Management of Patients with Cardiogenic Shock on Temporary Mechanical Circulatory Support: Urgent Transplantation or on to the Next Pump?

This article refers to 'Clinical outcomes of temporary mechanical circulatory support as a direct bridge to heart transplantation: a nationwide Spanish registry' by E. Barge-Caballero et al., published in this issue on pages 178–186. Cardiogenic shock (CS) remains a major challenge for heart failure specialists, cardiac surgeons and intensivists. Acute coronary syndrome with or without mechanical complications is the underlying aetiology in the majority of patients and a smaller proportion develops CS on the basis of worsening chronic heart failure or more rare causes, such as myocarditis.1 Despite some improvement in mortality rates in subgroups of patients with CS, mortality remains around 50% in most studies.1, 2 In the initial phase, inotropic support is usually applied, but this approach is unhelpful in many patients as most inotropes increase myocardial oxygen consumption and do not promote cardiac recovery or reversal of CS. A recent European consensus statement concludes that use of multiple inotropes or placement of an intra-aortic balloon pump are not effective interventions, but recognizes that short-term mechanical circulatory support (MCS) can be used in patients in whom medical therapy fails.3 Temporary MCS (t-MCS) may be provided by short-term percutaneous devices, including easily placed pumps that give partial left ventricular (LV) support, such as the Impella 2.5® and Impella CP® (Abiomed, Inc., Danvers, MA, USA). In some patients, pumps that provide greater support, for instance the Impella 5.0® or the TandemHeart® (CardiacAssist Technologies, Inc., Pittsburgh, PA, USA), are required. In the event of biventricular failure, a veno-arterial extracorporeal membrane oxygenation (VA-ECMO) device can be placed percutaneously or via sternotomy. Sternotomy is also required for intermediate-term extracorporeal biventricular assist systems, such as two Levitonix® CentriMag® (Levitronix LLC, Waltham, MA, USA) or Excor® (Berlin Heart GmbH, Berlin, Germany) pumps. Finally, temporary isolated LV support can be provided by systems like the Levitronix® CentriMag® pump implanted via sternotomy.4, 5 The latter systems can be left in place for several weeks and enable patients to ambulate to some extent while supported by the pump(s). In a substantial proportion of patients, end-organ function recovers after implantation of these devices, particularly with use of those that provide full LV or biventricular support according to the patient's needs. The real problem, however, remains: what do we offer to those patients who recover end-organ function but in whom cardiac recovery is inadequate to allow for ventricular assist device (VAD) removal? In reality there are three options: (i) exchange of the intermediate-term VAD for a durable LV assist device (LVAD); (ii) palliation (equivalent to death on the device), and (iii) urgent transplantation. Information about outcomes in patients managed with these strategies, especially urgent transplantation, is sparse in the literature. In this issue of the journal, Barge-Caballero and coworkers describe outcomes in 291 patients treated with t-MCS as a bridge to urgent heart transplantation.6 The cohort comprised patients treated in 16 Spanish transplant centres during 2010–15. Patients were supported with a temporary LVAD, biventricular assist device (BiVAD) or VA-ECMO. Impressively, almost 80% of patients underwent transplantation. Survival rates from listing to heart transplantation to discharge ranged from 54% to 79% and were highest among LVAD patients. Patients managed with ECMO had the highest risk for adverse events. In-hospital mortality rates in those who received a donor heart were 12%, 26% and 33% in patients bridged on a temporary LVAD, BiVAD and ECMO, respectively. Increasing age was an important, independent predictor of outcome. The authors conclude that it is reasonable to bridge patients with t-MCS for urgent listing for heart transplantation.6 The paper6 is important in several ways. Firstly, it represents a large, well-accounted, multicentre experience with urgent listing for heart transplantation in patients on temporary circulatory support. Secondly, it highlights the fact that high-level co-operation among transplant centres enables the provision of donor organ offers to a large proportion of patients on t-MCS within a reasonable timeframe. Thirdly, it clearly confirms that isolated LVAD support is associated with far fewer complications than BiVAD or ECMO, which are truly plagued by complications. The paper6 also raises several questions. It is clear that the results described can be obtained only in areas with both a high organ donation rate and a well-functioning allocation system, such as the highly renowned Spanish system. However, even if both these conditions are available, whether or not the strategy presented is reasonable is debatable, given the outcome, especially in patients transplanted from BiVAD and VA-ECMO. Donor organs are scarce, even in areas with high levels of donation, and in the vast