Normothermic regional perfusion (NRP) has revolutionized organ utilization and graft outcomes following donation after circulatory determination of death (DCDD). NRP involves the in situ regional perfusion of the abdomen or chest and abdomen with oxygenated blood at normothermic temperature range (35 °C–37 °C), initiated after the formal declaration of death and prior to organ recovery [1].

The growing adoption of NRP in DCDD is primarily due to significant improvements in post-transplant outcomes and organ utilization compared to the conventional super rapid recovery (SRR) technique [2–8]. Unlike organ-specific ex situ perfusion technologies, NRP allows for the simultaneous perfusion of multiple organs, thereby potentially reducing costs, complexity and improving patient access to high-quality DCDD organs [5, 9, 10].

Broader implementation of NRP remains limited by organizational, legal, and ethical challenges. Most notably, the absence of standardized criteria and definitions for many technical and clinical aspects are key barriers to its global expansion. This heterogeneity hinders the comparison of outcomes and the harmonization of protocols across centers and countries.

A panel of international experts with extensive experience in the field of organ transplantation and particularly in the use of NRP participated in this Delphi study (cfr. Table 1 and Martin et al.; [11]) specifically focused upon controlled DCDD.

The objective was to identify and analyze areas of ongoing controversy regarding the application of NRP where there was a lack of standardized consensus across centers and countries (see Results for participant details). Therefore, the aim of this consensus was to address this lack of uniformity by proposing consensus-based recommendations to facilitate the safe and effective global implementation of NRP in controlled DCDD.

The Delphi process applied in the overarching DCDD Consensus project has been outlined in detail [11]. In essence, a steering committee of 10 members with expertise in NRP led by two coordinators, developed a questionnaire addressing a range of topics identified as priorities through literature review and group discussions. The committee also identified expert panelists according to the previously described criteria [11], who were invited to participate in the Delphi process. The two coordinators did not participate in the expert panel.

Two survey waves were conducted using an online questionnaire administered by the independent company Adelphi Targis. All panelists who completed the first questionnaire were invited to complete the second wave survey. The questionnaire was refined for the second wave as described [11].

In each survey round panelists indicated their level of agreement with a series of statements using a Likert scale (1–9, strongly disagree–strongly agree); responses were analysed using descriptive statistics with ‘disagreement’ assigned to ratings 1–3, “neither agree nor disagree” 4-6, and “agreement” to ratings 7–9. Participants could indicate if a question was not relevant to their expertise and their responses were removed from the denominator when evaluating consensus on specific questions. Statements that reached 75% agreement were deemed to achieve consensus.

Consistent with the design of the Bucharest consensus, paediatric DCDD was addressed separately by a dedicated panel.

Thirty-one experts from nine countries completed the first round of the Delphi process, all of whom completed the second round. Panel demographics are shown in Table 1.

Consensus was achieved on 105 statements after the first round, and on an additional 25 statements after the second round. Following the first and the second round, 52 and 74 statements respectively were eliminated due to lack of consensus. In the second round 48 statements of the first round were reproposed or remodulated whereas 25 new statements were added. The recommendations summarised in Tables 2–14 below reflect the statements for which consensus was achieved, noting some were combined for efficiency.

There is broad expert agreement that perfusion techniques—either in situ (NRP) or ex situ (NMP or HMP)—should be routinely implemented in all cDCDD donors to optimize post-transplant graft function if they are available. In countries unable to incorporate these techniques, SRR and static cold storage can be adopted provided a strict control for other factors that affect post-transplant outcomes.

Recent studies have consistently demonstrated the positive impact of these perfusion strategies on recipient outcomes when compared with the standard rapid recovery technique [3–5, 7, 8, 25]. This Consensus did not aim at any point to assess the combined use of NRP with ex situ preservation techniques such as NMP or HMP. This topic has been recently described in the European Liver and Intestine Transplant Association (ELITA) consensus [17].

NRP is strongly recommended in many different situations (Table 3). However, no consensus was reached regarding scenarios of deceased donation involving conscious patients who can make their own decisions about end-of-life care (e.g., persons with end-stage respiratory failure or neurodegenerative diseases, or those in whom medical assistance in dying has been authorized). In countries where DCDD is possible under these circumstances, patients may require intensive care to facilitate organ donation (ICOD), including elective ventilation [40–42]. The limited clinical experience with these specific subgroups of donors and ICOD is likely the main reason for the lack of agreement.

These discrepancies regarding the patient groups in which NRP should be applied highlight the heterogeneity of current practices and underscore the urgent need for consensus statements and evidence-based clinical guidelines. This Delphi Consensus highlights the substantial differences observed in the implementation of NRP worldwide, all the more striking given that the participating experts have extensive experience in this field.

This study demonstrates a high degree of uniformity in the technical and human resource requirements to implement NRP in any center. It also shows broad consensus regarding the training needs of newly established NRP teams.

We believe there is an urgent need to agree on certain terms, such as functional warm ischemic time (FWIT), in order to promote uniformity across studies. Unfortunately, no consensus was achieved in defining the maximum acceptable limit of warm ischemic time (WIT) among experts.

One of the most important aspects of this Consensus is the agreement that the use of certain premortem measures, such as the administration of heparin or the placement of guidewires or cannulas before WLSM, improve post-transplant outcomes. This consensus may help to change the regulatory framework in many countries where these practices are still not permitted [15]. One of the most surprising findings in this area was that peripheral cannulation before WLSM was generally considered appropriate in the setting of TA-NRP, but not in A-NRP (Table 6).

Undoubtedly, the area that generates the greatest disagreement among experts is the use of TA-NRP, where there is considerable divergence regarding how to ensure the absence of cerebral perfusion during the procedure.

