Skip main navigation

Prospective Multicenter Study of Myocardial Recovery Using Left Ventricular Assist Devices (RESTAGE-HF [Remission from Stage D Heart Failure])

Medium-Term and Primary End Point Results
Originally published 2020;142:2016–2028



Left ventricular assist device (LVAD) unloading and hemodynamic support in patients with advanced chronic heart failure can result in significant improvement in cardiac function allowing LVAD removal; however, the rate of this is generally considered to be low. This prospective multicenter nonrandomized study (RESTAGE-HF [Remission from Stage D Heart Failure]) investigated whether a protocol of optimized LVAD mechanical unloading, combined with standardized specific pharmacological therapy to induce reverse remodeling and regular testing of underlying myocardial function, could produce a higher incidence of LVAD explantation.


Forty patients with chronic advanced heart failure from nonischemic cardiomyopathy receiving the Heartmate II LVAD were enrolled from 6 centers. LVAD speed was optimized with an aggressive pharmacological regimen, and regular echocardiograms were performed at reduced LVAD speed (6000 rpm, no net flow) to test underlying myocardial function. The primary end point was the proportion of patients with sufficient improvement of myocardial function to reach criteria for explantation within 18 months with sustained remission from heart failure (freedom from transplant/ventricular assist device/death) at 12 months.


Before LVAD, age was 35.1±10.8 years, 67.5% were men, heart failure mean duration was 20.8±20.6 months, 95% required inotropic and 20% temporary mechanical support, left ventricular ejection fraction was 14.5±5.3%, end-diastolic diameter was 7.33±0.89 cm, end-systolic diameter was 6.74±0.88 cm, pulmonary artery saturations were 46.7±9.2%, and pulmonary capillary wedge pressure was 26.2±7.6 mm Hg. Four enrolled patients did not undergo the protocol because of medical complications unrelated to the study procedures. Overall, 40% of all enrolled (16/40) patients achieved the primary end point, P<0.0001, with 50% (18/36) of patients receiving the protocol being explanted within 18 months (pre-explant left ventricular ejection fraction, 57±8%; end-diastolic diameter, 4.81±0.58 cm; end-systolic diameter, 3.53±0.51 cm; pulmonary capillary wedge pressure, 8.1±3.1 mm Hg; pulmonary artery saturations 63.6±6.8% at 6000 rpm). Overall, 19 patients were explanted (19/36, 52.3% of those receiving the protocol). The 15 ongoing explanted patients are now 2.26±0.97 years after explant. After explantation survival free from LVAD or transplantation was 90% at 1-year and 77% at 2 and 3 years.


In this multicenter prospective study, this strategy of LVAD support combined with a standardized pharmacological and cardiac function monitoring protocol resulted in a high rate of LVAD explantation and was feasible and reproducible with explants occurring in all 6 participating sites.


URL:; Unique identifier: NCT01774656.

Clinical Perspective

What Is New?

  • RESTAGE-HF (Remission from Stage D Heart Failure) demonstrates that optimized left ventricular assist device (LVAD) mechanical hemodynamic unloading, combined with a standardized specific aggressive pharmacological regimen designed to induce reverse remodeling and regular testing of underlying myocardial function, enhances the incidence of LVAD explantation in a prospective multicenter study in patients with chronic advanced heart failure.

  • Forty percent of all enrolled (16/40) patients achieved the primary end point (alive free from mechanical support/heart transplantation 1 year after LVAD explant), P<0.0001, and 52.3% (19/36) receiving the protocol were explanted overall.

  • The RESTAGE-HF protocol was reproducible, with explants occurring in all 6 participating sites, a key component for broader application of this strategy.

What Are the Clinical Implications?

  • This strategy, which enhanced the rate of recovery, was feasible and reproducible in all participating centers and could be widely applied across LVAD centers, likely resulting in a higher rate of LVAD explantation from myocardial recovery, as currently few centers systematically promote recovery like this.

  • This supports a structured strategy of widely promoting and testing systematically for recovery after LVAD implantation.

  • The next steps likely will focus on using the LVAD as a platform to induce myocardial recovery using this protocol as a “Stage 1” base that centers could more broadly adopt and subsequently add further adjuvant regimens to.

Left ventricular assist devices (LVADs) are now being increasingly used to treat patients with advanced heart failure (HF). The significant hemodynamic unloading provided by LVAD mechanical support can result in structural reverse remodeling and improvement in cardiac function.1–4 This can be significant enough, even in patients with chronic HF, to allow pump removal5–9 with sustained improved myocardial function.10,11 However, the rate at which this occurs is reported as only 1% to 2% in the INTERMACS registry (Interagency Registry for Mechanically Assisted Circulatory Support).12 This low rate occurs because LVADs are usually implanted as a bridge to transplant (BTT) or destination therapy (DT) in patients considered to have failed medical therapy. Few centers prospectively test or look for evidence of recovery, and often HF medications are not continued or optimized after LVAD implantation.

However, significantly enhanced rates of recovery have been shown by using a strategy that combines optimized prolonged LVAD hemodynamic unloading with aggressive pharmacological therapy, specifically designed to enhance reverse remodeling, along with regular testing of underlying myocardial function.13–15 An increasing number of institutions have been adopting a more aggressive approach using some or all of these components to facilitate/test for recovery,6,16–19 and these have seen higher rates of recovery, suggesting the potential for recovery is higher and currently underestimated.

Here we report the primary end point results from RESTAGE-HF (Remission from Stage D Heart Failure), a multicenter prospective study using a uniform protocol of LVAD pump speed optimization combined with an aggressive drug regimen and regular testing of underlying function, to determine if an enhanced rate of myocardial recovery can be obtained.


The authors will make the methods used in the analysis and materials used to conduct the research available to any researcher for the purposes of reproducing the results or replicating the procedure. The authors declare that all supporting data are available within the article and the Data Supplement.

Study Design

RESTAGE-HF is a prospective nonrandomized multicenter study (URL:; Unique identifier: NCT01774656) of 40subjects enrolled from 6 centers with demonstrated previous experience in recovering and explanting patients with an LVAD: the University of Louisville (also the Data Coordinating Center), the University of Utah, the University of Pennsylvania, the University of Nebraska, Montefiore Medical Center, and the Cleveland Clinic (for enrollment by center, see the Data Supplement). Echocardiography and tissue core laboratories were at Montefiore Medical Center and the University of Pennsylvania, respectively. Institutional review board approval was obtained, the study was approved by an institutional review committee at each site, and all patients provided written informed consent. Coordinators at each site collected all study data, which was then forwarded to the data coordinating center.

