Supplementary Table reports the results of the review on original pre-clinical studies involving EVHP in the context of DCD. A total of 34 original articles were identified from 2013 to 2023, involving 774 DCD hearts perfused with MP after procurement. The overall study objectives can be classified into three categories.

• Primarily graft evaluation

Historically, early studies aimed to evaluate graft function in DCD pre-transplant settings. The main objectives of these studies were to assess cardiac viability and performance following a period of warm ischemia, as well as to identify valid indicators of graft injury/quality. Through organ reperfusion, the goal was to identify criteria that could distinguish grafts amenable to heart transplantation from those that were not. The first reported study, dating back to 1997, utilized a custom-made MP device with an erythrocyte-enriched perfusate solution to investigate the effects of normothermic ischemic insults of varying durations [19]. In this study, coronary flow response during reperfusion was considered a key parameter for assessing recovery of power output. Subsequent studies identified additional parameters reflecting organ damage and predictive of myocardial function quality, such as direct hemodynamic or echocardiographic parameters, in particular systolic function, diastolic function, left-ventricular end systolic pressure-volume relationship, left ventricular maximal developed pressure, preload recruitable stroke volume, isovolumic relaxation constant [20–45], histological markers [23, 24, 26, 27, 29, 31, 32, 38–41, 45–47], vascular and microvascular analysis [25–27, 32, 48, 49], laboratory markers indicating metabolism, oxidative stress, etc. of whom the most and easiest used was lactate level [21–24, 30–33, 38–43, 45–48, 50–55] and gene and protein expression [26, 43, 44, 55]. These early studies laid the groundwork for the second category of studies, which focused on the optimization of the primary graft preservation protocol. These studies aimed to identify the most effective organ preservation protocol, including optimal preservation temperature, cardioplegic solution type, and enrichment of the solution with erythrocytes, albumin, etc. Additionally, these studies sought to identify organ evaluation metrics that could predict organ quality for transplant use. Various parameters were identified, including laboratory indices of myocardial damage, oxidative stress, histological markers such as tissue edema and electron microscopy alterations, and the maintenance of adequate echocardiographic or hemodynamic parameters.

Progressive advances in this field shifted research focus from graft evaluation to potential therapies that could be associated with MP to preserve myocardial function and oxygen consumption, reduce ischemia-reperfusion injury (IRI) mechanisms, and facilitate organ reconditioning. Among the therapeutic approaches explored were the use of CytoSorb filters [43], IL-11 [45], methylprednisolone [40], Intralipid [50], autologous mitochondrial transplantation [41], melatonin [42], the NLRP3 inflammasome inhibitor Mcc950 [55], and HSP90 inhibitors [44]. In the early studies, the rat model was predominantly used, whereas later, more advanced translational studies employed porcine models.

Table 2 presents the results of the review on original human studies involving EVHP in the context of DCD. We identified 21 original articles from 2013 to 2023, involving 866 DCD hearts perfused with MP after procurement and subsequently transplanted. A total of 268 DCD hearts were declined for transplantation. All studies used the OCS Heart as the sole MP device in controlled DCD heart transplantation settings. Initially, the purpose of using MP was to reduce ischemia time in distantly-procured transplantations or in recipients with adverse risk factors, such as those on ventricular assist device support [13, 59, 61, 65].

Parallel to these studies, parameters were examined that characterize DCD grafts in comparison to DBD settings [57, 62, 64, 66, 67, 69–71] or between DCD grafts preserved via cold storage versus those preserved using MP [12, 60, 69, 71, 72]. Among the most relevant parameters, Dhital et al. demonstrated a differential role of lactates during OCS in DCD hearts, showing that higher lactate values (>5 mM) in DCD grafts were still acceptable as long as there was a downward trend in lactate levels and an arteriovenous differential favoring lactate metabolism as an energy source, rather than its accumulation, which would indicate myocardial injury [61]. Additionally, studies have been published examining diagnostic insights into DCD organs under MP, such as coronary angiography or cardiac magnetic resonance imaging [62, 65]. Both short- and long-term outcomes, including survival rates and rejection episodes, were found to be comparable to outcomes from DBD donors [13, 56–72].

In DCD the inevitable warm ischemia time (time between withdrawal of life support and circulatory arrest) makes the already energy-depleted heart poorly tolerant of additional cold ischemia caused by transport with well-established cold static storage (CSS) [10]. Alternative ways of organ preservation can be preferred in this context to improve patients’ outcomes. Ex-situ MP can be divided into normothermic MP (NMP; 34°C–37.5°C), hypothermic MP (HMP; 1–8°C) and subnormothermic MP (sNMP; 20°C–33.5°C) [74]. NMP is defined as heart preservation in a beating, unloaded, metabolically active state on a portable device. Currently, the only commercially available option for clinical practice is the Organ Care System (OCS; TransMedics, Andover, MA) [8, 70, 75]. The OCS device has recently been awarded approval by the Food and Drug Administration for use in DCD transplantation. HMP is defined as heart preservation in a non-beating state with low myocardial metabolism and perfused with oxygenated nutrient-rich perfusate. Current available devices are.

