Abstract
Background
Left ventricular assist devices (LVADs) mechanically unload the heart and coupled with neurohormonal therapy can promote reverse cardiac remodeling and myocardial recovery. Minimally invasive LVAD decommissioning with the device left in place has been reported to be safe over short-term follow-up. Whether device retention reduces long-term safety, or sustainability of recovery is unknown.
Methods
This is a dual-center retrospective analysis of patients who had achieved responder status ((left ventricular ejection fraction, LVEF ≥40% and left ventricular internal diastolic diameter, LVIDd ≤6.0cm) and underwent elective LVAD decommissioning for myocardial recovery from May 2010 to January 2020. All patients had outflow graft closure and driveline resection with the LVAD left in place. Emergent LVAD decommissioning for an infection or device thrombosis was excluded. Patients were followed with serial echocardiography for up to three-years. The primary clinical outcome was survival free of heart failure hospitalization, LVAD reimplantation, or transplant.
Results
During the study period 515 patients received an LVAD and 29 (5.6%) achieved myocardial recovery, 12 patients underwent total device explantation or urgent device decommissioning, 17 patients underwent elective LVAD decommissioning and were included in the analysis. Median age of patients at LVAD implantation was 42 years (interquartile range, IQR: 25 to 54 years), all had a nonischemic cardiomyopathy and 5 (29%) were female. At LVAD implantation, median LVEF was 10% (IQR: 5 to 15%) and LVIDd 6.6 cm (IQR: 5.8 to 7.1cm). There were 11 hydrodynamically levitated centrifugal-flow (65%) and 6 axial-flow LVADs (35%). The median duration of LVAD support prior to decommissioning was 28.7 months (13.5 to 36.2 months). As compared to the turndown study parameters, one-month post-decommissioning, median LVEF decreased from 55% to 48% (p=0.03) and LVIDd increased from 4.8cm to 5.2 cm (p=0.10). There was gradual remodeling until 6 months, after which there was no statistical difference on follow-up through 3-years (LVEF 42%, LVIDd 5.6cm). Recurrent infections affected 41% of patients leading to 3 deaths and 1 complete device explant. Recurrent HF occurred in 1 patient who required a transplant. Probability of survival free of HF, LVAD, or transplant was 94% at 1-year and 78% at 3-years.
Conclusions
LVAD decommissioning for myocardial recovery was associated with excellent long-term survival free from recurrent heart failure and preservation of ventricular size and function up to 3-years. Reducing the risk of recurrent infections, remains an important therapeutic goal for this management strategy.
Keywords: LVAD, myocardial recovery, heart failure, mechanical circulatory support
Introduction
Left Ventricular Assist Devices (LVADs) have evolved into an essential therapy for patients with advanced heart failure (HF) and recalcitrant symptoms despite optimal medical management. Through mechanical unloading, LVADs reduce myocardial volume and pressure and promote reverse cardiac remodeling when coupled with HF neurohormonal therapy.1 Ultimately these effects can culminate in myocardial remission or recovery, the later implying long-term freedom from recurrent HF.2 Myocardial recovery with LVAD support has been recapitulated by several investigators.3–9
Following evidence of myocardial remission and satisfactory hemodynamic and echocardiographic findings during an LVAD turndown study, the pump is typically explanted through a median sternotomy followed by complete removal of the pump housing, inflow cannula, percutaneous driveline, and outflow graft with closure of the ventriculotomy site and aortic anastomosis. This approach, however, requires a repeat sternotomy which is an independent risk factor for cardiac injury, perioperative complications, and increases operative risk if a subsequent sternotomy is required. Recently we have shown that a minimally invasive approach to LVAD decommissioning with the device left in place avoids a resternotomy and is safe with respect to infection and thromboembolic events up to one-year post-decommissioning.10, 11 Long-term sustainability of this approach and freedom from adverse events remains undefined.
After discontinuation of LVAD support, the risk for recurrent heart failure and other cardiovascular events is highest in the first year. Long-term survival free from LVAD, transplant or death after LVAD explantation for myocardial recovery is approximately 90% at one-year, and 77% at three-years post-explant.12 However, for patients that undergo minimally invasive LVAD decommissioning, the effect of the retained pump on the sustainability of myocardial recovery is unknown. In addition, freedom from recurrent HF and longer-term survival has yet to be determined. We sought to describe the long-term outcomes and sustainability of myocardial recovery following elective LVAD decommissioning. Our primary hypothesis was that myocardial recovery patients undergoing LVAD decommissioning would have similar freedom from HF and long-term survival as total device explantation. We also describe the long-term risk of infection and thromboembolic complications in patients undergoing LVAD decommissioning for myocardial recovery.
Methods
Study Design
This is a dual-center retrospective analysis of patients undergoing elective LVAD decommissioning following evidence of myocardial recovery on clinical testing during an LVAD weaning study. The patients were implanted and followed at the Inova Heart & Vascular Institute in Falls Church, Virginia, United States and Freeman Hospital in Newcastle upon Tyne, United Kingdom. All patients from with myocardial recovery underwent LVAD decommissioning with either outflow graft surgical ligation through a right anterior mini-thoracotomy or percutaneous closure with an Amplatzer device (n=1, Abbott, Abbott Park IL). The driveline was then resected, and the device left in place. The institutional review boards at each center approved these analyses. The Newcastle data adhered to the terms of the United Kingdom Data and Protection Act and Freedom of Information Act and was approved to obtain confidential information by the local Caldicott Guardian. At Inova a protocol exemption was granted by the IRB (#11.086).
Study Population and Medical Therapy
LVAD patients with echocardiographic findings of reverse remodeling on routine echocardiography were selected for aggressive heart failure medication and speed titration to maximize mechanical unloading (e.g., reduce severity of mitral regurgitation and/or left ventricular size). The medical therapy protocol implemented at our centers follows that of the Remission from Stage D Heart Failure (RESTAGE-HF) protocol.12 Eligible patients had elective LVAD decommissioning for myocardial recovery from May 2010 to January 2020. Demographic, medical history, medication, hemodynamic, and echocardiography data of patients at time of LVAD implantation and serially after LVAD was collected. Patients with urgent/emergent LVAD decommissioning for an active infection or device thrombosis were excluded from the analysis as they did not have sufficient time for maximization of HF medications and to undergo formal recovery testing.
