Optimizing Graft Survival by Pretreatment of the Donor

• delayed graft function
• transplant outcomes
• Graft Survival
• kidney transplantation
• Tissue Donors

Over many decades, organ transplantation has overcome remarkable barriers to evolve into a highly successful enterprise. The dominant limitation faced today is, without a doubt, the insufficient supply of transplantable organs. When faced with any difficult quandary, the medical community relies on research to provide solutions. Any innovation that has proven efficacy to improve graft function and/or enhance graft survival should be highly valued and widely promoted.

In 2009, Schnuelle et al. (1) reported on a randomized, open label trial conducted at 60 European centers to determine whether donor treatment with dopamine infusion (4 μg/kg per minute) might improve recipient kidney function. The trial enrolled 264 donation after brain death (DBD) donors and studied 487 resultant kidney transplant recipients. Donor exposure to dopamine reduced the proportion of recipients who required one or more dialysis treatments: 56 of 227 (24.7%) versus 92 of 260 (35.4%; P=0.01). Those who required multiple compared with no dialysis treatments experienced substantially higher risk of 3-year graft failure (hazard ratio [HR], 3.61; 95% confidence interval [95% CI], 2.39 to 5.45; P<0.001).

Now, after extended follow-up, Schnuelle et al. (2) report in this issue of the Clinical Journal of the American Society of Nephrology that 5-year graft survival did not differ between treatment and control arms. However, because dopamine infusion times ranged widely from 0 to 32 hours, they explored a dose-response relationship. Regression analyses showed that longer periods of donor dopamine administration reduced dialysis need and graft failure. The maximal increase in efficacy occurred at 7 hours of dopamine administration for both end points: multiple dialysis treatments and 5-year graft failure. Remarkably, at 5 years, recipients of kidneys from donors who received dopamine for >7.1 hours (n=76) compared with all other recipients (260 recipients of kidneys from untreated donors combined with 151 kidneys from donors treated for <7.1 hours) had superior overall (81.5% versus 68.5%; log rank P=0.03) and death-censored (90.3% versus 80.2%; log rank P=0.04) graft survival.

The authors should be congratulated for performing detailed post hoc analyses after extended follow-up. Identifying a dose-response effect for both end points—multiple dialysis treatments and midterm graft survival—imparts additional credence and texture to their original findings. It is reassuring that the threshold effects for both of these end points converge. The consistent downward trend of adjusted HRs for hourly increases in dopamine infusion times with lower graft failure risk also tells a coherent story. Finally, the associations of optimal duration of dopamine exposure with 13% and 10% reduction in all-cause and death-censored graft failure, respectively, are definitely impressive. However, although these are tantalizing results, there are important caveats.

First and foremost, it is natural to wonder whether longer duration of dopamine infusion may simply be a surrogate for donor characteristics that predispose to less dialysis requirement and better graft survival. For example, dopamine duration may reflect stable donor hemodynamic status, because that was a requirement for trial enrollment, drug initiation, and drug continuation. Dopamine infusion was continued into the operating theater until donor crossclamp. Organ placement processes may be more protracted for higher-quality donors, thereby prolonging dopamine exposure. Although the authors attempt to mitigate these concerns, performing analyses of important time intervals (between intensive care unit admission, brain death declaration, treatment initiation, and organ procurement), analyses of subjects treated “per protocol,” and analyses incorporating the kidney donor profile index as an indicator of donor quality, lingering concerns remain (2).

A second question that arises is, in the (near) absence of data regarding the post-transplant course, including immunosuppression and rejection, the quality of renal function during follow-up, and the causes of graft failure, whether it is reasonable to believe that donor dopamine administration might translate into a significant and substantial effect on allograft survival. Can an intervention implemented in deceased donors improve recipient outcomes and improve not only short-term outcomes—delayed graft function (DGF)—but also, midterm outcomes—graft survival? For the transplant community, the desirable answer is a resounding yes. However, this remains unproven.

There is no doubt that DGF in and of itself is clinically undesirable. It is medically and psychologically burdensome, not to mention costly, for all parties involved—patients, physicians, and society. Beyond these concrete negatives, DGF connotes a lower-quality graft. Donor, recipient, and transplant risk factors that predispose to DGF also predispose to inferior graft survival (3). However, the key question for time immemorial is whether DGF is an independent risk factor for inferior graft survival. Both pro (4–6) and con (7–9) arguments for kidneys from DBD donors are well represented in the literature. Although this study does not help to resolve this conundrum, it is notable, because it suggests that an optimal dose of an intervention was associated with a reduction in both DGF and midterm graft failure.

