The ‘Two, One, Zero’ Decision: What to Do with Suboptimal Deceased Donor Kidneys

Every day around the world clinicians face tough decisions: what should they do when a pair of suboptimal deceased donor kidneys are available for waiting recipients, many of whom have faced long wait times? Should they transplant two into one recipient, one each into two recipients, or discard both: the two, one, zero countdown? Often they discard them: in 2009 in the USA, 2762 (19%) of recovered deceased donor kidneys were discarded (Table 1) (http://www.ustransplant.org/annual_reports/current/302_ord.htm), and this number rises every year. Surprisingly, nearly 987 of the discarded kidneys were from standard criteria donors.

Too often this decision follows a biopsy, even though the biopsy findings may have poor predictive value beyond what information is already known. Comparison of the United States to European practices suggests that the biopsy actually causes discards (1), and the success of the European old-for-old program shows what can be achieved without relying on biopsies (2). Without scientific data about biopsy features, the biopsy can be a dangerous step toward an idiosyncratic decision.

If the clinicians choose to transplant the kidneys, should they transplant two kidneys into one recipient, or one kidney each into two recipients? In the current issue of the AJT, Ekser and colleagues from the University of Padua describe the technical aspects and outcomes of unilateral transplantation of two kidneys from ECD's into one recipient, and compare the outcomes with transplantation of a single ECD kidney (3). They focus on the advantages of unilateral placement as sparing the contralateral side in case of the need for retransplantation, and describe nicely the technical nuances of transplanting both kidneys on one side. The limitations include the fact that it is a single center, nonrandomized experience. In addition, a detailed analysis of the pretransplantation pathology was not performed, and these kidneys were not pumped, so that the pulsatile preservation characteristics could not be described. They report excellent outcomes, with 3-year patient and graft survival rates of 95% and 90%, respectively, similar to that seen in the single kidney patients, and comparable rates of complications. This report, like essentially all of the reports in the two-into-one literature to date, demonstrates that this approach is effective, that is it can be done. Thus, it should be done, compared to discarding both kidneys.

However, it is not clear that it should be done compared to the one-into-one approach, and if so when it should be done. From the first reports, no description has even been given of the detailed histology of the kidneys utilized as two-into-one compared with kidneys transplanted one-into-one for two recipients. Thus, we lack precise criteria to guide clinicians. It is not enough that two-into-one gives recipient results similar to one-into one for two recipients: one patient remains on dialysis. Thus, two-into-one must be much better than one-into-one for two recipients.

Many believe that a rational basis for making the two-one-zero decision can be developed by systematic analysis of the renal function of the donor (creatinine clearance <60 cc), histopathology, demographics and perfusion characteristics. Histopathologic lesions of interest include the percentage of glomerulosclerosis, the degree of interstitial and fibrosis, fibrous intimal thickening of small arteries, as well as flow and resistance in pulsatile preservation, donor serum creatinine, cold ischemia time, or some combination of factors. None of the published papers, including this one, give us sufficient information to make the decision with confidence. Histology may have poor predictive value and is not easy to assess optimally at 3AM, when many of these decisions must be made. The published studies of the histopathology or demographic and perfusion characteristics indicate poor predictive value, and most lack appropriate validation reports. We are thus left with essentially ad hoc decisions by individual surgeons and programs, with little uniformity of approach. The literature to date, including the very nice study published in this issue, continues to raise more questions than it answers.

Equally important to defining donor variables are those variables in the recipient which may influence outcomes for these organs. Thus, recipient age, BMI, primary diagnosis for renal disease, cardiovascular comorbidities, time on dialysis, immunosuppressive regimen, and immunologic risk as defined by the presence and titer of DSAs are all elements that could influence patient and graft survival for suboptimal kidneys.

Oneway forward would be a large-scale multicenter prospective study sufficiently powered to account for all these critical variables: a data-driven prospective research project including histology, pump parameters, clinical parameters and potentially molecular parameters in sufficient detail to allow rational decisions to be made in the future. Indeed, it is time for a consensus conference to develop such a data-driven approach. Quantitative risk scores based on objective donor and recipient variables—and if validated, biopsy variables—could be derived from such a study and could truly yield important and interpretable results that will move the entire field forward. Once risks are defined on a continuous scale, future studies could determine whether transplants with defined risks should be managed differently to improve postoperative physiologic and immunologic management. Providing more support for the front-line clinician who must make the two-one-zero call could be an effective investment for improving the numbers and effectiveness of deceased donor kidney transplants, and would help to create an environment of accountability and rational choices. The result could be hundreds or thousands more of the 2762 discarded kidneys being transplanted, and a scientific basis for the ‘two, one, zero’ decision.