LVAD Weaning Study & Decommissioning Criteria
For Inova patients, after achieving myocardial remission on routine echocardiography and after providing informed consent, patients underwent an LVAD weaning study to analyze myocardial structure, function, and hemodynamics.13 In brief, patients were maintained on antithrombotic therapy with aspirin and warfarin. They were also treated with supplemental intravenous heparin to achieve a therapeutic partial thromboplastin time (PTT). In the cardiac catheterization suite, a right heart catheterization was performed with simultaneous echocardiographic imaging to assess. Several hemodynamic and echocardiographic parameters were obtained (Supplemental Table 1). The HeartMate II (Abbott, Abbott Park IL) speed was reduced in 1,000 revolutions per minutes (RPM) increments every 2–3 minutes until a nadir of 6,000 RPM. For the HVAD (Medtronic, Minneapolis MN), the speed was reduced in 200 RPM increments until a negative deflection of the flow wave indicating regurgitant flow in the outflow graft (as seen on the system monitor), typically 1,800 RPM. After the nadir device speed is reached, the patient remains at that speed for a minimum of ten minutes before echocardiographic imaging and hemodynamic testing.
Patients are then transported to the cardiopulmonary stress testing lab for a treadmill exercise test with simultaneous assessment of minute-by-minute gas exchange. Finally, a 6-minute walk test is performed. The patients are maintained at low speed for approximately 4–6 hours to complete the outlined testing.
Newcastle patients underwent a five-day inpatient low-flow assessment. On the first day, echocardiography was performed on full support and then reduced support. If there was no change in cardiac function, the patient was maintained on reduced support and could proceed with the complete assessment. By day four of reduced support, the patient underwent right heart catheterization and cardiopulmonary stress testing. On day 5, the patient underwent computed tomography (CT) angiogram and final echocardiographic assessment.
Patients at both centers proceeded with device decommissioning when they meet all hemodynamic and echocardiographic criteria (left ventricular internal diastolic diameter (LVIDd < 60mm), left ventricular end-systolic diameter <50 mm, left ventricular rejection fraction (LVEF >45%), pulmonary capillary wedge pressure ≤15 mmHg, cardiac index >2.4L/min/m2, maximum oxygen consumption (VO2) with exercise >16 ml/kg/min. The VO2 criterion was considered optional and not mandated at both centers to proceed with LVAD decommissioning.
Post-Implant Management and Monitoring after Decommissioning
After LVAD decommissioning, patients underwent routine transthoracic echocardiography every 3 months for the first year and every 6 months thereafter. Post-decommissioning antithrombotic therapy differed between the two sites. At Newcastle, all patients were continued on systemic anticoagulation with warfarin for an INR goal of 1.5 – 2.0 and no aspirin was used. At Inova, all patients were managed with aspirin 325 mg daily for the first year and 81mg daily thereafter unless there was another indication for systemic anticoagulation (e.g., atrial fibrillation or history of thromboembolism). Patients were maintained on neurohormonal therapy for HF using maximally tolerated doses. Compliance with optimal HF medical therapy was also monitored throughout follow-up.
Statistical Analysis
The primary outcome of the analysis was survival free of recurrent HF requiring hospitalization, LVAD reimplantation, or transplantation. Recurrent HF was defined as new signs or symptoms of HF requiring hospitalization. Secondary outcomes collected included device or wound infection, thromboembolism, and cardiac operative procedures for any cause. Study data were collected and managed using REDCap electronic data capture system hosted at the Inova Health System.14 Since the data were skewed, continuous values are reported as median with interquartile range (IQR). Categorial values are reported as number and percentage. The results of serial echocardiography prior to and after device decommissioning were compared using the Mann-Whitney U Test or Kruskal-Wallis Test. Survival free of the primary outcome was estimated using Kaplan-Meier methods. Statistical analysis was performed using GraphPad Prism version 8.4.1 (GraphPad Software, San Diego, CA, USA) and p-values < 0.05 were considered statistically significant.
Results
Patient Characteristics
During the study period, 515 patients received an LVAD at both institutions. A total of 29 patients (5.6%) had evidence of myocardial recovery (LVEF ≥ 40% and LVIDd ≤ 6.0cm) on routine echocardiography. In this group, 12 patients underwent total device explantation or urgent device decommissioning for a device infection or thrombus. The 17 remaining patients had an elective minimally invasive device decommissioning and were included in this analysis (Figure 1). The median age of the myocardial recovery patients at time of LVAD implantation was 42.0 years (IQR: 25.0 to 54.0) years and they were younger and more likely to be female than the remainder of the LVAD patient population at Inova and Newcastle (Supplemental Table 2). All myocardial recovery patients had a non-ischemic cardiomyopathy (NICM). At the time of LVAD implantation, the median LVEF was 10% (IQR: 5 to 15), the LVIDd was 6.6 cm (IQR: 5.8 to 7.1) and 88% of patients were supported with inotropes. Patients received either the HVAD (n=11, 65%) or HeartMate II (n=6, 35%) (Table 1).
Figure 1.
Recovery & Device Explantation Selection
Consort diagram provides follow-up data on all continuous-flow LVAD implants at the Inova Heart and Vascular Institute and Newcastle during the study period. In an unselected population 5.6% of LVAD patients had evidence of reverse cardiac remodeling and myocardial recovery on routine echocardiography achieving an LVEF ≥ 40% and LVIDd ≤ 6.0cm (i.e., Responder). For patients undergoing device decommissioning the median time on support was 28.7 months.
Table 1.
