Improving Long-Term Outcomes with Left Ventricular Assist Devices—Referral, Selection, Experience, and Technology

This article refers to ‘Long-term evaluation of a fully magnetically levitated circulatory support device for advanced heart failure—two-year results from the HeartMate 3 CE Mark Study’ by J.D. Schmitto et al., published in this issue on pages xxx. Advanced heart failure (HF) was recently defined and characterized in a position statement of the Heart Failure Association of the European Society of Cardiology.1 It affects 1–10% of patients with HF and entails progression toward organ failure and death despite maximum tolerated medical therapy and cardiac resynchronization therapy if indicated. For these patients, the only remaining alternatives are heart transplantation (HTx) which is available only to a minority of patients,2 and durable mechanical circulatory support, primarily in the form of an implantable left ventricular assist device (LVAD), as permanent so-called destination therapy (DT-LVAD).3, 4 Is DT-LVAD the future in advanced HF? This will require favourable long-term outcomes. The long-term benefit of LVAD therapy is limited primarily by thrombosis in the pump, causing pump malfunction and stroke, and requiring anticoagulation leading to increased risk of bleeding; infection primarily due to the external driveline connecting the internal pump to the external controller and batteries; and by late right ventricular failure. Until recently, the predominantly used LVADs were the HeartMate II and the HVAD. The novel HeartMate 3 has a fully magnetically levitated centrifugal rotor and an artificial pulse feature, both intended to reduce the risk of pump thrombosis, and the pulse also intended to reintroduce to some extent the physiological pulsatility. Recently, the 2-year outcomes of 366 patients in the randomized non-blinded MOMENTUM 3 trial were published,5 showing that the novel HeartMate 3 was superior compared to HeartMate II regarding the composite of survival free of stroke or reoperation due to pump malfunction (which is generally attributed to device thrombosis; there were only two patients who had suspected pump thrombosis in the HeartMate 3 group), and that this superiority was driven primarily by reduced pump replacement and reduced non-disabling stroke. This was hailed as a great success and considered a major development in HF,6 even though death, disabling stroke and rates of other complications were not reduced. In this issue of the Journal, Schmitto et al.7 report detailed 2-year outcomes in 50 patients from the observational European CE Mark trial for HeartMate 3. In this small sample, there were no episodes of pump thrombosis. MOMENTUM 3 was conducted at 69 US centres whereas the CE Mark study was conducted at 10 European centres. Advanced HF care is differently organized in different parts of Europe and between Europe and the US. Most centres in the US were early to organize HTx and LVAD selection and postoperative care around an advanced HF service, whereas in Europe local routines have been highly variable and the role of cardiologists more limited in the early years. As LVADs have become accepted therapy in most of Europe, LVAD patient care has also become more integrated within the HF team and there is now indeed a global registry for durable mechanical circulatory support which will assess quality of care and outcomes and hopefully lead to improvements worldwide.8 It was reassuring that both patients and outcomes with HeartMate 3 in the observational CE Mark study and in the randomized MOMENTUM 3 trial were roughly similar. Patients were a mix of bridge to transplantation and destination therapy candidates, mean age was approximately 60 years, ejection fraction 17–19%, systolic blood pressure 105–110 mmHg, cardiac index 1.8–2.2 L/min/m2, central venous pressure 10 mmHg, and pulmonary capillary wedge pressure 22 mmHg. The major baseline differences regarded inotrope use: >80% in the US vs. 58% in Europe; and intra-aortic balloon pump use 13–15% in MOMENTUM 3 vs. none in the CE Mark study. Regarding outcomes, stroke and sepsis occurred slightly more commonly in the CE Mark study, whereas right HF and gastrointestinal bleeding occurred more frequently in MOMENTUM 3, although in MOMENTUM, right HF was driven almost exclusively by prolonged early inotrope use (>7 days) rather than by right ventricular assist device implantation (3.2%).5 Mortality was slightly greater in the CE Mark study but the sample size too small to make any conclusions. The 6-month outcomes in the CE Mark study compared favourably against a hypothetical expected outcome based on the Seattle Heart Failure Model,9 but no comparison data were reported in the present 2-year outcomes report. Thus, durable LVAD therapy is established in Europe and the US and outcomes have improved, but outcomes are not good enough for universal acceptance of long-term destination therapy. What is still needed? There has been extensive research to improve patient selection for HTx and LVAD, moving from peak oxygen consumption and inotrope dependence to multi-marker scores incorporating clinical characteristics and laboratory, haemodynamic and imaging data.10, 11 These are used to assess both indications and contraindications and to maximize the net clinical benefit.12, 13 However, many patients are referred too late and simpler tools14 and better organization of care are needed to ensure timely referrals (which is different from selection) to the advanced HF centre and team, and it may even be useful to screen patients with New York Heart Association III–IV HF for potential indications for HTx and/or LVAD.15 Finally, technology has come far but leaves much to be desired. It is disappointing that most complications were not improved with HeartMate 3 as compared to older technology, that there is still an external driveline and as of yet no clinically available completely implantable system, and that costs and complications are excessive, diminishing the otherwise excellent quality of life improvement experienced by most LVAD recipients, and consuming excessive resources making cost-effectiveness of DT-LVAD difficult to establish. Conflict of interest: L.H.L. is an investigator in the ongoing randomized SweVAD trial, evaluating the HeartMate 3 as destination therapy, sponsored by Abbott, the Swedish Research Council, and the Swedish Heart-Lung Foundation (ClinicalTrials.gov Identifier: NCT02592499).