Article Text
Statistics from Altmetric.com
Geographical variations in healthcare lend us valuable insight in how a complex array of patient and external factors affects disease processes, treatments and outcomes. Differences in patient demographics, genetics, socioeconomic conditions, clinical practice patterns and regulatory policies provide us with empirical data and sometimes natural experiments to answer questions that are either impractical or unethical to answer with randomised control trials. In the realm of advanced heart failure therapies, ventricular assist devices and heart transplantation have been widely adopted therapies in the USA, Canada and Europe. Facing different regulations on device indication (bridge to transplant (BTT) vs destination therapy (DT)), donor heart availability and heart allocation priority systems, how does the experience of left ventricular assist device (LVAD) therapy compare in these different regions? Does longer transplant waiting time significantly impact the effectiveness and complications of BTT LVAD therapy? And more importantly, what lessons can we learn from each other’s experience?
In their Heart paper, Parameshwar et al add to our collective LVAD experience by describing the survival and outcomes of the BTT LVAD population in the UK.1 In the UK, LVAD therapy is funded as a BTT at six approved centres and destination therapy is not funded. All centres are required to submit LVAD therapy data to a national database, adding strength to the comprehensiveness of the study. Utilising this database, the authors analysed all 342 patients who received a HeartMate II or HeartWare HVAD from January 2007 to December 2013 (HeartMate 3 not yet available) and made a few important observations. First, compared with the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) registry in the USA, BTT LVAD patients in the UK had lower rates of heart transplantation (cumulative incidence of 20% at 3 years in the UK vs 31% transplant rate at 1 year and 50% at 2 years in INTERMACS) and lower overall survival rate (49.6% vs 59% at 3 years). Second, there were no significant differences in survival or adverse event rates (stroke) between the HeartMate II and HVAD devices. Last and perhaps most intriguing, while overall stroke event rates appear similar to INTERMACS, stroke was the most frequent cause of death in the UK registry accounting for 34% of deaths in contrast to only 18% of deaths in INTERMACS.
There are several important limitations the authors appropriately acknowledge and are worth emphasising. First, the number of patients in the UK registry is small compared with large registries such as INTERMACS. The small number of adverse events particularly limits statistical power. Second, even though patient inclusion is mandatory, the registry is not comprehensive and hence analysis lacks information on many key types of adverse events and treatment metrics. For example, infection, bleeding, right heart failure and hospitalisation rates are absent. Also, as strokes have diverse presentations and consequences, clear stroke definitions and documentation of severity would have been helpful to explain the high rate of stroke-related death in the UK population. In addition, blood pressure has a strong association with stroke rates and information on its management was not collected by the registry. Nevertheless, these limitations should not distract us from the central insight of this study: with lower transplant rates and longer device support time, LVAD patients suffer more complications and death, most notably due to strokes.
The UK experience may be the harbinger of what is to come in the USA. According to the INTERMACS registry, the overall rate of LVAD implantation in the USA is increasing, and despite destination therapy growing as a share of all LVAD implantations, the absolute number of both BTT and DT LVAD implantations per year is rising.2 In the context of near static heart transplant rates, the median waiting time to transplantation and the percentage of LVAD patients on the wait list has steadily increased since 2006.3 In addition, the new United Network for Organ Sharing heart allocation system coming into effect this fall aims to further stratify the risk of LVAD patients by lowering the relative priority of stable LVAD outpatients (https://optn.transplant.hrsa.gov/media/2414/adult_heart_infographic.pdf).4 Thus, the overall transplant rate of the BTT LVAD population in the USA may trend towards the UK rate, making the UK BTT LVAD experience especially relevant.
In this context of more LVAD patients waiting for a longer time on the transplant list, where shall we focus our efforts to further reduce LVAD complication rates and improve survival? If we use Parameshwar et al’s analysis as a guide, then reducing neurological events is the most important, as stroke accounts for 34% of deaths in the UK registry. There was hope that newer generation devices such as the HeartMate 3 may have lower risk of stroke. However, a comparison of neurological event rates in key registries and clinical trials shows fairly consistent rates of total neurological events and stroke across devices (table 1). For the HeartMate 3, even though total stroke rate appears lower compared with the HeartMate 2 in the Multi-centre Study of MagLev Technology in Patients Undergoing MCS Therapy With HeartMate 3 (MOMENTUM 3) clinical trial, disabling stroke, fatal stroke and other neurological events including transient ischaemic attack, actually occurred more often compared with HeartMate II.5 Device engineering, therefore, has thus far not had a significant impact on stroke rates, and a randomised clinical trial between the HeartMate 3 and HVAD devices is unfortunately lacking.
The good news is that we now have evidence that better medical therapy, in this case intensive blood pressure control, can reduce stroke risk. The recently published A Prospective, Randomised, Controlled, Unblinded, Multi-Centre Clinical Trial to Evaluate the HeartWare Ventricular Assist System for Destination Therapy of Advanced Heart Failure (ENDURANCE) Supplemental Trial showed that a blood pressure control protocol to lower mean arterial pressure to 85 mm Hg or less reduced overall stroke rate and cut haemorrhagic stroke rate by half.6 The bad news is that many clinical trials still have varying definitions of neurological events and different reporting methods (percentage of patients with event vs event rate per patient-year), making cross-study comparison extremely difficult. To have the best chance of success, stakeholders need to agree on a universal set of definitions for neurological events stratified by type and clinical severity to be used in all future clinical trials and registries. Additionally, proactive blood pressure management protocols should be implemented at every LVAD centre and blood pressure data reported in every clinical trial.
For future research, we should redouble our efforts of optimising medical management strategies in LVAD patients aimed at reducing complication rates and improving survival on support. The blood pressure management protocol to reduce stroke risk in the ENDURANCE Supplemental Trial is an encouraging example, but also underscores how little we know about this important subject. We still lack the evidence base to answer fundamental questions such as how to most accurately determine blood pressure in continuous flow LVAD patients non-invasively and what mean arterial pressure target is optimal. Going forward, these questions are going to be the key to success in further reducing stroke risk.
The UK BTT LVAD experience described by Parameshwar et al grants us a window to peak at a landscape where BTT LVAD patients succumb to device complications due to a paucity of donor hearts and relatively unfavourable heart allocation priority, a view that is ever more familiar to the rest of the world. It is our opinion that a deliberate and proactive effort to research and implement optimal medical management strategies to reduce LVAD complications, especially strokes, is urgently needed.
Footnotes
Contributors Both authors contributed equally.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Disclaimer Dr. Mahr is a consultant for Abbott, Medtronic and Abiomed
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Commissioned; internally peer reviewed.
Linked Articles
- Heart failure and cardiomyopathies