To ECMO or not to ECMO: Do we all Transplant in the Same Theater?

During lung transplantation, several factors intrinsic to lung transplantation itself, and out of any surgical accident, may impair the recipient’s hematose and hemodynamic status. These intraoperative situations may be anticipated either before or intraoperatively and are related to the orthotopic characteristic of lung transplantation. To perform lung transplantation, one has first to perform pneumonectomy of 1 of the diseased lungs before implanting the new lung graft. Before and during pneumonectomy, the first surgical steps are reversible and the surgeon will be able to test whether extracorporeal support is needed to operate further on. In pulmonary hypertension, extracorporeal support is conditioning pneumonectomy before implantation. Once pneumonectomy is completed, hemodynamic failure, surgical tear, or accident during implantation or massive reperfusion injury will represent an acute situation for which extracorporeal membrane oxygenation (ECMO)/extracorporeal life support (ECLS) was not expected but needs to be decided and implanted rapidly, mostly unplanned. An important point to notice is that extracorporeal support may eventually be needed even if all tests were supporting an off-pump strategy. Moreover, extracorporeal support may be urgently needed intraoperatively despite a perfectly conducted surgery, especially in case of acute primary graft dysfunction. While most of the lung transplantation community does agree a form of ECLS is mandatory in selected indications such as pulmonary hypertension, the lung transplantation community is still wondering what to do for the other recipients. The latest consensus statement of the International Society for Heart Lung Transplantation advocates for a specific preemptive discussion during the dedicated listing multi disciplinary team. Risk factors surrounding the surgical procedure may help decide whether ECMO should be used during transplantation. The ECMO would not be systematically assigned but several factors may then decide in favor of the ECMO: the underlying lung disease, the pulmonary artery pressure regimen, and the overall risk for impaired hemodynamics during the surgery (preoperative right and left ventricle function).1 Besides, another risk surrounding lung transplantation is much more difficult to predict: to date, no biomarker and no clinical factor is strong enough to precisely predict whether severe primary graft dysfunction is going to occur when unclamping the first implanted lung. Clinical series describe up to 20%–30% of all patients experiencing primary graft dysfunction (PGD). All lung transplant surgeons have at least 1 story to tell of an uneventful lung transplantation ending with sudden severe edema, fluid in the intubation tube, and a rough half hour after unclamping the first graft. No specific treatment is available, and only symptomatic measures, such as ECMO and waiting for the inflammatory storm to calm down, will help salvage the patient Bermudez et al.2,3 Should we, therefore, advocate for systematic ECMO in all lung transplantation recipients as a protection against PGD, but also provide enhanced intraoperative hemodynamic stability and protection of the first unclamped lung graft? Yet, ECMO is no magic formula: it will need heparinization of the recipient during the surgery and will activate inflammation per se. There is a world between systematic preemptive ECMO in all recipients and salvage ECLS for acute reperfusion injury or surgical accident. In high-volume, well-trained teams dedicating a whole part of their logistics to lung transplantation, the systematic use of ECMO ends with excellent results.4 But we are all different. Centers perform a range of volumes from 15 to 120 procedures a year, some have dedicated highly experienced anesthesiologists, some do not, some rely on high-volume center cardiac surgeons, others on purely thoracic surgeons with various ranging skills with all forms of circulatory bypasses.5,6 Could the high heterogeneity of practices explain for the differences of reports in international literature and the absence of definitive conclusion. That is what meta-analysis are for. In their meta-analysis, Passos et al6 pool a large number of patients with a time to event analysis7 differentiating 3 groups: planned ECMO, unplanned ECMO, and off-pump strategy. Differentiating planned and unplanned strategy gives a detailed look at what usually is the main hurdle when analyzing the outcomes of ECLS in lung transplantation: the unplanned emergency. The difference was categorized after analyzing the studies included in the meta-analysis to distinguish the different subgroup of ECMO implantation. ECMO as a bridge to transplantation was excluded from the final analysis and was not considered to define the different subgroups. Planned ECMO was defined as the preemptive initiation of mechanical support in all lung transplant recipients or in patients showing instability during intraoperative hemodynamic assessments (eg, pulmonary artery clamping before pneumonectomy). In other words, ECMO was considered as planned in each situation where the surgeon may test the need for intraoperative support. In contrast, unplanned ECMO referred to a rescue intervention in response to complications such as severe hypoxemia, hypercapnia, or hemodynamic deterioration occurring intraoperatively with no prior test anticipating the emergency. In studies where it was not possible to distinguish between planned and unplanned ECMO, cases were classified simply as ECMO. So far, except for pulmonary hypertension, most of the studies aiming at assessing the help provided by the use of ECMO in lung transplantation have assigned ECMO to all recipients or to none. Yet in their study, when compared 2 strategies relying on ECMO, the authors found the off-pump strategy turns better when considering intubation duration, ICU stay, and survival during the first 6 mo. Compared against planned ECMO or against unplanned ECMO, the off-pump strategy performs the best overall survival. In their meta-analysis, Passos et al6 make the demonstration of a room for an off-pump strategy with better outcomes. In other words, the off-pump strategy has to be considered as a valid indication in a group of patients that does not benefit from ECMO and that group should not be included in studies anymore. Besides, the meta-analysis shows ECMO confirms as a surrogate of the high-risk profile of patients being associated with PGD. Indeed, the high-risk profile of those patients probably explains why they are actually being analyzed in the planned ECMO group7 so studies should be aimed at identifying more precisely this high-risk profile. The planned ECMO strategy for high-risk patients has yet to be worked on by further studies aiming at identifying and sorting out risk factors to reduce as much as possible the numbers in the unplanned strategy.