Improving Utilization of Hard-to-Place Kidneys

Robert J. Stratta, MD


December 03, 2010

Improving Distribution Efficiency of Hard-to-Place Deceased Donor Kidneys: Predicting Probability of Discard or Delay

Massie AB, Desai NM, Montgomery RA, Singer AL, Segev DL
Am J Transplant. 2010;10:1613-1620

Study Summary

Massie and associates queried the United Network for Organ Sharing database and retrospectively analyzed 39,035 adult kidneys recovered from deceased donors for possible transplantation between 2005 and 2008. A total of 9262 recovered kidneys (24.4%) were ultimately not transplanted (discarded), and an additional 1173 organs (3.0%) had a cold ischemic time in excess of 36 hours prior to transplantation, which for study purposes was defined as a delay in placement. Compared with kidneys transplanted within 36 hours of recovery, kidneys that were discarded or that experienced a delayed placement came from older donors (mean 55 vs 43 years) who were more likely to have a history of smoking (42.2% vs 31.6%), hypertension (59.4% vs. 29.0%), diabetes (20.5% vs 7.0%), death by cerebrovascular accident (65.4% vs 43.7%), hepatitis C (10.4% vs 2.8%), or a biopsy showing > 20% glomerulosclerosis (32.4% vs 6.0%; all P < 0.01 except death from cerebrovascular accident).

A regression model was created that reliably (area under curve 0.83) quantified the probability that a given kidney was either discarded or delayed. Compared with easy-to-place kidneys, hard-to-place kidneys came from donors with a higher median age (60-62 vs 42-44 years, depending on statistical cut-off chosen), with greater chance of being an expanded criteria donor (76%-82% vs 20%-23%), having a history of hypertension (74%-79% vs 31%-32%), or having > 20% glomerulosclerosis on biopsy (53%-70% vs 4%-6%).

Hard-to-place kidneys also had a higher median number of patient-level refusals (62-65 vs 8-9), center-level refusals (7 vs 2-3), and were less commonly used locally (14%-16% vs 68%-72%). Of the hard-to-place kidneys that were used locally, 40%-47% were used by centers that were among the 20 highest users of hard-to-place kidneys in the country, whereas only 7%-12% were used locally at centers that were not among the most aggressive centers in the country.

Finally 67%-78% of hard-to-place kidneys were eventually discarded compared with only 12%-16% of easy-to-place kidneys. On the basis of this analysis, the investigators concluded that their proposed model reliably identifies hard-to-place kidneys that are poorly served by the current allocation system. Consequently, these kidneys should be distributed through an alternate expedited system to direct these kidneys to centers that are most likely to use them for transplantation.


The issue of kidney discard following deceased donor organ recovery is multifactorial and complex. Previous studies have demonstrated tremendous overlap in the factors that characterize kidneys that are actually used for transplant compared with those that are eventually discarded. No single factor can reliably determine the ultimate fate of a recovered kidney. In addition, discard rates vary widely according to donation service area, acceptance rates vary widely according to transplant center, and variation in these rates is greater when attempting to place organs of marginal quality.

Electronic organ offers through Donor Net 2007 were implemented in April 2007, in an attempt to expedite placement of organs. Deceased donor organs are currently offered to one center at a time, according to a model that does not take into account the willingness of a transplant center to accept organs of marginal quality for a particular patient. Moreover, a contacted center may be willing to use a given organ for a patient farther down on their list but unwilling to use the organ for their patients at the top of the waiting list (for example, they might not want to place a kidney from an expanded criteria donor into an obese, sensitized, or younger recipient).

Under the current system, the kidney is offered to other centers before it can be released to the center willing to use the kidney selectively. Unfortunately, an unintended consequence of this system of distribution is a delay in either contacting centers or releasing kidneys to centers that have a high likelihood of actually transplanting them. By wasting time and systematically offering hard-to-place kidneys to centers that are unlikely to accept them, delays in cold ischemia are amplified and the organs become even more marginal (particularly if they are not preserved by machine perfusion), which contributes to the discard rate.

The proposal to "flag" kidneys that are identified as hard-to-place and expedite placement (perhaps regionally) of these kidneys to centers that are most likely to actually transplant them is an attempt to prioritize utility above justice. However, one might contend that justice would ultimately be served if tweaking of the current system results in lowering of the organ discard rate and increases the number of patients actually transplanted. Tailoring organ distribution to organ quality appears to be a reasonable compromise in spite of potential logistical and political hurdles.