Forty patients with chronic advanced HF from nonischemic cardiomyopathy receiving the Heartmate II continuous axial flow LVAD were enrolled. Key inclusion criteria (see the Data Supplement for full details) were age 18 to 59 years, LVAD indication as either BTT or DT, left ventricular (LV) ejection fraction (LVEF) <25%, nonischemic cardiomyopathy with cardiomegaly, and duration of HF of 5 years or less (HF duration was defined as the earlier of HF symptoms or imaging with LVEF recorded as <40%). Key exclusion criteria were histological evidence of active acute myocarditis, LV end-diastolic diameter (LVEDD) within normal range, hypertrophic cardiomyopathy or sarcoidosis, mechanical aortic or mitral valve(s), aortic surgical valve closure, and significant motor deficit from previous cerebrovascular accident limiting ability to perform exercise testing. Patients agreed not to undergo heart transplantation for a minimum of 4.5 months after LVAD implantation (beyond that, the trend of improvement was discussed with the patient by the investigator and a treatment strategy decision made locally).

Study End Points

The primary end point was the proportion of subjects who met explant criteria after a standardized LVAD plus aggressive pharmacological reverse-remodeling treatment and testing protocol within 18 months with subsequent freedom from death/mechanical circulatory support/heart transplantation at 1 year after LVAD removal.

The secondary end points were the proportion of subjects meeting explant criteria and subsequently explanted, the durability of remission from HF at 12 months and up to 3 years, the time course of reverse remodeling, changes in ejection fraction measured at 6000 rpm, and predictors of recovery.

Study Protocol

Optimization of Pump Speed

After implant, before discharge, or at the first outpatient visit after discharge, the pump speed was optimized by echocardiography to maximize LV unloading. The pump speed was increased from baseline in increments of 200 rpm to optimize LV dimension reduction, and this was continued at subsequent visits until the LVEDD was <6 cm and mitral regurgitation <2 (if possible).

Pharmacological Management

Aggressive pharmacological management was initiated consisting of drugs intended to enhance reverse remodeling. Five drugs were used with the first initiated immediately after the weaning of inotropic support after LVAD implantation once there was adequate end-organ recovery. The drugs were titrated to a mean arterial pressure >60 mm Hg, as long as the patient was asymptomatic with adequate renal function and electrolytes within the normal range, to the following maximally tolerated target doses: lisinopril 20 mg PO BID, carvedilol 50 mg PO BID, spironolactone 25 mg PO daily, digoxin 125 μg PO daily, and losartan 150 mg PO daily. The patients underwent close monitoring of their renal function and electrolytes. This was based on phase I of the Harefield protocol, which proved to be both effective and safe in previous studies13–15 (except in the RESTAGE-HF protocol, the target dose of losartan was increased to 150 mg PO daily based on the results of the HEAAL trial [Heart failure End point evaluation of Angiotensin II Antagonist Losartan]20). For the detailed drug optimization protocol, see the Data Supplement.

Monitoring and Subject Follow-Up

Before LVAD implantation, subjects underwent a thorough clinical assessment including an echocardiogram and right heart catheterization. Follow-up visits were performed at 6 weeks and 3, 4, 5, 6, and 12 to 18 months after implantation and consisted of a physical examination, blood tests, and standard of care echocardiograms. At week 6 and at month 4, 6, and 9 visits, a full low-speed echocardiogram was performed in conjunction with a 6-minute walk test to measure “inotropic reserve.” At the month 12 to 18 visit, the low-speed echocardiogram was optional depending on the improvement in cardiac function seen. For the low-speed echocardiograms (based on previous studies21 showing there to be no net flow with the Heartmate II at 6000 rpm), a full echocardiogram was performed at the patient’s baseline speed, and then, with an International Normalized Ratio >2 (or after 10 000 IU heparin if International Normalized Ratio subtherapeutic), the pump speed was reduced to 6000 rpm in increments of 1000 rpm over 1 to 2 minutes. A limited echocardiogram was then performed at 5 minutes (LVEDD, left ventricular end-systolic diameter [LVESD], LVEF, and mitral regurgitation measurements) and then a full echo at 15 minutes of 6000 rpm. The Simpsons method performed on the apical 4-chamber view was selected for LVEF determination. If the subject tolerated the speed of 6000 rpm for 15 minutes, a 6-minute walk was performed at 6000 rpm followed by a repeat echocardiogram. Before explant, a cardiopulmonary exercise test was performed at 6000 rpm, and right ± left heart catheterization was performed at baseline speed and at 6000 rpm for 15 minutes.

For LVAD explantation by protocol, patients had to meet the minimum explant criteria (detailed in the Data Supplement), measured at 6000 rpm (zero net flow21) for 15 minutes:

  1. LVEDD <60 mm, LVESD <50 mm, LVEF >45%

  2. Left ventricular end-diastolic pressure or pulmonary capillary wedge pressure ≤15 mm Hg

  3. Resting cardiac index (CI) >2.4 L/min/m2

  4. ± Maximal oxygen consumption with exercise >16 mL/kg/min (per prespecified RESTAGE-HF protocol, this was an optional criterion, ie, the cardiopulmonary test data were always obtained, but maximal oxygen consumption was the 1 criterion not required to be met to proceed with device explantation)

It was predecided that should a center elect to explant with different criteria, the data would be reported but would not be part of the primary end point analysis but as a secondary end point.

Explantation and Postexplantation Pharmacological Therapy and Follow-Up

The method of surgical LVAD explantation versus decommission was left to the standard practice at each center. The same standardized pharmacological protocol was restarted after explantation and titrated up to the maximally tolerated doses as described above (detailed in the Data Supplement). A clinical assessment, physical examination, blood test, and echocardiogram were performed at 1, 2, 3, and 4 weeks, and 2, 3, 4, 6, 8, 10, 12, 24, and 36 months after explantation. Subjects were followed for 18 months after implant and up to 3 years after explant or until expired or transplanted. A data lock was set when the last explanted patient reached their primary end point (ie, 1 year after the last explant performed within the 18-month primary end point window) for the purposes of this article.

Statistical Analysis

The primary end point was predefined and statistically powered to be successfully met if this was greater than 10%, a liberal estimate of the proportion of subjects who would experience this outcome with usual clinical care. Hence the study was powered for a null hypothesis of 10%, and the alternate hypothesis was that it was significantly greater than this. The z-score was calculated based on the primary end point achieved.