Myocardial Recovery Patient Characteristics at Time of LVAD Implantation
| Decommissioned Patients (N = 17) | |
|---|---|
| Demographics | |
| Age, years, median (IQR) | 42.0 (25.0 to 54.0) |
| Sex, n (%) | |
| Race | |
| Medical History | |
| Cardiomyopathy Type | |
| INTERMACS Profile | |
| Heart Failure Duration, months (range) | 3 (1 – 180) |
| Diabetes Mellitus | 2 (12%) |
| Hypertension | 2 (12%) |
| Atrial Fibrillation | 3 (18%) |
| ICD or CRT-D | 3 (18%) |
| Medications | |
| Digoxin | 2 (12%) |
| Device implanted | |
| HMII | 6 (35%) |
| HVAD | 11 (65%) |
| Device Indication | |
| Bridge to Transplantation | 12 (71%) |
| Destination Therapy | 5 (29%) |
| Echo Parameters Pre-LVAD | |
| LVEF, % | 10.0 (5.0 to 15.0) |
| LVIDd, cm | 6.6 (5.8 to 7.1) |
| Laboratory Values | |
| Creatinine, mg/dL | 1.1 (0.9 to. 1.3) |
| Blood Urea Nitrogen, mg/dL | 20.1 (17.0 to 25.8) |
| Total Bilirubin, mg/dL | 1.5 (0.9 to 1.9) |
ACE-I: angiotensin converting enzyme inhibitors, HMII: heartmate II, ICD: implantable cardioverter-defibrillator, CRT-D: cardiac resynchronization therapy with defibrillator, LVEF: left-ventricular ejection fraction, LVIDd: left ventricular internal diastolic diameter.
Hemodynamics and Echocardiographic Parameters at Time of LVAD Weaning
Patients with echocardiographic evidence of reverse remodeling at full support, had medical therapy for HF optimized and were then selected for an LVAD weaning study as outlined in the method section. At the time of the LVAD weaning study, median LVEF was 55% (IQR: 50 to 60), LVIDd 4.8 cm (IQR: 4.3 to 5.1), pulmonary capillary wedge pressure 8.0 mmHg (IQR: 8.0 to 12.0), and cardiac index 2.4 L/m2 (IQR: 2.0 to 2.7). The peak VO2 with exercise was 21.2 mL/kg/min (IQR: 15.7 to 23.8). Of note 2 patients did not have a VO2 performed prior to device decommissioning and 5 patients had a peak VO2 < 16ml/kg/min. Other hemodynamic, echocardiographic, and exercise testing parameters are provided in Table 2. The median duration of LVAD support was 28.7 months (IQR: 13.5 to 36.2) prior to device decommissioning.
Table 2.
LVAD Weaning Study Parameters Prior to Device Decommissioning
| Decommissioned Patients (N = 17) | |
|---|---|
| Length of LVAD Support, months, median (IQR) | 28.7 (13.5 to 36.2) |
| Low-flow assessments prior to decommissioning, # | 1.0 (1.0 to 2.0) |
| Time from Testing to device decommissioning, days | 76.5 (55 to 123) |
| Echocardiographic Parameters | |
| LVEF, % | 55 (50 to 60) |
| LVIDd, cm | 4.8 (4.3 to 5.1) |
| LVIDs, cm | 3.3 (2.9 to 3.8) |
| Mitral Regurgitation | |
| Aortic Insufficiency | |
| Right Ventricular Function | |
| Hemodynamic Parameters | |
| Right Atrial Pressure, mmHg | 5.0 (2.0 to 7.0) |
| Pulmonary Capillary Wedge Pressure, mmHg | 8.5 (8.0 to 12.3) |
| Pulmonary Arterial Systolic Pressure, mmHg | 25.0 (20.0 to 28.0) |
| Pulmonary Arterial Diastolic Pressure, mmHg | 10.0 (8.0 to 14.0) |
| Mixed Venous Oxygen Saturation, % | 68.4 (66.0 to 71.3) |
| Cardiac Output (Fick), L/min | 4.5 (4.2 to 5.3) |
| Cardiac Index (Fick), L/m2 | 2.4 (2.0 to 2.7) |
| Exercise Testing Parameters | |
| Peak VO2, mL/kg/min | 21.2 (15.7 to 23.8) |
| VE/VCO2* | 25.8 (24.0 to 34.1) |
| 6-minute walk distance, meters* | 356.6 (337.5 to 426.0) |
Data only reported for Inova patients.
LVEF: left-ventricular ejection fraction, LVIDd: left ventricular internal diastolic diameter, LVIDs: left ventricular internal systolic diameter, VO2: peak oxygen consumption, Ve/VCO2: ventilatory efficiency.
With respect to medical therapy at time of the weaning study, 88% of patients were on a β-blocker, 94% on an angiotensin-converting enzyme inhibitor (ACE-I), angiotensin II receptor blocker (ARB), or angiotensin II receptor blocker / neprilysin inhibitor (ARNI), 76% on mineralocorticoid receptor antagonist (MRA), and 47% were treated with digoxin (Table 3).
Table 3.
Medical Therapy Before and After LVAD Decommissioning
| 3 years (n=6) | ||||||
|---|---|---|---|---|---|---|
| β-blocker, n (%) | 15 (88%) | 14 (100%) | 12 (100%) | 13 (100%) | 6 (100%) | 6 (100%) |
| Carvedilol, n, (median daily dose in mg) | 7 (50) | 5 (50) | 5 (50) | 5 (75) | 3 (75) | 3 (80) |
| Metoprolol succinate | 1 (300) | 1 (300) | 1 (300) | 1 (300) | 0 | 0 |
| Bisoprolol | 7 (10) | 8 (8.75) | 6 (8.75) | 7 (10) | 3 (10) | 3 (10) |
| ACE/ARB/ARNI, n (%) | 16 (94%) | 13 (93%) | 11 (92%) | 13 (100%) | 6 (100%) | 6 (100%) |
| Lisinopril, n, (median daily dose in mg) | 14 (10) | 11 (10) | 9 (10) | 11 (15) | 6 (10) | 6 (15) |
| Sacubitril-Valsartan, mg | 1 (200) | 1 (200) | 0 | 0 | 0 | 0 |
| Losartan, mg | 0 | 0 | 1 (100) | 1 (100) | 0 | 0 |
| Candesartan, mg | 1 (4) | 1 (4) | 1 (4) | 1 (4) | 0 | 0 |
| Aldosterone Blockade, n (%) | 13 (76%) | 9 (64%) | 9 (75%) | 9 (69%) | 4 (67%) | 3 (50%) |
| Digoxin, n (%) | 8 (47%) | 5 (36%) | 3 (25%) | 4 (31%) | 2 (33%) | 0 (0%) |
ACE-I: angiotensin-converting enzyme inhibitor (ACE-I), ARB: angiotensin II receptor blocker: ARNI: ARB with neprilysin inhibitor.
