The impact of normothermic machine perfusion on coagulation function and transfusion practice: Miracle machine or anticipatory bias

The use of normothermic machine perfusion (NMP) in liver transplantation (LT) over the traditional preservation method of static cold storage (SCS) has been shown to improve posttransplant outcomes1–3 while expanding the use of marginal allografts.4,5 However, studies with real-time intraoperative data demonstrating the quality of NMP allografts are sparse. In this issue of Liver Transplantation, Su et al6 reviewed their experience with 186 recipients of donations after circulatory death LT, of which 99 (53%) were preserved in SCS and 87 (47%) were placed on NMP using the TransMedics Organ Care System. The synthetic coagulation function of the allografts, as reflected in thromboelastography (TEG) monitoring and conventional coagulation testing, were compared between the NMP and SCS groups. In addition, the utilization of blood products and tranexamic acid (TXA) was compared. The authors found that NMP was associated with increased synthetic coagulation function and less fibrinolysis in the post-reperfusion period (measured at 10 and 70 min after reperfusion). Of note, this difference was only observed in the TEG, but not conventional coagulation testing, parameters. Furthermore, there was a significant decrease in blood product and TXA usage in the NMP group during both the anhepatic and post-reperfusion periods. In the last decade, NMP has emerged as a promising innovation in LT, with several studies suggesting a reduction of early graft dysfunction, primary nonfunction, and ischemic cholangiopathy.1–3 Based on the success of the landmark Liver PROTECT randomized controlled trial,1 the first NMP device was granted Food and Drug Administration (FDA) approval in September 2021. Since then, NMP has allowed transplant centers to enhance their use of extended criteria donors beyond traditional cutoffs and improve access to transplantation.4,5 Nevertheless, the high financial investment and operational burden associated with NMP continue to pose significant challenges in its broader implementation as the new clinical standard. This study by Su and colleagues is a welcome addition to the growing body of literature that supports the expanded utilization of NMP in LT. One of the key findings in this study is the reduction in blood product and TXA utilization in the NMP group compared with the SCS group. While this analysis is intriguing, it raises questions about whether the difference truly reflects NMP’s favorable impact on the allograft, or if this benefit is driven by clinicians’ perception of NMP’s superiority to SCS. This consideration is particularly relevant for the anhepatic phase of the operation, which precedes implantation and should theoretically be similar for both groups. Could the difference, then, stem from a bias in which clinicians anticipate greater post-reperfusion hemodynamic stability in the NMP group, leading to more judicious use of blood products and TXA? While the clinical interpretation of TEG parameters may indeed explain the observed difference during the reperfusion phase, the lack of standardized transfusion protocols introduces subjectivity and the potential for bias. In this study, red blood cells were transfused with the goal of maintaining hemoglobin above 8 g/dL, while other blood products were administered “according to institutional practices based on correcting TEG abnormalities, abnormal conventional coagulation testing parameters, and discussions with the surgical team regarding visualization of clot formation in the surgical field”6—guidelines which are open to varying interpretations by a diverse group of clinicians. In addition, as the authors noted themselves, the study was limited in matching the precise timing of transfusion relative to laboratory results, which further obscures our ability to understand the cause of the difference between the 2 groups. In general, there has been considerable speculation about NMP’s influence on clinician and transplant center behavior.4,5 This study, like many others, faces the challenge of distinguishing between NMP’s actual capabilities versus the impact that the perception of its effectiveness has had on our clinical decisions. In spite of these caveats, the transformative nature of NMP in the field of transplantation cannot be overstated. The work of Su et al advances our understanding of this innovative technology by providing real-time intraoperative data on allograft quality, an area which has previously been underexplored. By demonstrating improved synthetic coagulation function and reduced fibrinolysis in NMP-preserved livers, this study offers valuable insights into the physiologic benefits of NMP. A natural question that arises as a potential avenue for further work is whether TEG parameters could be used as indicators for allograft quality before utilization. Viability assessment during ex-situ machine perfusion is a topic of ongoing research with currently little consensus regarding the most suitable laboratory markers or their precise cutoff values.7 As the financial cost of NMP continues to raise questions about and barriers to its use, it is paramount that all the benefits of this new technology also be tracked to enable a comprehensive cost-benefit analysis. By demonstrating a reduction in resources such as blood products, this study could potentially justify broader utilization. While the fixed costs of NMP are readily transparent, it is critical to also demonstrate any cost savings both from the hospital and patient perspective. For example, potential reductions in hospital stay duration, intensive care unit utilization, operating room staffing needs, and rates of physician burnout are likely to lead to cost savings. Furthermore, creating opportunities for earlier transplantation and increasing access for patients with lower MELD scores could result in lower health care costs and increased productivity by optimizing quality-adjusted life years.