Outcomes after LVAD were calculated with a competing outcomes analysis, and the Kaplan-Meier method was used for postexplantation survival free from re-LVAD or transplantation. A post hoc logistic regression model was generated (using likely and known risk factors) for the multivariate logistic regression analysis of predictors, and a univariate analysis of predictors was also performed. The remaining data were analyzed as basic descriptive statistics. Data is given as mean+SD.


Between December 6, 2012, and December 14, 2015, a total of 40 patients with chronic advanced HF caused by nonischemic cardiomyopathy receiving the Heartmate II LVAD were enrolled at 6 centers (for enrollment by center, see Table I in the Data Supplement). The baseline characteristics before LVAD implantation for all 40 patients are shown in Table 1. The mean age of the subjects was 35.1±10.8 years, and 67.5% were men. The mean duration of HF was 20.8±20.6 months. All patients were considered to have failed medical therapy, and 38 (95%) were on inotropic support, with 38 (95%) in New York Heart Association (NYHA) Class IV and 2 (5%) in NYHA Class IIIB, 5 (12.5%) required an intra-aortic balloon pump, 1 (2.5%) required an Impella, and 1 (2.5%) required a Centrimag pump before LVAD implantation. Twenty-two subjects (55%) were implanted as DT and 18 (45%) as a BTT. Before LVAD implantation, LVEF was 14.5±5.3%, LVEDD was 7.33±0.89 cm, and LVESD was 6.74±0.88 cm. Fick CI was 1.87±0.43 L/min/m2 (measured on inotropes in 38 [95%] and intra-aortic balloon pump or temporary mechanical circulatory support in 7 [17.5%]), pulmonary artery (PA) saturations were 46.7±9.2%, and pulmonary capillary wedge pressure was 26.2±7.6 mm Hg (Table 1).

Table 1. Demographic Information, Severity of Heart Failure, and Hemodynamics of 40 Enrolled Study Patients Before Heartmate II LVAD Implantation

DemographicAll enrolled (n=40)
Age, y35.1±10.8
Men/women, n (% of males)27/13 (67.5%)
White/Hispanic/Black/other, n (%)25 (62.5%)/5 (12.5%)/8 (20%)/2 (5%)
Implanted as BTT/DT, n (%)18 (45%)/22 (55%)
History of ventricular arrhythmias, n (%)6 (15%)
Presence of CRT, n (%)4 (10%)
ICD, n (%)24 (60%)
History of hypertension, n (%)12 (30%)
Duration of heart failure, mo20.8±20.6
Family history/genetic defects, n (%)5 (12.5%)
Likely postpartum component, n (%)6 (15%)
After chemotherapy, n (%)2 (5%)
NYHA Class IV, n (%)38 (95%)
NYHA Class IIIB, n (%)2 (5%)
INTERMACS Class 1, n (%)5 (12.5%)
INTERMACS Class 2, n (%)8 (20%)
INTERMACS Class 3, n (%)25 (62.5%)
INTERMACS Class 4, n (%)2 (5%)
Inotropic support, n (%)38 (95%)
IABP/Impella/Centrimag before LVAD, n (%)5 (12.5%)/ 1 (2.5%)/ 1 (2.5%)
LVEF, %14.5±5.3
LVEDD, cm7.33±0.89
LVESD, cm6.74±0.88
LVPWTd, cm0.94±0.27
PCWP, mm Hg26.2±7.6*
Fick Cardiac Output, L/min3.85±0.87*
Fick Cardiac Index, L/min/m21.87±0.43*
PA saturations, %46.7±9.2*

Data provided is mean±SD. BTT indicates bridge to transplant; CRT, cardiac resynchronization therapy; DT, destination therapy; IABP, intra-aortic balloon pump; ICD, implantable cardiac defibrillator; INTERMACS, Interagency Registry for Mechanically Assisted Circulatory Support; LVAD, left ventricular assist device; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LVESD, left ventricular end-systolic diameter; LVPWTd, left ventricle posterior wall thickness in diastole; NYHA, New York Heart Association; PA, pulmonary artery; and PCWP, pulmonary capillary wedge pressure.

* A total of 38 (95%) patients were on inotropic support, 5 (12.5%) on IABP, 1 (2.5%) on Centrimag and 1 (2.5%) on Impella

Of note, the patients with a HF duration <6 months had an end-diastolic diameter of 7.9±0.5 cm, end-systolic diameter of 7.3±0.6 cm, and ejection fraction 12.7±4.1%; all were in NYHA Class IV and all were on inotropes, 2 had an intra-aortic balloon pump, 1 an Impella, and 1 a Centrimag, with 1 requiring hemodialysis.

Patient Outcomes

Of the 40 enrolled patients, 4 did not undergo any of the protocol study procedures—not the pharmacological therapy, LVAD speed optimization, or cardiac function monitoring/study testing. Two of these patients died at 14 and 148 days having never left the intensive care unit after LVAD implantation, and a third died at 63 days of a cerebral hemorrhage secondary to methicillin-resistant Staphylococcus aureus septicemia. The fourth patient required pump ligation at 218 days while still in the intensive care unit after implantation after a pump disconnect/stoppage, and 36 patients completed the study protocol.

LVAD Explant for Myocardial Recovery

Overall, a total of 19 patients (19/36, 52.8% of those receiving the protocol) were explanted or had their LVAD decommissioned (Figures 1 and 2). Of these, 12 had idiopathic cardiomyopathy, 2 had familial cardiomyopathy, 3 had a peripartum component, and 2 had chemotherapy-induced cardiomyopathy.

Figure 1.

Figure 1. Flow chart showing the outcome of all enrolled patients. n=40. LVAD indicates left ventricular assist device.

Eighteen patients reached explant criteria within the 18-month time period after LVAD implantation required by the primary end point (ie, 50% [18/36] of those receiving the protocol), and 1 patient (who required pump exchange at 1.3 years) had his LVAD decommissioned at 1089 days after the original implant (Figures 2 and 3).

Figure 2.

Figure 2. Competing outcomes curves showing the competing outcomes of all enrolled patients (n = 40) over time. Explantation for recovery (n=19, 47.5%), ongoing left ventricular assist device (LVAD) support (n=9, 22.5%), transplantation (n=4, 10.0%), withdrawn (n=1, 2.5%), or death (n=7, 17.5%).

Twelve patients had the pump fully explanted (11 by sternotomy and 1 by combined sternotomy and anterolateral thoracotomy). Six patients were left with a residual inflow cannula (4 by left subcostal incision, 1 by lower sternotomy, and 1 by combined anterolateral thoracotomy and sternotomy), all without using cardiopulmonary bypass, and these patients were kept on coumadin and aspirin. One patient had the pump decommissioned percutaneously with an Amplatzer plug, and he was maintained on coumadin. One of the patients with a residual inflow cannula had ventricular tachycardia after 6 weeks and required resternotomy and removal of the inflow cannula and elbow (and mitral and aortic valve repair). Table 2 shows the explant details and the anticoagulation used after explantation.