Post-Decommissioning Characteristics
Median follow-up time after LVAD decommissioning was 28 months (IQR: 14 to 49). Patients were anticoagulated with either warfarin (53%) or aspirin alone (47%), depending on the respective institutional protocol. At one-year post-decommissioning, all patients were treated with β-blockers (100%), ACE/ARB/ARNI (100%), and most were continued on an MRA (69%), and 4 patients remained on digoxin (31%) (Table 3).
Immediately after discontinuation of LVAD support, the median LVEF decreased by 7% to 48% (IQR: 39 to 55, p = 0.03, Figure 2A) and the LVIDd increased by 0.4 cm to 5.2 cm (IQR: 4.6 to 5.8, p = 0.10, Figure 2B). There was a mild reduction in LVEF and increase in LVIDd up until approximately 6 months after device decommissioning reaching an LVEF of 43% (IQR: 35 to 54) and LVIDd 5.3 cm (IQR: 4.7 to 6.0). Afterwards the LVEF and LVIDd stabilized until 3-years post-decommissioning; the LVEF was 42% (IQR: 28 to 49) and the LVIDd 5.8cm (IQR: 5.0 to 6.6, Table 4). Echocardiographic remodeling after LVAD decommissioning and individual patient changes in LVEF and LVIDd are plotted in Figure 3.
Figure 2.
Serial Echocardiographic Follow-Up for Assessment of Cardiac Function (A) and Structure (B) Following LVAD Decommissioning
Serial echocardiographic imaging after LVAD decommissioning depicts the longitudinal change in median LVEF and LVIDd during follow-up. There was a reduction in the LVEF and LVIDd from time of weaning study (55% and 4.8cm) to 6-months (40% and 5.4cm, respectively). Thereafter, both parameters stabilized until 3-year follow-up (42% and 5.8cm).
Table 4.
Echocardiographic Assessment of Myocardial Function Following LVAD Decommissioning
| 3 years (n=5) | ||||||
|---|---|---|---|---|---|---|
| LVEF, %, median (IQR) | 55.0 (50.0 to 60.0) | 42.5 (38.8 to 54.8) | 42.5 (35.0 to 53.8) | 37.5 (30.0 to 45.0) | 32.5 (27.5 to 45.3) | 41.5 (27.5 to 48.75) |
| LVIDd, cm | 4.9 (4.4 to 5.2) | 5.5 (5.0 to 5.9) | 5.3 (4.7 to 6.0) | 5.7 (5.4 to 6.2) | 5.6 (5.1 to 6.0) | 5.8 (5.0 to 6.6) |
| Mitral Regurgitation | ||||||
| None, n (%) | 12 (71%) | 10 (71%) | 6 (46%) | 4 (29%) | 4 (67%) | 2 (40%) |
| Mild | 4 (24%) | 4 (29%) | 6 (46%) | 9 (64%) | 2 (33%) | 2 (40%) |
| Moderate-Severe | 1 (6%) | 0 (0%) | 0 (0%) | 1 (7%) | 0 (0%) | 1 (20%) |
| Aortic Insufficiency | ||||||
| None | 16 (94%) | 12 (86%) | 11 (85%) | 14 (100%) | 6 (100%) | 4 (80%) |
| Mild | 1 (6%) | 2 (14%) | 1 (8%) | 0 (0%) | 0 (0%) | 1 (20%) |
| Moderate-Severe | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Right Ventricular Function | ||||||
| Normal | 10 (59%) | 5 (36%) | 2 (15%) | 4 (29%) | 1 (17%) | 0 (0%) |
| Mildly Reduced | 3 (18%) | 2 (14%) | 6 (46%) | 4 (29%) | 3 (50%) | 4 (80%) |
| Moderately Reduced | 4 (24%) | 5 (36%) | 4 (31%) | 4 (29%) | 1 (17%) | 0 (0%) |
| Severely Reduced | 0 (0%) | 2 (14%) | 0 (0%) | 2 (14%) | 1 (17%) | 0 (0%) |
LVEF: left-ventricular ejection fraction, LVIDd: left ventricular internal diastolic diameter.
Figure 3.
Individual Patient Trajectories for LVEF and LVIDd after LVAD Decommissioning
Longitudinal echocardiographic imaging follow-up of individual patient trajectories of left ventricular structure and function after device decommissioning.
Long-term Adverse Events, Heart Failure Recurrence, and Patient Survival
After device decommissioning, no patients required LVAD reimplantation. One patient (6%) had recurrent heart failure requiring a heart transplant at 73 days post-decommissioning. Survival free from HF reoccurrence, LVAD reimplantation, or transplant at one-, two- and three-years was 94%, 87%, and 78%, respectively (Figure 4).
Figure 4.
Survival Free of HF Recurrence Requiring Hospitalization, LVAD Reimplantation, or Transplantation Following Decommissioning of LVAD Support
Estimated survival free of recurrent HF, LVAD, transplant or death after device decommissioning with number at risk at each time point. Dashed line represents the 95% confidence interval of the survival estimate.
While on active LVAD support, 8 patients (47%) had either a driveline infection or device-specific infection. After device decommissioning, 88% of these patients (n=7) developed a recurrent driveline exit site (n = 4) or right anterior mini-thoracotomy infection (n = 3) at a median of 239 days (range: 77–809, Supplemental Figure 1). Three patients died at a median of 840 days (range 413 to 1422) after device decommissioning (Figure 4), and the last patient with a device infection had the device fully explanted without any subsequent events.
