Faculty Bibliography

U.S. organ allocation policy sequesters livers from deceased donors within arbitrary geographic boundaries, frustrating the intent of those who wish to offer the livers to transplant candidates based on medical urgency. We used a zero-one integer program to partition 58 donor service areas into between four and eight sharing districts that minimize the disparity in liver availability among districts. Because the integer program necessarily suppressed clinically significant differences among patients and organs, we tested the optimized district maps with a discrete-event simulation tool that represents liver allocation at a per-person, per-organ level of detail. In April 2014, the liver committee of the Organ Procurement and Transplantation Network (OPTN) decided in a unanimous vote of 22-0-0 to write a policy proposal based on our eight-district and four-district maps. The OPTN board of directors could implement the policy after the proposal and public-comment period.Redistricting liver allocation would save hundreds of lives over the next five years and would attenuate the serious geographic inequity in liver transplant offers.

Liver allocation is based on current Model for End-Stage Liver Disease (MELD) scores, with priority in the case of a tie being given to those waiting the longest with a given MELD score. We hypothesized that this priority might not reflect risk: registrants whose MELD score has recently increased receive lower priority but might have higher wait-list mortality. We studied wait-list and posttransplant mortality in 69,643 adult registrants from 2002 to 2013. By likelihood maximization, we empirically defined a MELD spike as a MELD increase ≥ 30% over the previous 7 days. At any given time, only 0.6% of wait-list patients experienced a spike; however, these patients accounted for 25% of all wait-list deaths. Registrants who reached a given MELD score after a spike had higher wait-list mortality in the ensuing 7 days than those with the same resulting MELD score who did not spike, but they had no difference in posttransplant mortality. The spike-associated wait-list mortality increase was highest for registrants with medium MELD scores: specifically, 2.3-fold higher (spike versus no spike) for a MELD score of 10, 4.0-fold higher for a MELD score of 20, and 2.5-fold higher for a MELD score of 30. A model incorporating the MELD score and spikes predicted wait-list mortality risk much better than a model incorporating only the MELD score. Registrants with a sudden MELD increase have a higher risk of short-term wait-list mortality than is indicated by their current MELD score but have no increased risk of posttransplant mortality; allocation policy should be adjusted accordingly.

BACKGROUND:Frailty, a validated measure of physiologic reserve, predicts adverse health outcomes among adults with end-stage renal disease. Frailty typically is not measured clinically; instead, a surrogate-perceived frailty-is used to inform clinical decision-making. Because correlations between perceived and measured frailty remain unknown, the aim of this study was to assess their relationship. METHODS:146 adults undergoing hemodialysis were recruited from a single dialysis center in Baltimore, Maryland. Patient characteristics associated with perceived (reported by nephrologists, nurse practitioners (NPs), or patients) or measured frailty (using the Fried criteria) were identified using ordered logistic regression. The relationship between perceived and measured frailty was assessed using percent agreement, kappa statistic, Pearson's correlation coefficient, and prevalence of misclassification of frailty. Patient characteristics associated with misclassification were determined using Fisher's exact tests, t-tests, or median tests. RESULTS:Older age (adjusted OR [aOR] = 1.36, 95%CI:1.11-1.68, P = 0.003 per 5-years older) and comorbidity (aOR = 1.49, 95%CI:1.27-1.75, P < 0.001 per additional comorbidity) were associated with greater likelihood of nephrologist-perceived frailty. Being non-African American was associated with greater likelihood of NP- (aOR = 5.51, 95%CI:3.21-9.48, P = 0.003) and patient- (aOR = 4.20, 95%CI:1.61-10.9, P = 0.003) perceived frailty. Percent agreement between perceived and measured frailty was poor (nephrologist, NP, and patient: 64.1%, 67.0%, and 55.5%). Among non-frail participants, 34.4%, 30.0%, and 31.6% were perceived as frail by a nephrologist, NP, or themselves. Older adults (P < 0.001) were more likely to be misclassified as frail by a nephrologist; women (P = 0.04) and non-African Americans (P = 0.02) were more likely to be misclassified by an NP. Neither age, sex, nor race was associated with patient misclassification. CONCLUSIONS:Perceived frailty is an inadequate proxy for measured frailty among patients undergoing hemodialysis.

In June 2013, a change to the liver waitlist priority algorithm was implemented. Under Share 35, regional candidates with MELD ≥ 35 receive higher priority than local candidates with MELD < 35. We compared liver distribution and mortality in the first 12 months of Share 35 to an equivalent time period before. Under Share 35, new listings with MELD ≥ 35 increased slightly from 752 (9.2% of listings) to 820 (9.7%, p = 0.3), but the proportion of deceased-donor liver transplants (DDLTs) allocated to recipients with MELD ≥ 35 increased from 23.1% to 30.1% (p < 0.001). The proportion of regional shares increased from 18.9% to 30.4% (p < 0.001). Sharing of exports was less clustered among a handful of centers (Gini coefficient decreased from 0.49 to 0.34), but there was no evidence of change in CIT (p = 0.8). Total adult DDLT volume increased from 4133 to 4369, and adjusted odds of discard decreased by 14% (p = 0.03). Waitlist mortality decreased by 30% among patients with baseline MELD > 30 (SHR = 0.70, p < 0.001) with no change for patients with lower baseline MELD (p = 0.9). Posttransplant length-of-stay (p = 0.2) and posttransplant mortality (p = 0.9) remained unchanged. In the first 12 months, Share 35 was associated with more transplants, fewer discards, and lower waitlist mortality, but not at the expense of CIT or early posttransplant outcomes.

Whether the liver allocation system shifts organs from better performing organ procurement organizations (OPOs) to poorer performing OPOs has been debated for many years. Models of OPO performance from the Scientific Registry of Transplant Recipients make it possible to study this question in a data-driven manner. We investigated whether each OPO's net liver import was correlated with 2 performance metrics [observed to expected (O:E) liver yield and liver donor conversion ratio] as well as 2 alternative explanations [eligible deaths and incident listings above a Model for End-Stage Liver Disease (MELD) score of 15]. We found no evidence to support the hypothesis that the allocation system transfers livers from better performing OPOs to centers with poorer performing OPOs. Also, having fewer eligible deaths was not associated with a net import. However, having more incident listings was strongly correlated with the net import, both before and after Share 35. Most importantly, the magnitude of the variation in OPO performance was much lower than the variation in demand: although the poorest performing OPOs differed from the best ones by less than 2-fold in the O:E liver yield, incident listings above a MELD score of 15 varied nearly 14-fold. Although it is imperative that all OPOs achieve the best possible results, the flow of livers is not explained by OPO performance metrics, and instead, it appears to be strongly related to differences in demand.