Table 2. Surgical Explant Details and Postexplant Anticoagulation

Type of explantNumberCardiopumonary bypassApproachPostoperative anticoagulationComplications
Full12 (infection present in 3)CPB in 811 sternotomyNoneNone
No CPB in 41 combined sternotomy and anterolateral thoracotomy
Retained inflow cannula6None4 left subcostal incisionCoumadin and aspirin1 had VT requiring resternotomy and removal of the inflow cannula and elbow (and mitral and aortic valve repair)
1 lower sternotomy and 1 combined anterolateral thoracotomy and sternotomy
Decommissioned with an Amplatzer plug1NonePercutaneousCoumadinNone

CPB indicates cardiopulmonary bypass; and VT, ventricular tachycardia.

Of the other 17 patients who were not explanted (Figure 1), 9 remain on ongoing LVAD support at a mean of 3.97±0.54 (range, 3.1–4.6) years after LVAD implantation, 1 patient relocated and withdrew from the study at 9.5 months, 4 were transplanted 1.8±0.8 (range, 0.8–2.8) years after implantation (all alive 795±445 days after transplant), and 3 died at 2.0, 2.1, and 2.8 years after LVAD implant from consequences of noncompliance, cerebral hemorrhage, and endocarditis. Competing outcomes after LVAD are shown in Figure 2.

Echocardiograhic Response During LVAD Support

The echocardiographic response of the 36 patients over time is shown in Figure 4A and 4B. Except for the baseline echocardiographic measurements, all of the other measurements were taken after 15 minutes of the pump being turned down to 6000 rpm (ie, no net flow through the Heartmate II pump).

Figure 3.

Figure 3. Flow chart summarizing the primary end point results for all enrolled patients (n=40; ie, the patients meeting explant criteria within 18 months and remaining free from death/transplantation or mechanical support at 1 year after explant). LVAD indicates left ventricular assist device; Tx, transplantation; and VAD, ventricular assist device.

Echocardiograhy and Hemodynamics Pre-Explant in the Explanted Patients

The mean time to explantation of the 18 patients who reached explant criteria within 18 months (required by the primary end point) was 352±165 (range, 116–656) days. Before explantation for these patients, LVEF (at 6000 rpm for 15 minutes) was 57±8%, LVEDD was 4.81±0.58 cm, and LVESD was 3.53±0.51 cm. Hemodynamic assessment before explantation at 6000 rpm showed PA systolic pressure of 26.5±7.1 mm Hg (versus 24.6±6.8 mm Hg at baseline speed), PA diastolic pressure of 9.3±3.3 mm Hg (versus 8.7±3.5 mm Hg at baseline speed), pulmonary capillary wedge pressure of 8.1±3.1 mm Hg (versus 6.9±3.5 mm Hg), Fick CI was 2.43±0.3 L/min/m2 (versus 2.69±0.47 L/min/m2 at baseline speed), and PA saturations were 63.6±6.8% at 6000 rpm (versus 64.2±6.6% at baseline speed).

Another patient who reached echocardiographic (LVEDD 4.1, LVESD 3.1, and LVEF 49%) but not hemodynamic criteria on low speed testing (CI dropped from 1.66 to 1.5) was not initially explanted but later reached hemodynamic criteria (had improved to CI 2.5 on repeat low-speed testing), and had his LVAD decommissioned at 1089 days (3 years) after the original implant.

Outcomes After Explantation

Primary End Point

Of the 19 explanted patients, 18 met explant criteria within the 18-month period that was prespecified in the primary end point, and of these, 16 patients (40% of the original 40 enrolled) reached the primary end point of LVAD explant within 18 months with freedom from mechanical circulatory support/heart transplantation at 1 year after explant (Figure 3). Of the 2 patients not reaching the event-free 1-year mark, 1 patient committed suicide 15 days after explantation (and an echocardiogram performed 3 days before he died showed an LVEDD of 5.8 cm, LVESD of 4.2 cm, and LVEF of 57%), and the other patient developed recurrent heart failure and was transplanted at 227 days after explantation (he remains alive and well 3.5 years [1282 days] later). The study was powered for a null hypothesis of 10%, hence the null hypothesis was rejected with the primary end point statistically met (calculated z value 6.8, P<0.0001).

Long-Term Outcomes Beyond 1 Year After Explant

Two patients died after the 1-year postexplant primary end point. One developed recurrent heart failure and died 577 days (1.6 years) after explant. Another patient with bipolar disorder stopped his HF medications, did not contact the LVAD implanting center, and died of a pneumonia (Figure 5) at 639 days after explant.

Current Status

The remaining 15 explanted patients are currently alive and well at 2.1±0.9 years after LVAD explant. Fourteen of these were explanted within the 18-month period required by the primary end point and are now 2.4±0.9 (range, 1.0–4.1) years after explant with LVEF of 47.7±11.2%, LVEDD 5.55±0.47 cm, and LVESD 4.5±0.74 cm. The other patient who had his LVAD decommissioned after 3 years is now 242 days after explant with a LVEF of 45%, LVEDD 5.5 cm, and LVESD 4.3 cm. Postexplantation survival free from re-LVAD or transplantation is shown in Figure 6.

Figure 4.

Figure 4. Echocardiographic changes in underlying cardiac function over time after left ventricular assist device (LVAD) implantation in all evaluable patients. A, Left ventricular ejection fraction measured by Simpsons method with the LVAD speed turned down to 6000 rpm (no net flow21) for 15 minutes at serial time points after LVAD implantation in all evaluable patients (n=36). B, Left ventricular end diastolic diameter and left ventricular end systolic diameter measured by with the LVAD speed turned down to 6000 rpm (no net flow21) for 15 minutes at serial time points after LVAD implantation in all evaluable patients (n=36).

Figure 5.

Figure 5. Flow chart showing the current status of all 40 enrolled patients. LVAD indicates left ventricular assist device.

Figure 6.

Figure 6. Postexplant survival free of transplantation or left ventricular assist device (LVAD).

Predictors of Recovery

Multivariate logistic regression analysis showed that age; sex; duration of HF; presence of cardiac resynchronization therapy; ventricular arrythmias; hypertension; an underlying diagnosis of familial, postpartum, or chemotherapy-induced cardiomyopathy; indication BTT versus DT; pre-LVAD ejection fraction; creatinine; and PA diastolic pressure all did not predict whether patients recovered or not. A univariate analysis showed that only lower preoperative creatinine was associated with a higher chance of recovery (P<0.05).


