Mechanical circulatory support in the UK

Abstract

It is time to do a trial of left ventricular assist devices for lifetime use

In June the National Institute for Health and Clinical Excellence (NICE) published welcome but bewildering guidelines for short term circulatory support with left ventricular assist devices (LVADs) as a bridge to cardiac transplantation or recovery.¹ Welcome because the guidelines will support funding of these devices but bewildering because few, if any, guidelines for use were actually provided. The limited evidence was derived from the USA and Europe, where LVADs have been used for 20 years, and the guidelines are silent on a third potential use for these devices—their longer term use as a lifetime treatment.

First generation LVADs were designed to replace the failing left ventricle by providing stroke volume and pulsatile blood flow. Blood is taken from the ventricle and pumped in a pulsatile manner into the aorta at a rate of 4-10 litres per minute. These devices provide symptomatic relief, reverse multi-organ dysfunction, and reduce the cytokine and humoral responses to heart failure.² Transplant survival is improved following the use of a device.³ Resting the heart and increasing coronary flow with an LVAD has marked effects on the diseased myocardium. Reduced wall tension and stroke work contribute by decreasing myocyte hypertrophy, apoptosis, myocytolysis, and fibrosis.⁴ Myocyte genetic expression and metabolism change towards normal. As a result LVADs can occasionally be removed after function improves in the native heart (bridge to recovery).⁵ This occurs more often with reversible disease such as myocarditis or intoxication.

Currently a transplant is the only operation recognised by NICE for chronic heart failure and is the sole option in congenitally malformed hearts.⁶ Survival after a heart transplant is currently over 80% at one year and 50% at 10 years.⁷ But in the UK around 150 donor hearts per year are shared among 15 000 patients aged under 65 with stage D chronic heart failure. LVADs are rarely used as a bridge to transplantation. In the EVAD (evaluation of the ventricular assist device programme in the UK) study 70 LVADs were implanted over 32 months.⁸ Thirty patients died before transplantation; 32 completed the treatment plan, by later undergoing a transplant; and four recovered without a transplant. The others remain on LVAD support. Overall 12 month survival was 52%. The mean cost of the LVAD implant alone was £63 830 (97 290 euros, $126 060).⁸ Ongoing costs were not provided but many of the patients remained in hospital pending transplantation.

Rather than provide a bridge to transplantation, LVADs may prove more economically viable for lifetime support in patients not eligible for transplantation. Chronic heart failure generates enormous costs through hospital admissions and the use of pacemakers, implantable defibrillators, and drugs. Yet the impact on symptoms and life expectancy of these interventions is modest. The treatment of renal disease sets the precedent. Haemodialysis, with a 60% two year survival, is offered irrespective of age or eligibility for transplantation at a cost of £34 000 per year.⁹

Lifetime use of LVAD has an evidence base in the REMATCH trial (randomised evaluation of mechanical assistance for the treatment of congestive heart failure).¹⁰ Patients with New York Heart Association class IV heart failure who were not candidates for transplantation were randomly assigned to a pulsatile first generation LVAD or continuing medical treatment. At enrolment 68% required intravenous inotropes; the remainder had a peak myocardial oxygen consumption of 9.18 ml/kg/min, highly predictive of early mortality. Median survival in those assigned to the LVAD group was 409 days versus 150 days for controls.⁸ The 75% annual mortality for controls exceeded that for AIDS and many cancers. The LVAD provided a 48% reduction in mortality during follow up and 27% reduction at one year. With improved selection of patients an initial 21% two year survival improved to 43% later in the trial. The University of Utah now achieves 85% one year and 65% two year survival, similar to that achieved by haemodialysis in patients with renal disease.¹¹

With improved technology, the strategic boundaries between LVAD use for bridge to transplantation, bridge to recovery, and lifetime use no longer exist. The LVAD sustains life while the patient's response determines the clinical course. For instance patients not eligible for a transplant may be rescued with a temporary LAVD and then be switched to lifetime treatment. In future LVADs may also provide the platform for myocardial regeneration by neoangiogenesis or gene or stem cell therapy.

Last July one of us (SW) reported the six year survival in the first patient to receive a miniaturised axial flow pump for lifetime use.¹² The Jarvik 2000 LVAD was tested in laboratory programmes in Houston and Oxford. The 61 year old English patient had idiopathic dilated cardiomyopathy with biventricular failure. He was breathless at rest with pitting oedema to the thighs, ulcerated legs, and ascites. Left ventricular ejection fraction was less than 10%. He was rejected for transplantation because of renal impairment and subsequently declined the procedure. Six years later he has an active life and is in New York Heart Association class II. Pump output is 5.5 l/min against a mean blood pressure of 70-80 mm Hg and usually a pulse pressure of 10-15 mm Hg. Less than 5% of follow up was spent in hospital. The total cost over the six years has been £115 000. This LVAD has had 100% mechanical reliability in the first 150 implants, and lower complication rates than pulsatile pumps.

A clinical trial of lifetime therapy is now indicated. There are no ethical dilemmas: the technology is proved and the patients have short wretched lives. Moreover, there are questions of equity: is it reasonable to treat end stage renal disease but not heart failure at a similar cost? Should people with potentially recoverable hearts (with idiopathic dilated cardiomyopathy or myocarditis) be directed towards transplantation or offered an LVAD first? These issues should be addressed before this technology is absorbed into the health service without proper assessment. Clinical trials rather than observational studies are essential, though modifications of the classical controlled design may be necessary to study patients whose stage D symptoms have already shown the failure of medical treatment. The National Health Service is almost the perfect system within which to organise these trials.