In surveying previous attempts to reduce DGF, the landscape has frankly been bleak, populated largely by trials with negative results. Prospective randomized trials have shown that many agents (including furosemide, pentoxifylline, erythropoietin, IGF1, recombinant P-selectin glycoprotein ligand Ig fusion protein, and antibody against intercellular adhesion molecule 1) typically administered to recipients around the time of allograft reperfusion failed to reduce DGF (10–15). A single randomized, open label, single-center trial reported that thymoglobulin infused before compared with after kidney reperfusion decreased DGF incidence (16). Disappointingly, this finding has not been confirmed, perhaps with a caveat that the timing of thymoglobulin may not have been optimized (17).

Although multiple treatments administered to recipients as delineated above have failed, different organ preservation approaches have affected DGF rates. Hypothermic machine perfusion compared with cold static storage has decreased DGF incidence for both DBD and donation after cardiac death (DCD) kidneys (18). This definitive trial smartly used a paired kidney design, thereby neutralizing the numerous donor factors that affect DGF. Machine perfusion reduced DGF risk (adjusted odds ratio, 0.57; 95% CI, 0.36 to 0.88; P=0.01). Moreover, perfusion was associated with lower risk of 1-year graft failure (adjusted HR, 0.52; 95% CI, 0.29 to 0.93; P=0.03) that persisted through 3 years (adjusted HR, 0.60; 95% CI, 0.37 to 0.97; P=0.04) (19). Although the magnitude of DGF risk reduction was comparable for kidneys from all donor flavors—standard versus expanded criteria and DBD versus DCD—the magnitude of graft survival benefit differed. The greatest gains in graft survival were observed in kidneys from DBD donors and in particular, expanded criteria DBD donors. This uncoupling of DGF and graft survival outcomes has been previously observed: DCD compared with DBD kidneys experience considerably higher DGF rates but enjoy equivalent graft survival rates (20,21). The disconnect emphasizes that interventions that reduce DGF may or may not enhance graft survival and if so, may enhance it only for a subset of kidneys.

Although hypothermic machine perfusion has shown efficacy to mitigate DGF, arguably the most elegant (and certainly the cheapest) intervention to reduce DGF is the induction of mild donor hypothermia after declaration of brain death (22). Compared with normothermia (36.5°C to 37.5°C), mild hypothermia (34°C to 35°C) substantially reduced DGF (odds ratio, 0.62; 95% CI, 0.43 to 0.92; P=0.02), with greater effect on expanded versus standard criteria donor kidneys. Unfortunately, no data regarding graft survival are yet available.

In surveying the spectrum of efforts targeting DGF, there seems to be an effect gradient, with interventions most proximal to organ injury—those delivered to donors—showing the greatest efficacy and interventions most distal to organ injury—those delivered to recipients—showing the least efficacy. Biologically and intuitively, this makes sense: intervening in the deceased donor can mitigate injury, whereas intervening in the transplant recipient can only accelerate recovery from injury. It is striking, however, that many more studies are conducted in recipients than donors. The design, organization, and execution of high-quality, prospective, randomized, multicenter trials in deceased donors have been stymied by a myriad of ethical, regulatory, and logistical challenges (23,24). Donor intervention trials represent a unique research paradigm; the investigational treatment is delivered to donors, but its effect is studied in recipients: the research does not occur in subjects who received the intervention. For both the dopamine and the hypothermia trials, where the intervention was arguably minimal or low risk, the overseeing ethical commission/institutional review board deemed that informed consent from transplant recipients was unnecessary. However, for interventions with higher-risk profiles, waitlisted candidates offered organs from exposed donors will need to provide consent, indicating willingness to accept these exposed organs for transplantation. Actual recipients will also need to provide consent, indicating willingness to participate in research. The broad geographic reach of contemporary organ allocation and distribution policies renders it unlikely, if not impossible, to identify potential recipients in advance of administering the intervention to deceased donors. The time constraints imposed by donation and transplantation challenge basic tenets of informed consent. These along with other challenging issues that obstruct research in deceased donors have been acknowledged and embraced by the Institute of Medicine as worthy of study (http://www.nationalacademies.org/hmd/Activities/Research/IssuesinOrganDonorInterventionResearch.aspx). The transplant community should anticipate with bated breath emerging guidance that, ideally, will not only enable but also, facilitate the ethical and safe conduct of trials to test innovative therapies in deceased donors. The impressive results of the donor dopamine and hypothermia trials emphatically show that interventions in deceased donors hold promise to improve the quality and increase the quantity of organs for transplantation.

• Copyright © 2017 by the American Society of Nephrology