We have demonstrated that pharmacological therapy combined with optimal LVAD unloading to promote myocardial recovery resulted in 40% of all enrolled (16/40) patients achieving the primary end point (alive free from mechanical support/heart transplantation at 1 year after LVAD explant), P<0.0001. Furthermore, 50% (18/36) of patients receiving the protocol reached the explant criteria within the predefined 18-month period, and 52.3% (19/36) of these were explanted overall. This result was reproducible with explants occurring in all 6 participating centers. These patients demonstrated remission from advanced chronic HF with an LVAD, and these results are important because myocardial recovery and remission from HF would be the preferred outcome, if the patient can retain their own heart over heart transplant or permanent LVAD support.

The patients in this study had chronic advanced HF as specified in our inclusion criteria (which required cardiomegaly and set any histological evidence of myocarditis on the core at the time of implantation as an exclusion). Echocardiography before LVAD implant confirmed chronic dilated disease in this cohort who had an LVEDD of 7.33±0.89 cm and LVESD of 6.74±0.88 cm. Furthermore, these patients were sick at the time of LVAD implantation (mean INTERMACS profile 2.6±0.8), all reached criteria for advanced therapies, and 95% were on inotropic support with 17.5% on temporary mechanical circulatory support, making recovery without proceeding to durable mechanical support unlikely. Of note, for the patients with a HF duration <6 months, their end-diastolic diameter was 7.9±0.5 cm, end-systolic diameter was 7.3±0.6 cm, and ejection fraction was 12.7±4.1%, suggesting they had all had chronic disease, all were NYHA Class IV, and all were on inotropes with 2 having an intra-aortic balloon pump, 1 an Impella, and 1 a Centrimag, and 1 requiring dialysis, suggesting they could not have survived without proceeding to LVAD support.

Historically, the percentage of patients who have had their LVAD successfully removed has been perceived to be low—reported to be approximately 4.5% to 24% in most institutional series5–9; a recent United Network for Organ Sharing analysis22 showed a rate of 5%, and publications from INTERMACS have shown the explant rate to be only 1% to 2%.12 However, few centers systematically, based on protocol, examine for evidence of recovery. Patients in most centers are implanted as either a BTT or as DT and progress along that path without their underlying myocardial function being tested.19 LVAD-induced recovery is therefore underevaluated and underpromoted. Centers that have implemented protocols for systematic evaluation of myocardial function after LVAD implant have seen higher rates of recovery.9,16–18,23 Transplantation has a defined benefit and a well-known natural history, and trepidation about recurrence of HF after LVAD removal has been understandably prevalent in those that care for these patients, but unfortunately the number of useable donor hearts has not been increasing over recent years, necessitating an alternative approach for these patients.

The strategy of optimizing recovery of myocardial function after LVAD implant was developed at Harefield Hospital in England. It consisted of the LVAD being set at a speed for optimal unloading, combined with high doses of drugs known to enhance reverse remodeling (Phase 1), along with regular testing of underlying cardiac function (with the pump off, or essentially off), followed by the use of clenbuterol (Phase 2) to improve the durability of recovery after explantation.13 This approach was associated with a 70% explant rate in a prospective study with the Heartmate I pulsatile LVAD, with more than 3 years of follow-up,14 and 60% in a subsequent study using the Heartmate II continuous flow LVAD.15

Much of the rationale of this protocol is to combine mechanical unloading with drugs known to enhance reverse remodeling.13–15,24 Most stop reverse-remodeling HF drugs after LVAD implant, believing them to have “failed” in that patient, only using them for hypertension control. These are a key part of the RESTAGE-HF protocol, where the reverse-remodeling HF drugs are aggressively initiated when there is adequate end-organ recovery and titrated to high doses. A common thought process is that the LVAD will provide the output necessary for a patient with lifetime use, and the focus is on the device rather than the heart. Whereas patients often do not tolerate large doses of angiotensin-converting enzyme inhibitors (ACEi), β-blockers, mineralocorticoid receptor antagonists, and angiotensin II receptor blockers while in severe HF because of renal failure or hypotension, once they achieve good cardiac output and adequate blood pressure and renal function on LVAD support, they tolerate the HF drugs well, and these can often be used at high, not previously tolerated doses. This is an important part of the protocol. Angiotensin receptor–neprilysin inhibitors had not been approved when this trial was designed, which may yield further benefit.

These drugs and optimized neurohormonal blockade (NHB), the cornerstone for HF therapy, are usually grossly underused25 in patients with LVAD. Two recent single-center studies have shown ACEi to improve B-type natriuretic peptide, NYHA class, 6-minute walk distance, and reverse remodeling parameters26 and improve mortality27 in patients with LVAD. Furthermore, mechanical unloading increases myocardial angiotensin II, collagen cross-linking, and myocardial stiffness,28 and addition of ACEi in these patients decreases cross-linked collagen, LV mass, and myocardial stiffness.29 There may be other benefits as ACEi and digoxin have also been shown to reduce the rate of gastrointestinal bleeding in patients with LVAD.30 A recent large multicenter analysis from INTERMACS31 has identified improved long-term survival in patients with LVAD who are receiving NHB. In that study, the use of any NHB was associated with significantly improved survival 4 years after implant and also with a higher Kansas City Cardiomyopathy Questionnaire score and a longer 6-minute walk test at 2 years. Furthermore, in this INTERMACS study, patients receiving triple therapy with an ACEi or angiotensin II receptor blockers, β-blockers, and mineralocorticoid receptor antagonists had the lowest hazard of death and the lowest N-terminal pro-B-type natriuretic peptide and creatinine. This underscores that the strategy to maximally induce reverse remodeling with established HF therapeutics is not only safe, but also beneficial for success with long-term mechanical circulatory support. A wider and more aggressive attempt to promote and look for recovery in a larger population of LVADs is likely to result in a much higher incidence and identify a broader group of patients that can recover.

Pump speed optimization was also a priority in the RESTAGE-HF trial. Most centers do not optimize pump speed for unloading, and many leave the aortic valve opening, with partial unloading sometimes wanting to retain some pulsatility. Testing of underlying myocardial function is also crucial and necessitates an accurate and safe way of testing function while on mechanical support. The 6000 rpm testing speed has now been reliably and safely used in several centers, which is major progress toward more widespread evaluation of recovery.

