Recovery of Myocardial Function with Long-Term Ventricular Assist Device Support

A rapidly evolving area of investigation in long-term use of the ventricular assist device (VAD) is the potential for left ventricular recovery. Elegant bench and clinical investigative efforts have characterized the changes in histology, physiology, and hormonal milieu when the left ventricle is chronically decompressed. As this new application for mechanical circulatory support continues to evolve, I believe the goal of both basic science and clinical research related to myocardial recovery should be to answer three questions. First, we must define the disease process, or processes, that are most likely to benefit from long-term mechanical circulatory support. Although blood pumps have been successfully removed from patients with ischemic, 1, 2 idiopathic, 1, 3, 4 and peripartum 4 cardiomyopathies, as well as acute myocarditis, 5, 6 the patient population with the greatest potential for recovery of left ventricular function with extended VAD support is unknown. Is it true that patients with a dilated cardiomyopathy are more likely to recover function? Should patients with ischemic cardiomyopathy ever be considered candidates for device removal? How is it best to define, let alone treat, patients with acute myocarditis? Second, we must define reliable indicators of myocardial recovery. A wide variety of hemodynamic and morphologic left ventricular changes have been associated with improved left ventricular function. Long-term VAD support leads to a reduction in cardiothoracic ratio on chest roentgenogram, decrease in left ventricular end-diastolic dimension, reduction in pulmonary capillary wedge pressure and pulmonary vascular resistance, and increase in left ventricular ejection fraction and cardiac index. 7 Are changes in left ventricular geometry indicative of a recovery of left ventricular function or do they simply reflect a reduction in preload conditions? The reduction in cell length, width, and myocyte volume noted in conjunction with the decrease in left ventricular dilation and mass observed by echocardiography led Zafeiridis 8 to conclude that “alterations in geometry that occur in a parallel fashion at both organ and cellular levels may contribute to reduced wall stress and improved mechanical performance after left VAD (LVAD) support.” Alterations in cellular and humoral immunity, a reduction in pro-inflammatory cytokines, and improvement in myocardial ultrastructure have been observed in patients who have been weaned from mechanical blood pumps. Torre-Amione 9 described a significant reduction in intracardiac tumor necrosis factor-α, a cytokine capable of producing myocardial hypertrophy and fibrosis, after VAD support. Goldstein 10 noted that serum interleukin 6 and interleukin 8 levels rise with cardiogenic shock, but normalize with hemodynamic recovery, suggesting that “these cytokines may be markers of tissue damage....” Altemose 11 described a close relationship between left ventricular mass and atrial natriuretic peptide immunopositivity, both of which declined with 6 weeks of LVAD support. Failing myocardium exhibits a depressed response to β-adrenergic stimulation. Dipla 12 demonstrated that myocyte contractility and β-adrenergic responsiveness improved with LVAD support. Although the myocardium responds favorably to chronic unloading, the time course of ventricular recovery, if it is to occur, is unknown. Weaning protocols and reliable endpoints have yet to be defined. Loebe 3 carefully outlines a multi-week weaning process with the decision to explant the VAD based on a decline in anti-β1-adrenoreceptor-autoantibody titer and transthoracic echocardiographic-derived cardiac dimensions. Mancini, 1 and subsequently Slaughter, 13 combined echocardiographic information with oxygen consumption data to help make a decision as to when to initiate the weaning process. Most authors would agree that after a ventricle has “recovered” following an extended period of VAD support, the weaning process is best accomplished over a period of days to weeks. This allows the clinician to observe cardiac performance with increasing workload. A protracted weaning process avoids the sudden increase in ventricular wall stress and chamber dilation associated with acute “pump off” tests. Third, and perhaps most importantly, will the recovery of myocardial function after long-term VAD support be sustained? Initial anecdotal reports of device explantation have been followed by small series numbering tens of patients who have survived for up to 2 or 3 years after device removal, with no recurrence of left ventricular dilation. Other patients have succumbed to sudden death or redilated and have gone on to require VAD replacement or transplantation. Houel 14 reported two patients who received 12 and 50 days of biventricular support, respectively, for acute lymphocytic myocarditis. Ventricular function improved in both patients, permitting blood pump removal. At the time of his report, one patient was doing well 2 months after VAD explantation. The second patient required urgent cardiac transplantation for acute, recurrent biventricular failure. The authors conclude that “long-lasting recovery after mechanical support for acute myocarditis remains unpredictable....”14 Helman 15 coins the term “recurrent remodeling” to describe the recurrence of heart failure after device explantation in two patients in whom myocardial recovery was thought to have occurred after a period of LVAD support. On the other hand, Reedy 16 recently described 22 patients who received VAD support for an average of nearly 2 months before device removal. Twenty-one patients were discharged from the hospital. The 2-year actuarial survival in this patient group was 80%. Hetzer’s 17 series includes 23 patients with dilated cardiomyopathy who received left or biventricular support for a mean of 6 months. All patients seemingly recovered left ventricular function and underwent device explantation. Three patients subsequently died of noncardiac causes, seven patients developed recurrent heart failure and were referred for transplantation, and 13 patients experienced stable cardiac recovery at a mean follow-up of 23 months. The benchmark for surgical management of the end-stage heart failure patient has been defined. If the “bridge to recovery” application is to become a standard of care for selected patients with end-stage heart failure, temporary, long-term VAD implantation must it lead to the 85% 1-year and 75% 3-year survival seen after cardiac transplantation. I believe this is a very exciting time for the use of mechanical blood pumps in patients with end-stage heart failure. The successful use of the VAD as a bridge to recovery will have a profound impact upon our ability to manage this ever increasing patient population. Long-term temporary device implantation will permit pharmacologic manipulation of the resting, failing heart, reduce the burden on an already limited donor organ population, and expand blood pump use while eliminating many of the problems associated with permanent device insertion. I applaud the efforts of the basic scientists and clinicians that have contributed to the burgeoning literature concerning the recovery of myocardial function with long-term VAD support. I look forward to their future research efforts.