Due to the uncertainty that complete cessation of blood flow to the brain results by only clamping the arch vessels, current protocols in the United Kingdom and Spain have added the step of severing the arch vessels distal to the clamps and draining blood from the cephalad ends to atmospheric pressure [39, 43, 44]. However, no consensus was achieved regarding whether the vessels should be transected and left to drain to atmospheric pressure. Some authors, particularly in Europe, argue that clamping alone may be insufficient to ensure the absence of postmortem cerebral perfusion [43–46], whereas U.S. guidelines consider both clamping alone and clamping with transection as acceptable options [45, 47]. The American Society of Transplant Surgeons produced an updated statement supporting clamping and venting the supra-aortic trunks when using TA-NRP [26].

The differences observed between authors, and particularly between European and American guidelines regarding clamping alone or clamping and sectioning the supra-aortic trunks when using TA-NRP represent a point of controversy that will need to be solved [43–47]. A Spanish study evaluating the intracerebral blood pressure at the Circle of Willis and the basilar artery—which encompasses the anterior, middle and posterior cerebral circulations—provides strong evidence to alleviate concerns regarding reperfusion of the brain during the TA-NRP procedure, as long as the arch vessels are clamped and the cephalad ends are vented to the atmosphere [48].

We believe that one reason for the significant divergences regarding TA-NRP is the limited global experience with this technique. In fact, in some of the countries represented by the experts in this consensus, TA-NRP has not yet been implemented. Furthermore, there is significant inconsistency in the recommendation to clamp or to ligate the aortic arch vessels (see Table 8, point 2) in order to improve the technical success of the TA-NRP, with no agreement on whether they should be also sectioned. On the contrary, when the focus is on prevention of brain perfusion during TA-NRP (Table 13), there was consensus that it is precluded by clamping or ligation of the aortic arch vessels and sectioning of the cephalic ends of the vessels, plus drainage and aspiration of blood from their caudal ends to achieve atmospheric pressure in the arteries is sufficient to maintain cessation of cerebral perfusion.

This study found broad consensus on how to assess grafts during A-NRP or TA-NRP (Table 10) and how to perform the sequence of a multiorgan donation with NRP, as well as on how to successfully retrieve the heart and lungs using TA-NRP.

There was a great consensus in the recommendations regarding the evaluation of organs in donors undergoing NRP, with routine biopsies for the evaluation of kidneys and liver considered not necessary. There was a wide consensus in the use of serial transaminases and serial lactate as useful ancillary test to assess liver function during NRP as was also detailed in the recent ELITA consensus report [17]. However, the systematic use of EVLP for the evaluation of lungs in DCDD donors remains a matter of considerable debate. Certain programs report a high proportion of lungs being assessed with EVLP [49], whereas others employ this modality only infrequently [18, 19, 50, 51]. The authors of this report considered that EVLP should be recommended for lung evaluation or reconditioning only in special cases, and not for routine cases (Table 10, point 2).

Since the use of TA-NRP is still in its early stages, different protocols exist regarding how to assess the heart and how to coordinate its maintenance with that of the other organs [25, 50, 52]. There was a clear agreement that when optimal cardiac function is achieved, TA-NRP should be discontinued as soon as possible to facilitate the recovery of the stunned heart (Table 9).

One of the major challenges in TA-NRP for the panelists were those situations where the procedure involves multiorgan recovery, which could risk the other grafts because of problems with the heart. The panelists agreed that a designated person in the team should ideally be responsible for determining that TA-NRP has been unsuccessful and shifting the team’s focus to the recovery of lungs and abdominal organs. This named person has demonstrated a great benefit in Spain, where the donor transplant coordinator, mostly an intensive care physician, has a pivotal role in organizing the different surgical teams participating in NRP procedures, particularly when TA-NRP is used [9].

The ethical concerns regarding NRP have substantially impeded the adoption of this technique worldwide [53, 54]. It was notable that more than 90% of panellists disagreed with a series of normative and empirical claims that are often presented as justifications for prohibiting NRP (Table 12). It is evident that the most critical ethical concern relates to the respect for the “dead donor rule”, with consensus among panellists that interventions are essential to ensure that the permanency standard for the determination of death is maintained during TA-NRP by preventing cerebral reperfusion.

There was a consensus that effective techniques must be used to maintain permanent cessation of cerebral perfusion and the delivery of oxygenated perfusate to the brain during NRP. Indeed, clamping or ligation of the aortic arch vessels and sectioning of the cephalic ends of the vessels, plus drainage and aspiration of blood to atmospheric pressure during TA-NRP was considered sufficient to ensure this. However, ligation of the arch vessels alone was not considered an effective technique to maintain permanent cessation of brain circulation during TA-NRP (Table 13). There is thus a critical need for research that will establish interventions that conclusively demonstrate the absence of brain perfusion after TA-NRP [55]. Two recent studies monitoring intracerebral blood pressure circle of Willis and using perfusion scintigraphy with technetium-99m hexamethylpropyleneamine oxime as a radiotracer during TA-NRP provided the first strong evidence that permanent cessation of brain perfusion could be maintained by clamping and sectioning the supra-aortic vessels [39, 46]. It remains unclear whether clamping of the supra-aortic vessels in isolation, which has been considered acceptable in some parts of the U.S, excludes any risk of postmortem cerebral perfusion [45, 46]. The US guidelines accepted both options [47].

Emerging evidence regarding the absence of cerebral perfusion during TA-NRP [39, 46], together with recent studies providing direct evidence of the temporal relationship between the loss of brain blood flow, brain activity, and systemic circulation in humans [56], helps strengthening the confidence within the scientific community—and may facilitate a more widespread adoption of NRP in the setting of DCDD.