Three patients (18%) had a thromboembolic complication at a median of 412 days (range: 286 to 862) after LVAD decommissioning. These patients all had an HVAD device/retained inflow cannula and two occurred despite therapeutic anticoagulation with warfarin and one occurred while on aspirin monotherapy; this later patient was found to have a prothrombin gene mutation and had the device completely explanted (Supplemental Figures 2 & 3).15 Of these three patients, 2 were diagnosed with a left ventricular thrombus, one which was complicated by a stroke and one without. The last patient had a stroke without an identified ventricular thrombus.
Discussion
In this dual-center analysis of 17 patients that underwent LVAD decommissioning with the device left in place for myocardial recovery, after decommissioning using a minimally invasive method: 1) the LV function showed a mild reduction in left ventricular function and increase in ventricular size that starts immediately after LVAD decommissioning and then stabilizes at 6-months, being maintained until 3-years follow-up; 2) excellent one- and three-year survival free of recurrent heart failure requiring hospitalization, LVAD reimplantation, or cardiac transplantation at 94% and 78% respectively; and finally 3) a high burden of recurrent infections leading to device explant or death which warrants careful consideration and implementation of novel strategies to mitigate infections in future patients being considered for device decommissioning.
Time Course of Reverse Cardiac Remodeling and Myocardial Recovery
Approximately 10% of LVAD patients have evidence of reverse cardiac remodeling and are termed responders (LVEF ≥ 40%, LVIDd ≤ 6.0cm) to mechanical unloading.7, 8, 16 Despite, this the reported frequency of device explant/decommissioning for myocardial recovery is only ~1% at one-year and grows to ~4% at 5-years.7, 8, 17 Patients with evidence of myocardial recovery leading to device explant/decomissioning, constituted 5.6% of our unselected patient population. We identified that the median support duration prior to device decommissioning was approximately 2.5 years, which is much longer than previous publications which report a median time to explant of 9-months.3, 12 Importantly, these are prospective studies that enriched for patients with a high propensity for recovery (e.g., younger age, NICM), and immediately after LVAD implantation started aggressive titration of neurohormonal therapy with simultaneous adjustment of LVAD-induced mechanical unloading. As LVAD patients remain on support for longer periods of time, we expect to see higher rates of reverse cardiac remodeling and myocardial recovery.
Long-term Freedom from Heart Failure
A number of previous publications have reported excellent freedom from recurrent HF after device explantation with both pulsatile,18 and continuous-flow LVADs.3, 4, 6 The recently published RESTAGE-HF study suggested that in a multicenter cohort of NICM patients, who received an aggressive prospective protocol that couples mechanical unloading coupled with neurohormonal therapy; device explantation could be successfully achieved in 50% of patients.12 The long-term survival free of HF in RESTAGE-HF requiring an LVAD or transplant was 90% at one-year and 77% at two-years post-explant. This study which included predominantly patients who had their entire LVAD explanted, parallels our analysis with a 94% one-year, and 78% three-year survival free of recurrent HF, LVAD or transplant. This suggests that device decommissioning is not associated with higher rates of recurrent heart failure that lead to LVAD, transplant, or death. Only 1 patient in our series developed symptomatic heart failure ultimately requiring a heart transplant. It is important to note that across all studies, when patients achieve clinical criteria for device explantation and are maintained on long-term neurohormonal therapy, the recurrence rate of HF is quite low,19 and survival after explantation parallels survival after cardiac transplantation.20 In addition, functional capacity as measured by peak oxygen consumption parallels normal patients and is higher than patients who remain on LVAD support or receive a cardiac transplant.9
Myocardial Remission and Durability of Recovery
Aside from HF recurrence, we also evaluated serial changes in left ventricular size and function in our patient cohort after LVAD decommissioning. All patients in the cohort were considered responders and met criteria for discontinuation of device support achieving a median LVEF of 55% and LVIDd of 4.8cm. We observed that immediately after LVAD decommissioning, there is a mild drop in LVEF and increase in LV size which parallels changes that occur when a new LVAD is initially implanted.16 Afterwards we noticed mild adverse remodeling up until 6-months, after which time the LVEF and LVIDd stabilizes, and at 3-years were 42% and 5.8cm, respectively. These patients would still be considered responders and enjoy long-term freedom from recurrent heart failure without the complications of ongoing LVAD therapy or after cardiac transplant.16, 20
The concept of both molecular and clinical partial recovery has been noted by numerous investigators.2 Even in patients meeting the clinical criteria for LVAD explantation, there is a persistent molecular signature of HF, which suggests only partial response. Long-term freedom from HF recurrence is an important goal, but even in medically-treated HF patients with a recovered LVEF, there is an ongoing risk for HF recurrence, suggesting myocardial remission and not complete recovery.21 Given post-explantation survival that parallels transplantation after device explantation, myocardial remission may be a reasonable goal for many LVAD patients with informed decision making. Factors that promote long-term sustainability of myocardial recovery need ongoing research to enhance this overall therapeutic strategy.22
Complications
We noted a high incidence of infection after device decommissioning. Of these patients who developed recurrent infections, three subsequently died and one required complete device explant. In a recent systematic analysis by Choi et al, that compared clinical outcomes of LVAD patients who had recovery and underwent total device explant compared to patients who had device decommissioning, they noted similar rates of infection amongst both patient cohorts, 14.3% vs. 13.3%.23 The reason for a high rate of infections in our study may be attributed to a large portion of patients having a history of driveline infection while on LVAD support (47%). This is consistent with prior publications that suggest a 42% incidence of infection for LVAD patients at one-year which increases with longer support durations.24 In our series, 4 patients experienced an infection at the right anterior mini-thoracotomy site.