Our study protocol required that patients agree not to undergo heart transplantation for a minimum of 4.5 months after LVAD implantation. This was a point of discussion and a significant concern for the investigators at the time of trial design, as it was expected that most patients would be BTT, and we did not want them to miss their chance of a heart transplant. However, in reality and with the limited donor supply, this was not an issue, as 55% of the patients were implanted as DT, and over the course of the study, transplant wait times became prolonged.

A multivariate analysis showed that none of the clinical factors examined predicted recovery. This included age; duration of HF; presence of cardiac resynchronization therapy; or an underlying diagnosis of familial, postpartum, or chemotherapy-induced cardiomyopathy. Of note, 2 patients with a family history or known genetic defect and 2 patients with chemotherapy-induced cardiomyopathy recovered. A univariate analysis did show that lower preoperative creatinine was associated with a higher subsequent chance of recovery (P<0.05), which might be related to an ability to tolerate higher doses of the neurohormonal antagonists.

A cohort of patients now exist who have had their LVADs removed with sustained recovery for many years with a normal quality of life.32 The long-term durability of recovery or remission from HF still has to be determined for the cohort explanted in this prospective study. We locked the data at 1 year after the last patient who satisfied the primary end point was explanted for the purposes of this publication, but will subsequently follow all patients to 3 years after explant and beyond. Recently published data33 show excellent functional status longer-term after explant from the Harefield series, with many explanted patients having cardiac and physical functional capacities within the normal range of healthy controls, suggesting that the functional recovery described is sustained. A recently published clinical trial randomizing chronic patients with HF to withdrawal of pharmacological therapy (TRED-HF) identified relapse with myocardial dysfunction/HF in 44% of patients after withdrawal of HF therapy, compared with none of the patients with ongoing therapy, underscoring the importance of continuing NHB in preventing relapse of recurrent ventricular dysfunction.34 The explanted patients in RESTAGE-HF were restarted on the aggressive NHB as tolerated, and this was maintained.

The rate of recovery using this strategy is significantly higher than previously reported. In the previous studies, clenbuterol was mostly aimed to improve the durability of recovery and was initiated after the heart was mostly reverse-remodeled and there was already significant improvement in myocardial function.13–15 Clenbuterol was not used in this study to specifically assess the effect and reproducibility of the Phase 1 part of the protocol (which centers could adopt on a more widespread basis). Intermittently lowering the pump speed to allow sufficient contractility to open the aortic valve and increase myocardial work might be an alternative way to retrain the LV in the future, simulating the clenbuterol concept. After an initial phase of reverse remodeling and shrinking the ventricle with maximal unloading, patients may benefit from periods of significantly reduced LVAD support, to “work” the ventricle before explant, simulating this concept. This might be a focus of future studies to further improve the durability of the recovery obtained.

The RESTAGE protocol appears to enhance rates of recovery and was reproducible with explants successfully occurring in all 6 centers, a key component for broader application. The next steps are likely to focus on using this protocol as a “Stage 1” base that centers could more broadly adopt and subsequently add further adjuvant regimens to. The hemodynamic stability provided by the newer LVADs allows creative and innovative approaches, and a variety of adjunctive agents could be used in combination with LVAD therapy. Using the LVAD as a platform to induce myocardial recovery is likely to lead to a significant increase in the rate of recovery.

Patients explanted because of myocardial recovery avoid the need for immunosuppression and its associated complications and spare the donor heart for another needy individual.10 Even if explanted patients should decompensate and require transplantation at a later stage, this approach is likely to extend their overall lifespan considerably.10 The newer continuous flow pumps have excellent durability, more patients are having these devices implanted as an alternative to transplantation, and many transplant candidates are waiting for prolonged times, allowing more time for promotion and testing for myocardial recovery. Our data suggest explantation should be considered in more patients. Furthermore, if LVADs start to be implanted at an earlier stage of HF, it is likely that more patients could recover.

Study Limitations

Although this study was prospective, it has a relatively small number of patients. It also had no control group (although to monitor hemodynamic changes in such complex patients, they often serve well as their own controls), and so although our observations are important, as they show that a protocol of optimizing pump speed for unloading combined with aggressive therapy to promote recovery and testing of underlying function can result in a higher rate of recovery than previously seen, which was reproducible in several centers, in the absence of a randomized comparator group, this is only hypothesis-generating and cannot be considered definitive. Our study focused on patients under the age of 60 years with nonischemic cardiomyopathy. This does not necessarily mean that older patients or those with other etiologies would not have recovered. It just means that we did not study them, so we cannot comment on whether this protocol would have had a similar effect in them or what percentage of them might recover.


In a multicenter prospective study, this protocol enhanced the rate of recovery compared with that otherwise reported with explants successfully occurring in all 6 centers, a key component for its broader application. This suggests that the explant rate after LVAD could be much higher if this strategy were more widely used and supports promoting and testing systematically for recovery after LVAD implantation.


The authors are indebted to the incredible research coordinators who were invaluable. In particular the authors acknowledge Jennifer Strege, Tamara Bernard, Judith Marble, Joanna Oviedo, Terri Blanton, and Barbara Gus.

Supplemental Materials

Data Supplement Table I

Supplemental Methods

IA. Inclusion Criteria

IB. Exclusion Criteria

II. Detailed Drug Protocol to Promote Recovery Post LVAD

III. Evaluation for LVAD Explantation

IV. Drug Protocol Post-Explantation


Sources of Funding, see page 2026

Continuing medical education (CME) credit is available for this article. Go to to take the quiz.

The Data Supplement, podcast, and transcript are available with this article at

Emma Birks, FRCP, PhD, MBBS, Gill Heart and Vascular Institute, University of Kentucky, 800 Rose Street, Lexington, KY 40536; Email
Stavros Drakos, MD, PhD, FACC, Division of Cardiology, University of Utah, 50 North Medical Drive, Salt Lake City, UT 84112. Email