majority of cases the allocation of an organ to one patient means the denial of transplantation to another. Donor organ shortage is increasing in Europe and waiting time for a heart transplant in the Eurotransplant area has tripled since the year 2000.7 Organ allocation should take into account the chance of long-term survival of the donated organ. Recent data from the International Society for Heart and Lung Transplantation (ISHLT) (2009–2014) reveal a 1-year survival rate of 85% after heart transplantation in general, and in transplant organizations with low proportions of urgent transplantation, the figure is even higher.8 Although survival rates in the Spanish registry6 of patients transplanted from t-LVAD may compare with the ISHLT data, the results of patients on VA-ECMO do not. Indeed, an inferior outcome in patients transplanted from VA-ECMO is also documented in the recent ISHLT report.9 Hence, the offering of potentially lifesaving therapy to desperately ill patients with CS in whom inferior outcomes are anticipated must be balanced with that offered to elective patients in whom excellent long-term survival after cardiac transplantation is expected. An alternative strategy to the urgent listing of patients with CS on t-MCS involves using a bridge-to-bridge approach. Although it is clear that the outcomes of implanting durable LVADs directly in unsupported, unstable patients with CS [Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) levels 1 and 2] are unsatisfactory, and the strategy is generally advised against,10 it is possible to transition patients on t-MCS to a durable LVAD with acceptable survival. A recent meta-analysis documented reasonable survival to discharge in such patients11 that reflected the range of outcomes reported in the study by Barge-Caballero et al.6 At the least, it is arguable that the survival achieved is more acceptable after LVAD than after transplantation as donor organs are scarcer than LVADs. The patient on a durable LVAD can subsequently be electively evaluated for transplantation when end-organ function has recovered (i.e. a bridge-to-decision implantation). Outcomes after transplantation in patients on long-term, continuous-flow LVADs are now similar to those in elective patients not transplanted from MCS.9 The strategy has several advantages. Most importantly, the use of donor organs in patients with inferior post-transplant outcomes is minimized compared with the 'direct from t-MCS listing' approach. Further, for some patients, remaining on LVAD as destination therapy may turn out to be preferable, either because of other medical issues or simply because of patient preference. Finally, some patients may experience cardiac recovery that enables removal of the LVAD and may not require a donor organ after long-term LVAD support.12 In this context it should be noted that the average time on LVAD support to explant for recovery is approximately 1 year.13 Disadvantages of the bridge-to-bridge method include the need for repeated surgery and consequent increase in costs. Moreover, the strategy is not applicable in all cases. Transition to a durable LVAD requires adequate right ventricular function to sustain satisfactory circulatory function on the LVAD. This is, obviously, available in most patients on t-LVAD, but is far from always so in patients on VA-ECMO and rarely in patients who initially require t-BiVAD support. Hence, in the two groups with the poorest outcomes after urgent transplantation, successful transition to a durable LVAD will often be challenging or impossible. One option in t-BiVAD and VA-ECMO patients is to 'downgrade' support and evaluate haemodynamics on t-LVAD prior to moving the patient on to a durable LVAD, but this is a risky journey. However, it may be preferable to take that road rather than to lose a large number of good donor hearts that may provide other waiting patients with perhaps more than 20 years of excellent quality of life. The discussion above highlights the need for better heart replacement systems, particularly with the possibility of biventricular support. So far, total artificial hearts have been associated with high incidences of complications,14 but new technology, such as that in the Carmat heart (Carmat SA, Velizy Villacoublay, France) or other emerging biventricular replacement or support solutions, may prove to allow better outcomes.15 Such devices, together with the current, highly efficient LVADs, may provide us with an armamentarium with which to bridge all patients with CS initially stabilized on t-MCS to an ambulatory life on a durable device. Then, decisions on transplantation may be taken under much better conditions than in the setting of a desperately ill patient on t-MCS in the intensive care unit. Until this technological step forward is made, heart transplant physicians and surgeons must remain critical when evaluating patients on BiVAD or VA-ECMO for urgent heart transplantation, particularly older patients, and should bear in mind the needs of not only the patient in front of them, but also those of the long line of elective patients on the waiting list. Conflict of interest: Speaker's fee: Abbott.