In our analysis 2 patients suffered a cerebrovascular accident, none of which was associated with long-term sequale. In a previous analysis the rate of stroke after device explantation was noted to be 6.0%, with no reported strokes in the patients who had device support discontinued with the LVAD retained.23 Frazier et al. reported a stroke rate of 11% after full LVAD explantation for recovery which is similar to our analysis.6 Of note, we observed that 2 patients developed a thrombus in the left ventricle or inflow thrombus, suggesting the retained LVAD inflow cannula is a potential nidus for thrombus formation. One patient was found to have a prothrombin gene mutation as the likely cause of the thrombotic event.15 Of note, in a prior report by Padera and colleagues, a high incidence of thrombus at the outer surface of the HVAD inflow cannula may explain the higher incidence of thromboembolic events with that device. All of our patients with a stroke or inflow cannula thrombus had an HVAD.25 Thromboembolism occurred despite warfarin anticoagulation in 2 patients and aspirin in 1 patient. Thromboembolism after device decommissioning could be reduced with a novel tip design that incorporates material that is less thrombogenic,26 optimal inflow cannula positioning within the ventricle,27 and a tipless inflow design may enhance the safety of device decommissioning after myocardial recovery.28
Alternative Approaches to Surgical Device Decommissioning
Given the high rate of infections identified in our analysis, it is important to consider alternative techniques to mitigate the risk of surgical site or pump infection which includes total device explantation or a hybrid approach that combines percutaneous closure of the outflow graft with the Amplatzer device (Abbott, Abbott Park IL) and surgical excision of the driveline site. If total device explant is performed the ventriculotomy site can be closed with a titatinium plug,29–31 or ventricular patch plasty.32 If a percutaenous approach is used, the endovascular device is deployed in the cardiac catheterization lab using a transaortic technique, with placement of the occluder in the distal outflow graft prior to the aortic anastamosis.33 The driveline can then be excised in the cardiac catheterization suite or operating room. Both endovascular and surgical outflow graft ligation carry a Class IIa (level of evidence C) recommendation for management of patients with myocardial recovery undergoing device decommissioning.13.
The percutaneous driveline is an important source of ascending infections in LVAD patients and this often occurs through biofilm formation. Approximately 30% of asymptomatic LVAD patients have evidence of a bacterial pathogens on the driveline, which have the potential to tunnel and extend to the pump.34 The process of biofilm propagation is facilitated by microgaps in the driveline, interface between the smooth portion of the driveline and velour fibers, and insufficient tissue ingrowth along the driveline.35 In the future, fully implantable devices which promise to reduce the risk of device-specific infections by eliminating peripheral components and the percutaneous driveline are likely to enhance the safety of device decomissioning. In our analysis, 3 patients had recurrent driveline site infections after decommissioning that required surgical debridement suggesting the driveline remanent is an important source of long-term infection. For current patients being considered for LVAD decommissioning with a history of driveline infections (active or past infection), one approach to assess candidacy for decommissioning or complete explantation, is to pursue fluorine-18 flurodeoxyglucose (FDG) or gallium-67 single-photon emission computed tomography (Ga-SPECT-CT).36, 37 These imaging techniques have a high sensitivity and specificity for detecting infections involving the internal component of the driveline, pump, inflow cannula, and outflow graft. Patients with positive scans concerning for infection could then be counseled that complete device removal may be more appropriate than device decommissioning or should be considered for long-term suppressive antibiotic therapy if complete device explant is not feasible.
Limitations
We report the largest series to date of patients undergoing LVAD decommissioning for myocardial recovery but appreciate that our sample size is limited which likely effects our reported estimates of survival and echocardiographic parameters especially at later time points. We are encouraged by HF event rates that seem to parallel the published experience with complete device explantation but recognize that the absence of a comparator group from our centers that underwent explantation is a limitation. This was a retrospective analysis, and we were unable to standardize neurohormonal therapy after decommissioning or the time intervals for follow-up clinical visits and monitoring. It is quite likely that with a prospective protocol that couples closer monitoring of interventricular filling pressures and monitors adherence to neurohormonal therapy; the sustainability of the reverse cardiac remodeling could be enhanced. Finally, each center had a varying antithrombotic therapy management protocol after LVAD decommissioning (Newcastle warfarin and Inova aspirin alone), which could have affected the rates of thromboembolism. Nevertheless, our rates of thromboembolic disease are like prior publications where patients had the device fully explanted.
Conclusions
These data show that LVAD decommissioning for myocardial remission appears to be associated with a similar risk of recurrent heart failure on long-term follow-up, as complete device explantation. Acute ventricular loading with device decommissioning led to adverse cardiac remodeling up until 6-months afterwards, thereafter the ventricular function and size stabilized until 3-years. For HVAD patients with myocardial recovery, complete device explant if feasible is recommended to reduce the risk of subsequent thromboembolic complications. A high rate of infection was noted, and the associated risk of mortality warrants implementation of novel strategies such as percutaneous closure outflow graft closure to reduce the risk of surgical site infections and imaging-based screening for deeper VAD-specific infections to reduce the risk of recurrent infections after device decommissioning.
Supplementary Material
Funding:
The work in this study is supported by a National Institutes of Health, Career Development Award K23 1K23HL143179 awarded to Dr. Shah
Disclosures:
Dr. Shah reports support for this work from an NIH K23 Career Development Award 1K23HL143179 and grant support from American Heart Association / Enduring Hearts, Abbott, Merck, Roche and Bayer and personal fees from Procyrion, Merck, Roche, Novartis, and Ortho Clinical Diagnostics for unrelated work. Dr. Wang is supported by the National Institute for Health Research (NIHR) Blood and Transplant Research Unit (BTRU) in Organ Donation and Transplantation at the University of Cambridge in collaboration with Newcastle University and in partnership with National Health Service Blood and Transplant (NHSBT). Dr. Psotka has received personal fees from Amgen, Cytokinetics, and Windtree, and grant support from the United States Food and Drug Administration. Dr. Desai receives honoraria for speaking from Abbott. Drs MacGowan and Jakovljevic are supported by European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement no. 777204 and the UK National Institute of Health Research.