  • 1. Frazier OH, Radovancevic B, Abou-Awdi NL, Wilansky S, Jin B, Capek P, Mac Allister HA, Seifein H, Macris MP. Ventricular remodelling after prolonged ventricular unloading “heart rest” experience with the Heartmate left ventricular assist device.J Heart Lung Transplant. 1994; 13:77Google Scholar
  • 2. Levin HR, Oz MC, Chen JM, Packer M, Rose EA, Burkhoff D.Reversal of chronic ventricular dilation in patients with end-stage cardiomyopathy by prolonged mechanical unloading.Circulation. 1995; 91:2717–2720. doi: 10.1161/01.cir.91.11.2717LinkGoogle Scholar
  • 3. Drakos SG, Wever-Pinzon O, Selzman CH, Gilbert EM, Alharethi R, Reid BB, Saidi A, Diakos NA, Stoker S, Davis ESet al; UCAR (Utah Cardiac Recovery Program) Investigators. Magnitude and time course of changes induced by continuous-flow left ventricular assist device unloading in chronic heart failure: insights into cardiac recovery.J Am Coll Cardiol. 2013; 61:1985–1994. doi: 10.1016/j.jacc.2013.01.072CrossrefMedlineGoogle Scholar
  • 4. Drakos SG, Pagani FD, Lundberg MS, Baldwin JT.Advancing the Science of Myocardial Recovery With Mechanical Circulatory Support: A Working Group of the National, Heart, Lung, and Blood Institute.JACC Basic Transl Sci. 2017; 2:335–340. doi: 10.1016/j.jacbts.2016.12.003CrossrefMedlineGoogle Scholar
  • 5. Frazier OH, Myers TJ.Left ventricular assist system as a bridge to myocardial recovery.Ann Thorac Surg. 1999; 68:734–741. doi: 10.1016/s0003-4975(99)00801-2CrossrefMedlineGoogle Scholar
  • 6. Dandel M, Weng Y, Siniawski H, Potapov E, Lehmkuhl HB, Hetzer R.Long-term results in patients with idiopathic dilated cardiomyopathy after weaning from left ventricular assist devices.Circulation. 2005; 1129 supplI37–I45. doi: 10.1161/CIRCULATIONAHA.104.525352LinkGoogle Scholar
  • 7. Mancini DM, Beniaminovitz A, Levin H, Catanese K, Flannery M, DiTullio M, Savin S, Cordisco ME, Rose E, Oz M.Low incidence of myocardial recovery after left ventricular assist device implantation in patients with chronic heart failure.Circulation. 1998; 98:2383–2389. doi: 10.1161/01.cir.98.22.2383LinkGoogle Scholar
  • 8. Farrar DJ, Holman WR, McBride LR, Kormos RL, Icenogle TB, Hendry PJ, Moore CH, Loisance DY, El-Banayosy A, Frazier H.Long-term follow-up of Thoratec ventricular assist device bridge-to-recovery patients successfully removed from support after recovery of ventricular function.J Heart Lung Transplant. 2002; 21:516–521. doi: 10.1016/s1053-2498(01)00408-9CrossrefMedlineGoogle Scholar
  • 9. Simon MA, Kormos RL, Murali S, Nair P, Heffernan M, Gorcsan J, Winowich S, McNamara DM.Myocardial recovery using ventricular assist devices: prevalence, clinical characteristics, and outcomes.Circulation. 2005; 1129 supplI32–I36. doi: 10.1161/CIRCULATIONAHA.104.524124LinkGoogle Scholar
  • 10. Birks EJ, George RS, Firouzi A, Wright G, Bahrami T, Yacoub MH, Khaghani A.Long-term outcomes of patients bridged to recovery versus patients bridged to transplantation.J Thorac Cardiovasc Surg. 2012; 144:190–196. doi: 10.1016/j.jtcvs.2012.03.021CrossrefMedlineGoogle Scholar
  • 11. Dandel M, Weng Y, Siniawski H, Potapov E, Drews T, Lehmkuhl HB, Knosalla C, Hetzer R.Prediction of cardiac stability after weaning from left ventricular assist devices in patients with idiopathic dilated cardiomyopathy.Circulation. 2008; 11814 supplS94–S105. doi: 10.1161/CIRCULATIONAHA.107.755983LinkGoogle Scholar
  • 12. Kirklin JK, Pagani FD, Kormos RL, Stevenson LW, Blume ED, Myers SL, Miller MA, Baldwin JT, Young JB, Naftel DC.Eighth annual INTERMACS report: special focus on framing the impact of adverse events.J Heart Lung Transplant. 2017; 36:1080–1086. doi: 10.1016/j.healun.2017.07.005CrossrefMedlineGoogle Scholar
  • 13. Yacoub MH.A novel strategy to maximize the efficacy of left ventricular assist devices as a bridge to recovery.Eur Heart J. 2001; 22:534–540. doi: 10.1053/euhj.2001.2613CrossrefMedlineGoogle Scholar
  • 14. Birks EJ, Tansley PD, Hardy J, George RS, Bowles CT, Burke M, Banner NR, Khaghani A, Yacoub MH.Left ventricular assist device and drug therapy for the reversal of heart failure.N Engl J Med. 2006; 355:1873–1884. doi: 10.1056/NEJMoa053063CrossrefMedlineGoogle Scholar
  • 15. Birks EJ, George RS, Hedger M, Bahrami T, Wilton P, Bowles CT, Webb C, Bougard R, Amrani M, Yacoub MHet alReversal of severe heart failure with a continuous-flow left ventricular assist device and pharmacological therapy: a prospective study.Circulation. 2011; 123:381–390. doi: 10.1161/CIRCULATIONAHA.109.933960LinkGoogle Scholar
  • 16. Maybaum S, Mancini D, Xydas S, Starling RC, Aaronson K, Pagani FD, Miller LW, Margulies K, McRee S, Frazier OHet al; LVAD Working Group. Cardiac improvement during mechanical circulatory support: a prospective multicenter study of the LVAD Working Group.Circulation. 2007; 115:2497–2505. doi: 10.1161/CIRCULATIONAHA.106.633180LinkGoogle Scholar
  • 17. Patel SR, Saeed O, Murthy S, Bhatia V, Shin JJ, Wang D, Negassa A, Pullman J, Goldstein DJ, Maybaum S.Combining neurohormonal blockade with continuous-flow left ventricular assist device support for myocardial recovery: a single-arm prospective study.J Heart Lung Transplant. 2013; 32:305–312. doi: 10.1016/j.healun.2012.11.019CrossrefMedlineGoogle Scholar
  • 18. Frazier OH, Baldwin AC, Demirozu ZT, Segura AM, Hernandez R, Taegtmeyer H, Mallidi H, Cohn WE.Ventricular reconditioning and pump explantation in patients supported by continuous-flow left ventricular assist devices.J Heart Lung Transplant. 2015; 34:766–772. doi: 10.1016/j.healun.2014.09.015CrossrefMedlineGoogle Scholar
