outcomes by considering the p-value or confidence interval of each estimate. We present the reader with a basic explanation of these tools and their interpretation in the context of reading the SRTR Program-Specific Reports. We describe the intended audience of these reports, including patients, monitoring and process improvement bodies, payers and others such as the media. Use of these statistics in a way that reflects a basic understanding of these concepts and their limitations is beneficial for all audiences.

Developing organ offer and acceptance measures: when 'good' organs are turned down.

Turndowns of offers of deceased donor kidneys for transplantation can contribute to inefficiencies in the organ distribution system and inequality in access to donated organs. Match run data were obtained for 4967 'good' kidneys placed and transplanted in 2005 after fewer than 50 offers. These kidneys were not recovered from donation after cardiac death or expanded criteria donors, or from donors with a history of substance abuse. On average, these good kidneys were not accepted until after seven offers to candidates and after offers to 2.4 programs. Models for the likelihood of acceptance found several donor and candidate characteristics to be significantly related to acceptance rates (p < 0.05). After accounting for these variables, there remained 2- to 3-fold differences among transplant programs in acceptance rates. These models could be used to identify kidney transplant centers with exceptional acceptance practices. Several strategies might be employed to increase acceptance rates for good organs.

Geographic variability in access to primary kidney transplantation in the United States, 1996-2005.

This article focuses on geographic variability in patient access to kidney transplantation in the United States. It examines geographic differences and trends in access rates to kidney transplantation, in the component rates of wait-listing, and of living and deceased donor transplantation. Using data from Centers for Medicare and Medicaid Services and the Organ Procurement and Transplantation Network/Scientific Registry of Transplant Recipients, we studied 700,000+ patients under 75, who began chronic dialysis treatment, received their first living donor kidney transplant, or were placed on the waiting list pre-emptively. Relative rates of wait-listing and transplantation by State were calculated using Cox regression models, adjusted for patient demographics. There were geographic differences in access to the kidney waiting list and to a kidney transplant. Adjusted wait-list rates ranged from 37% lower to 64% higher than the national average. The living donor rate ranged from 57% lower to 166% higher, while the deceased donor transplant rate ranged from 60% lower to 150% higher than the national average. In general, States with higher wait-listing rates tended to have lower transplantation rates and States with lower wait-listing rates had higher transplant rates. Six States demonstrated both high wait-listing and deceased donor transplantation rates while six others, plus D.C. and Puerto Rico, were below the national average for both parameters.

Recovery and utilization of deceased donor kidneys from small pediatric donors.

The optimal use of kidneys from small pediatric deceased donors remains undetermined. Using data from the Scientific Registry of Transplant Recipients, 2886 small (< 21 kg) pediatric donors between 1993 and 2002 were identified. Donor factors predictive of kidney recovery and transplantation (1343 en bloc; 1600 single) were identified by logistic regression. Multivariable Cox regression was used to assess the risk of graft loss. The rate of kidney recovery from small pediatric donors was significantly higher with increasing age, weight and height. The odds of transplant of recovered small donor kidneys were significantly higher with increasing age, weight, height and en bloc recovery (adjusted odds ratio = 65.8 vs. single; p < 0.0001), and significantly lower with increasing creatinine. Compared to en bloc, solitary transplants had a 78% higher risk of graft loss (p < 0.0001). En bloc transplants had a similar graft survival to ideal donors (p = 0.45) while solitary transplants had an increased risk of graft loss (p < 0.0001). En bloc recovery of kidneys from small pediatric donors may result in the highest probability of transplantation. Although limited by the retrospective nature of the study, kidneys transplanted en bloc had a similar graft survival to ideal donors but may not maximize the number of successfully transplanted recipients.

SRTR center-specific reporting tools: Posttransplant outcomes.