Abbreviations
- HF
Heart Failure
- HMII
HeartMate II
- HVAD
HeartWare HVAD
- LVAD
Left Ventricular Assist Device
- LVEF
Left Ventricular Ejection Fraction
- LVIDd
Left Ventricular Internal Diameter End Diastole
- LVIDs
Left Ventricular Internal Diameter End Systole
- NICM
Non-ischemic cardiomyopathy
- PCWP
Pulmonary Capillary Wedge Pressure
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Drakos SG, Pagani FD, Lundberg MS and Baldwin JT. Advancing the Science of Myocardial Recovery With Mechanical Circulatory Support: A Working Group of the National, Heart, Lung and Blood Institute. J Card Fail 2017;23:416–421. [DOI] [PubMed] [Google Scholar]
- 2.Mann DL, Barger PM and Burkhoff D. Myocardial recovery and the failing heart: myth, magic, or molecular target? J Am Coll Cardiol. 2012;60:2465–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Birks EJ, George RS, Hedger M, Bahrami T, Wilton P, Bowles CT, Webb C, Bougard R, Amrani M, Yacoub MH, et al. Reversal of severe heart failure with a continuous-flow left ventricular assist device and pharmacological therapy: a prospective study. Circulation. 2011;123:381–90. [DOI] [PubMed] [Google Scholar]
- 4.Birks EJ, Tansley PD, Hardy J, George RS, Bowles CT, Burke M, Banner NR, Khaghani A and Yacoub MH. Left ventricular assist device and drug therapy for the reversal of heart failure. N Engl J Med. 2006;355:1873–84. [DOI] [PubMed] [Google Scholar]
- 5.Drakos SG, Wever-Pinzon O, Selzman CH, Gilbert EM, Alharethi R, Reid BB, Saidi A, Diakos NA, Stoker S, Davis ES, et al. 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;61:1985–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Frazier OH, Baldwin AC, Demirozu ZT, Segura AM, Hernandez R, Taegtmeyer H, Mallidi H and Cohn WE. Ventricular reconditioning and pump explantation in patients supported by continuous-flow left ventricular assist devices. The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation. 2015;34:766–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Topkara VK, Garan AR, Fine B, Godier-Furnemont AF, Breskin A, Cagliostro B, Yuzefpolskaya M, Takeda K, Takayama H, Mancini DM, et al. Myocardial Recovery in Patients Receiving Contemporary Left Ventricular Assist Devices: Results From the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS). Circulation Heart failure. 2016;9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wever-Pinzon O, Drakos SG, McKellar SH, Horne BD, Caine WT, Kfoury AG, Li DY, Fang JC, Stehlik J and Selzman CH. Cardiac Recovery During Long-Term Left Ventricular Assist Device Support. J Am Coll Cardiol. 2016;68:1540–53. [DOI] [PubMed] [Google Scholar]
- 9.Jakovljevic DG, Yacoub MH, Schueler S, MacGowan GA, Velicki L, Seferovic PM, Hothi S, Tzeng BH, Brodie DA, Birks E, et al. Left Ventricular Assist Device as a Bridge to Recovery for Patients With Advanced Heart Failure. J Am Coll Cardiol. 2017;69:1924–1933. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kapoor K, Shah P, Bogar LJ, Edwards LG, Panjrath GS and Singh R. Long-term Safety of Minimally Invasive Left Ventricular Assist Device Discontinuation for Myocardial Recovery. The Annals of thoracic surgery. 2019;108:1398–1403. [DOI] [PubMed] [Google Scholar]
- 11.MacGowan GA, Wrightson N, Robinson-Smith N, Woods A, Parry G, Gould K and Schueler S. Myocardial Recovery Strategy with Decommissioning for the HeartWare Left Ventricular Assist Device. ASAIO journal. 2017;63:299–304. [DOI] [PubMed] [Google Scholar]
- 12.Birks EJ, Drakos SG, Patel SR, Lowes BD, Selzman CH, Starling RC, Trivedi J, Slaughter MS, Alturi P, Goldstein D, et al. 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. Circulation. 2020;142:2016–2028. [DOI] [PubMed] [Google Scholar]
- 13.Potapov EV, Antonides C, Crespo-Leiro MG, Combes A, Farber G, Hannan MM, Kukucka M, de Jonge N, Loforte A, Lund LH, et al. 2019 EACTS Expert Consensus on long-term mechanical circulatory support. Eur J Cardiothorac Surg. 2019;56:230–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N and Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. Journal of biomedical informatics. 2009;42:377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Eltelbany M, Singh R, Genovese L and Shah P. Delayed Presentation of Thrombophilia After Left Ventricular Assist Device Deactivation for Reverse Cardiac Remodeling. Circulation Heart failure. 2020;13:e007062. [DOI] [PubMed] [Google Scholar]
- 16.Shah P, Psotka M, Taleb I, Alharethi R, Shams MA, Wever-Pinzon O, Yin M, Latta F, Stehlik J, Fang JC, et al. Framework to Classify Reverse Cardiac Remodeling With Mechanical Circulatory Support: The Utah-Inova Stages. Circulation Heart failure. 2021;14:e007991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Molina EJ, Shah P, Kiernan MS, Cornwell WK 3rd, Copeland H, Takeda K, Fernandez FG, Badhwar V, Habib RH, Jacobs JP, et al. The Society of Thoracic Surgeons Intermacs 2020 Annual Report. The Annals of thoracic surgery. 2021. [DOI] [PubMed] [Google Scholar]