  • 19. Wever-Pinzon O, Drakos SG, McKellar SH, Horne BD, Caine WT, Kfoury AG, Li DY, Fang JC, Stehlik J, Selzman CH.Cardiac recovery during long-term left ventricular assist device support.J Am Coll Cardiol. 2016; 68:1540–1553. doi: 10.1016/j.jacc.2016.07.743CrossrefMedlineGoogle Scholar
  • 20. Konstam MA, Neaton JD, Dickstein K, Drexler H, Komajda M, Martinez FA, Riegger GA, Malbecq W, Smith RD, Guptha Set al; HEAAL Investigators. Effects of high-dose versus low-dose losartan on clinical outcomes in patients with heart failure (HEAAL study): a randomised, double-blind trial.Lancet. 2009; 374:1840–1848. doi: 10.1016/S0140-6736(09)61913-9CrossrefMedlineGoogle Scholar
  • 21. George RS, Sabharwal NK, Webb C, Yacoub MH, Bowles CT, Hedger M, Khaghani A, Birks EJ.Echocardiographic assessment of flow across continuous-flow ventricular assist devices at low speeds.J Heart Lung Transplant. 2010; 29:1245–1252. doi: 10.1016/j.healun.2010.05.032CrossrefMedlineGoogle Scholar
  • 22. Topkara VK, Garan AR, Fine B, Godier-Furnemont AF, Breskin A, Cagliostro B, Yuzefpolskaya M, Takeda K, Takayama H, Mancini DMet al.Myocardial recovery in patients receiving contemporary left ventricular assist devices: results from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS).Circ Heart Fail.2016; 9:10.1161/CIRCHEARTFAILURE.116.003157 e003157. doi: 10.1161/CIRCHEARTFAILURE.116.003157LinkGoogle Scholar
  • 23. Wever-Pinzon J, Selzman CH, Stoddard G, Wever-Pinzon O, Catino A, Kfoury AG, Diakos NA, Reid BB, McKellar S, Bonios Met alImpact of ischemic heart failure etiology on cardiac recovery during mechanical unloading.J Am Coll Cardiol. 2016; 68:1741–1752. doi: 10.1016/j.jacc.2016.07.756CrossrefMedlineGoogle Scholar
  • 24. Drakos SG, Terrovitis JV, Anastasiou-Nana MI, Nanas JN.Reverse remodeling during long-term mechanical unloading of the left ventricle.J Mol Cell Cardiol. 2007; 43:231–242. doi: 10.1016/j.yjmcc.2007.05.020CrossrefMedlineGoogle Scholar
  • 25. Khazanie P, Hammill BG, Patel CB, Kiernan MS, Cooper LB, Arnold SV, Fendler TJ, Spertus JA, Curtis LH, Hernandez AF.Use of heart failure medical therapies among patients with left ventricular assist devices: insights from INTERMACS.J Card Fail. 2016; 22:672–679. doi: 10.1016/j.cardfail.2016.02.004CrossrefMedlineGoogle Scholar
  • 26. Grupper A, Zhao YM, Sajgalik P, Joyce LD, Park SJ, Pereira NL, Stulak JM, Burnett JC, Edwards BS, Daly RCet alEffect of neurohormonal blockade drug therapy on outcomes and left ventricular function and structure after left ventricular assist device implantation.Am J Cardiol. 2016; 117:1765–1770. doi: 10.1016/j.amjcard.2016.03.011CrossrefMedlineGoogle Scholar
  • 27. Vaidya G, Birks E, Pillarella J, Salgado B, Vijayakrishnan R, Lenneman A, Slaughter M, Abramov D.Effects of beta blockers and ACE inhibitors after left ventricular assist device implantation.VAD J. 2018; 4:article9. doi: 10.13023/VAD.2018.09CrossrefGoogle Scholar
  • 28. Klotz S, Foronjy RF, Dickstein ML, Gu A, Garrelds IM, Danser AH, Oz MC, D’Armiento J, Burkhoff D.Mechanical unloading during left ventricular assist device support increases left ventricular collagen cross-linking and myocardial stiffness.Circulation. 2005; 112:364–374. doi: 10.1161/CIRCULATIONAHA.104.515106LinkGoogle Scholar
  • 29. Klotz S, Danser AH, Foronjy RF, Oz MC, Wang J, Mancini D, D’Armiento J, Burkhoff D.The impact of angiotensin-converting enzyme inhibitor therapy on the extracellular collagen matrix during left ventricular assist device support in patients with end-stage heart failure.J Am Coll Cardiol. 2007; 49:1166–1174. doi: 10.1016/j.jacc.2006.10.071CrossrefMedlineGoogle Scholar
  • 30. Houston BA, Schneider AL, Vaishnav J, Cromwell DM, Miller PE, Faridi KF, Shah A, Sciortino C, Whitman G, Tedford RJet alAngiotensin II antagonism is associated with reduced risk for gastrointestinal bleeding caused by arteriovenous malformations in patients with left ventricular assist devices.J Heart Lung Transplant. 2017; 36:380–385. doi: 10.1016/j.healun.2016.12.016CrossrefMedlineGoogle Scholar
  • 31. McCullough M, Caraballo C, Ravindra NG, Miller PE, Mezzacappa C, Levin A, Gruen J, Rodwin B, Reinhardt S, van Dijk Det alNeurohormonal blockade and clinical outcomes in patients with heart failure supported by left ventricular assist devices.JAMA Cardiol. 2020; 5:175–182. doi: 10.1001/jamacardio.2019.4965CrossrefMedlineGoogle Scholar
  • 32. George RS, Yacoub MH, Bowles CT, Hipkin M, Rogers P, Hallas C, Banner NR, Dreyfus G, Khaghani A, Birks EJ.Quality of life after removal of left ventricular assist device for myocardial recovery.J Heart Lung Transplant. 2008; 27:165–172. doi: 10.1016/j.healun.2007.09.029CrossrefMedlineGoogle Scholar
  • 33. Jakovljevic DG, Yacoub MH, Schueler S, MacGowan GA, Velicki L, Seferovic PM, Hothi S, Tzeng BH, Brodie DA, Birks Eet alLeft ventricular assist device as a bridge to recovery for patients with advanced heart failure.J Am Coll Cardiol. 2017; 69:1924–1933. doi: 10.1016/j.jacc.2017.02.018CrossrefMedlineGoogle Scholar
  • 34. Halliday BP, Wassall R, Lota AS, Khalique Z, Gregson J, Newsome S, Jackson R, Rahneva T, Wage R, Smith Get alWithdrawal of pharmacological treatment for heart failure in patients with recovered dilated cardiomyopathy (TRED-HF): an open-label, pilot, randomised trial.Lancet. 2019; 393:61–73. doi: 10.1016/S0140-6736(18)32484-XCrossrefMedlineGoogle Scholar


eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.