Measuring and monitoring performance--be it waiting list and posttransplant outcomes by a transplant center, or organ donation success by an organ procurement organization and its partnering hospitals--is an important component of ensuring good care for people with end-stage organ failure. Many parties have an interest in examining these outcomes, from patients and their families to payers such as insurance companies or the Centers for Medicare and Medicaid Services; from primary caregivers providing patient counseling to government agencies charged with protecting patients. The Scientific Registry of Transplant Recipients produces regular, public reports on the performance of transplant centers and organ procurement organizations. This article explains the statistical tools used to prepare these reports, with a focus on graft survival and patient survival rates of transplant centers--especially the methods used to fairly and usefully compare outcomes of centers that serve different populations. The article concludes with a practical application of these statistics--their use in screening transplant center performance to identify centers that may need remedial action by the OPTN/UNOS Membership and Professional Standards Committee.

Cardiovascular mortality in children and young adults with end-stage kidney disease.

OBJECTIVE: To analyze cardiovascular death in a national end-stage renal disease (ESRD) population. STUDY DESIGN: This retrospective, observational study with data from the US Renal Data Systems analyzed 1380 deaths from 1990 to 1996 among patients who started ESRD therapy as children and died before 30 years of age. RESULTS: Percentage of cardiac deaths (n = 311) varied by age and was higher among black patients (0-4 years, 36%; 5-9 years, 18%; 10-14 years, 35%; 15-19 years, 22%; 20-30 years, 32%) than white patients (18%, 12%, 17%, 14%, and 23%, respectively). Among black patients, cardiac deaths occurred in 11% of transplant recipients, 34% of dialysis patients, and among white patients 9% and 25%, respectively. Black patients were 1.6 times more likely to die of a cardiac death (P <.001) than white patients. Transplant recipients had 78% lower risk of cardiac death than dialysis patients (odds ratio = 0.22; P =.0001). The cardiac death rate among dialysis patients was 21.4 per 1000 patient-years in black patients compared with 20.5 in white patients. Transplantation cardiac death rates were lower in black patients, 2.1 per 1000 patient-years, and 1.3 in white patients. CONCLUSIONS: Cardiovascular death accounts for 23% of pediatric and young adult ESRD deaths. Black patients and dialysis patients are at higher risk of a cardiac death compared with white patients and transplant recipients. Further studies are needed to identify risk factors associated with cardiovascular death in patients with ESRD.

Mortality risk by hemodialyzer reuse practice and dialyzer membrane characteristics: results from the usrds dialysis morbidity and mortality study.

Hemodialyzer reuse is commonly practiced in the United States. Recent studies have raised concerns about the mortality risk associated with certain reuse practices. We evaluated adjusted mortality risk during 1- to 2-year follow-up in a representative sample of 12,791 chronic hemodialysis patients treated in 1,394 dialysis facilities from 1994 through 1995. Medical record abstraction provided data on reuse practice, use of bleach, dialyzer membrane, dialysis dose, and patient characteristics and comorbidity. Mortality risk was analyzed by bootstrapped Cox models by (1) no reuse versus reuse, (2) reuse agent, and (3) dialyzer membrane with and without the use of bleach, while considering dialysis and patient factors. The relative risk (RR) for mortality did not differ for patients in reuse versus no-reuse units (RR = 0.96; 95% confidence interval [CI], 0.86 to 1.08; P > 0.50), and similar results were found with different levels of adjustment and subgroups (RR = 1.01 to 1.05; 95% CI, lower bound > 0.90, upper bound < 1.19 each; each P > 0.40). The RR for peracetic acid mixture versus formalin varied significantly by membrane type and use of bleach during reprocessing, achieving borderline significance for synthetic membranes. Among synthetic membranes, mortality was greater with low-flux than high-flux membranes (RR = 1.24; 95% CI, 1.02 to 1.52; P = 0.04) and without than with bleach during reprocessing (RR = 1.24; 95% CI, 1.01 to 1.48; P = 0.04). Among all membranes, mortality was lowest for patients treated with high-flux synthetic membranes (RR = 0.82; 95% CI, 0.72 to 0.93; P = 0.002). Although mortality was not greater in reuse than no-reuse units overall, differences may exist in mortality risk by reuse agent. Use of high-flux synthetic membrane dialyzers was associated with lower mortality risk, particularly when exposed to bleach. Clearance of larger molecules may have a role.