- 18.Dandel M, Weng Y, Siniawski H, Stepanenko A, Krabatsch T, Potapov E, Lehmkuhl HB, Knosalla C and Hetzer R. Heart failure reversal by ventricular unloading in patients with chronic cardiomyopathy: criteria for weaning from ventricular assist devices. European heart journal. 2011;32:1148–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Antonides CFJ, Schoenrath F, de By T, Muslem R, Veen K, Yalcin YC, Netuka I, Gummert J, Potapov EV, Meyns B, et al. Outcomes of patients after successful left ventricular assist device explantation: a EUROMACS study. ESC Heart Fail. 2020;7:1085–1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Birks EJ, George RS, Firouzi A, Wright G, Bahrami T, Yacoub MH and Khaghani A. Long-term outcomes of patients bridged to recovery versus patients bridged to transplantation. The Journal of thoracic and cardiovascular surgery. 2012;144:190–6. [DOI] [PubMed] [Google Scholar]
- 21.Basuray A, French B, Ky B, Vorovich E, Olt C, Sweitzer NK, Cappola TP and Fang JC. Heart failure with recovered ejection fraction: clinical description, biomarkers, and outcomes. Circulation. 2014;129:2380–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dandel M, Weng Y, Siniawski H, Potapov E, Krabatsch T, Lehmkuhl HB, Drews T, Knosalla C and Hetzer R. Pre-explant stability of unloading-promoted cardiac improvement predicts outcome after weaning from ventricular assist devices. Circulation. 2012;126:S9–19. [DOI] [PubMed] [Google Scholar]
- 23.Choi JH, Weber MP, Horan DP, Luc JGY, Phan K, Patel S, Rizvi SA, Maynes EJ, Reeves GR, Entwistle JW, et al. Left Ventricular Assist Device Decommissioning Compared with Explantation for Ventricular Recovery: A Systematic Review. ASAIO journal. 2020;66:17–22. [DOI] [PubMed] [Google Scholar]
- 24.Shah P, Birk SE, Cooper LB, Psotka MA, Kirklin JK, Barnett SD, Katugaha SB, Phillips S, Looby MM, Pagani FD, et al. Stroke and death risk in ventricular assist device patients varies by ISHLT infection category: An INTERMACS analysis. The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation. 2019;38:721–730. [DOI] [PubMed] [Google Scholar]
- 25.Padera R Commentary from Dr. Robert Padera regarding thrombus on the inflow cannula of the heartware HVAD: An update. Cardiovasc Pathol. 2020;48:107220. [DOI] [PubMed] [Google Scholar]
- 26.Strickland KC, Watkins JC, Couper GS, Givertz MM and Padera RF. Thrombus around the redesigned HeartWare HVAD inflow cannula: A pathologic case series. The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation. 2016;35:926–30. [DOI] [PubMed] [Google Scholar]
- 27.Ghodrati M, Maurer A, Schloglhofer T, Khienwad T, Zimpfer D, Beitzke D, Zonta F, Moscato F, Schima H and Aigner P. The influence of left ventricular assist device inflow cannula position on thrombosis risk. Artificial organs. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Sonntag SJ, Lipinski E, Neidlin M, Hugenroth K, Benkowski R, Motomura T and Kaufmann TAS. Virtual Fitting and Hemodynamic Simulation of the EVAHEART 2 Left Ventricular Assist Device and Double-Cuff Tipless Inflow Cannula. ASAIO journal. 2019;65:698–706. [DOI] [PubMed] [Google Scholar]
- 29.Potapov EV, Stepanenko A, Hennig E, Hetzer R and Krabatsch T. A titanium plug simplifies left ventricular assist device removal after myocardial recovery. The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation. 2010;29:1316–7. [DOI] [PubMed] [Google Scholar]
- 30.Sajjad M, Butt T, Oezalp F, Siddique A, Wrightson N, Crawford D, Pillay T and Schueler S. An alternative approach to explantation and exchange of the HeartWare left ventricular assist device. Eur J Cardiothorac Surg. 2013;43:1247–50. [DOI] [PubMed] [Google Scholar]
- 31.Mulzer J, Faerber G, Kaufmann F and Potapov EV. Recovery plug for HeartMate 3 left ventricular assist device. The Journal of thoracic and cardiovascular surgery. 2019;157:e35–e37. [DOI] [PubMed] [Google Scholar]
- 32.Gyoten T, Rojas SV, Fox H, Hata M, Deutsch MA, Schramm R, Gummert JF and Morshuis M. Cardiac recovery following left ventricular assist device therapy: experience of complete device explantation including ventricular patch plasty. Eur J Cardiothorac Surg. 2021;59:855–862. [DOI] [PubMed] [Google Scholar]
- 33.Zeigler SM, Sheikh AY, Lee PH, Desai J, Banerjee D, Oyer P, Dake MD and Ha RV. A novel, catheter-based approach to left ventricular assist device deactivation after myocardial recovery. The Annals of thoracic surgery. 2014;98:710–3. [DOI] [PubMed] [Google Scholar]
- 34.Schoenrath F, Kursawe L, Nersesian G, Kikhney J, Schmidt J, Barthel F, Kaufmann F, Knierim J, Knosalla C, Hennig F, et al. Fluorescence In Situ Hybridization and Polymerase Chain Reaction to Detect Infections in Patients With Left Ventricular Assist Devices. ASAIO journal. 2021;67:536–545. [DOI] [PubMed] [Google Scholar]
- 35.Qu Y, McGiffin D, Hayward C, McLean J, Duncan C, Robson D, Kure C, Shen R, Williams H, Mayo S, et al. Characterization of infected, explanted ventricular assist device drivelines: The role of biofilms and microgaps in the driveline tunnel. The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation. 2020. [DOI] [PubMed] [Google Scholar]
- 36.Puius YA, Parkar F, Tlamsa AP, Nistico JA, Muggia VA, Minamoto GY, Jorde UP, Goldstein DJ and Moadel RM. Gallium-67 Single-Photon Emission Computed Tomography Affects Management of Infections of Left Ventricular Assist Devices. ASAIO journal. 2020. [DOI] [PubMed] [Google Scholar]
- 37.Tam MC, Patel VN, Weinberg RL, Hulten EA, Aaronson KD, Pagani FD, Corbett JR and Murthy VL. Diagnostic Accuracy of FDG PET/CT in Suspected LVAD Infections: A Case Series, Systematic Review, and Meta-Analysis. JACC Cardiovasc Imaging. 2020;13:1191–1202. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.