• The XVIVO NIHP system (XVIVO Perfusion AB, Göteborg, Sweden), which is still under clinical investigation (ongoing clinical trial),

One of the main issues in DCD is graft quality evaluation before implantation. During DCD, organs can either be directly procured (Direct procurement and perfusion, DPP) and Thoracoabdominal normothermic regional perfusion (TA-NRP) can be used. The latter consists in using veno-arterial extracorporeal membrane oxygenation (VA-ECMO) to re-perfuse the allograft in the donor immediately after pronouncement of circulatory death, with the goal of allowing the heart to recover from the warm ischemic injury, allowing for a detailed evaluation of cardiac function and pump function in situ. Then, the graft is procured if deemed suitable for use [76] and either maintained with ex situ MP or with CSS during transport.

If DPP is chosen, the pre-retrieval function is incompletely assessed. In this setting, NMP enables a quantitative heart assessment via measurement of heart rate and rhythm, pump flow, coronary flow rate, aortic pressure, hematocrit, temperature oxygen saturation and arteriovenous lactate levels trend. Nevertheless, ventricular contractility is evaluated only via subjective visual examination – epicardial echocardiography is technically demanding and not standardized – in the setting of an unloaded heart (as the left ventricle is actively vented during perfusion in resting mode) [10, 17, 75]. A recent systematic review highlighted the role of coronary angiography performed while the graft was already connected to NMP [77]. So far, nine total cases of ex‐situ coronary angiography of donor human hearts plus one experimental porcine model have been reported. This technique is particularly useful in DCD because coronary angiography is not always available due to ethical issues and availability in procurement centers [15, 57]. In some institutions even echocardiogram is not routinely available because it is considered an unacceptable antemortem intervention [59].

Although CSS is universally accepted as a safe and effective method of heart preservation, PGD continues to affect approximately 3% of clinical heart transplants performed worldwide, and accounts for 26% of deaths in the first 30 days after transplantation [78]. It has been demonstrated that a continuous perfusion with oxygen and metabolites is a physiological support of aerobic metabolism needed for the maintenance of cell integrity and vital cell functions during the transport period [79–83]. Other potential advantages are myocardial cooling through the native coronary circulation and the ongoing washout of metabolic byproducts resulting in improvement of myocardial preservation and microvascular protection. In CSS settings there is a change in myocardial metabolism and viability: ATP levels are depleted and fail to recover despite reperfusion with all necessary substrates for energy repletion, markers of oxidative injury (MDA), apoptosis (caspase-3), and endothelial dysfunction (ET-1) are induced and are further exacerbated by reperfusion [23, 51, 84]. In contrast, EVHP preserves ATP and maintains baseline tissue pH. As a result, these hearts show a rapid restoration of ATP and shorter period of acidosis during reperfusion, suggesting decreased oxygen debt and improved microvascular recovery with a strong correlation with systolic function [85]. Although EVHP is demonstrated to preserve systolic function, a consistent concern is represented by edema, which may negatively affect post-transplant diastolic recovery [21, 29, 86]. In addition, MP does not evaluate diastolic function.

NMP has also been considered in experimental settings to prevent IRI in DCD and subsequently leads to a lower incidence of PGD and delayed graft failure. This has been achieved in pre-clinical scenarios by enriching the MP perfusate with specific molecules interfering with core steps of IRI, as the nucleotide-binding oligomerization domain-like receptor pyrin domain-containing 3 (NLRP3) inflammasome [55]. Oxidative stress can also be targeted by adding melatonin to cardioplegia and perfusate during EVHP [42]. Even cardiac gene therapy is emerging as a promising approach. It can be obtained both with addition of siRNA in cardiac NMP or via adeno-associated virus specific gene transfer to myocytes [87]. Target genes are involved in inflammatory and apoptotic signal proteins (such as C3, nuclear factor κB-p56, caspase-8, and caspase-3). Additionally, a limited number of studies have reported on the use of mesenchymal stem cells during EVHP due to their anti-inflammatory and immunomodulatory effects both by cell-to-cell contact and by secreting substances [88]. At the moment the focus of these studies is still on feasibility and safety aspects.

To date, more than 50 DCD heart transplants have been performed in Australia and United Kingdom by means of NMP technology with excellent short-term and long-term (4 years) outcomes, comparable with those of DBD transplantation [12, 13, 59].

However, given the high cost of the MP devices themself, comparative cost analyses are needed to elucidate their cost effectiveness. The OCS-DCD US trial was the first randomized trial comparing DCD heart transplant to DBD standard criteria heart transplant clinical outcomes [75]: to date, 180 patients (90 DCD vs. 90 DBD) were enrolled and transplanted. One year patient and graft survival were greater than 90%, with a higher rate compared with DBD. Unfortunately, incidence of moderate-to-severe ISHLT PGD was around 20% in DCD versus 9.1% in DBD, raising the idea that warm ischemia (even though <30 min) can negatively affect early graft function. Chew et al. observed that in the recovery of hearts for DCD, WIT was a crucial determinant of outcome [58]. The need for post-transplant mechanical circulatory support rises when the asystole to cardioplegia (AP) time exceeds 15 min. The interval between circulatory arrest and AP time became a determinant of delayed graft function and the need for short-term ECMO. The full recovery of DCD hearts after short-term support using ECMO was suggestive of delayed graft function as opposed to PGD. On the other hand, Coniglio et al. showed that there were no differences in cardiac MRI findings realized <60 days from HT between those who underwent DCD transplantation using OCS device, DBD using OCS device and DBD transported via CSS, including presence of gadolinium hyperenhancement after transplant (all p > 0.050) [64].