Left Ventricular Reverse Remodeling With a Continuous Flow Left Ventricular Assist Device Measured by Left Ventricular End-Diastolic Dimensions and Severity of Mitral Regurgitation

Abstract

Pulsatile flow left ventricular assist devices (LVADs) maximally unload the left ventricle (LV), leading to reverse remodeling of the myopathic LV that manifests as decreased LV end-diastolic dimension (LVEDD) and decreased severity of mitral regurgitation (MR). There is a paucity of data, however, regarding the ability of continuous flow (CF) pumps to adequately decompress the LV to induce similar reverse remodeling. We sought to evaluate the effects of CF-LVADs on LV reverse remodeling. From March 2006 through July 2011, one hundred patients with chronic heart failure underwent implantation of CF-LVAD (93 HeartMate II LVADs and seven HeartWare LVADs) as bridge-to-transplant (n = 68) and destination therapies (n = 32). Echocardiograms and right heart catheterizations were reviewed preoperatively and at 1 and 6 months post-LVAD implantation. Mean age was 52.1 ± 12.1 years; etiology of heart failure was ischemic cardiomyopathy in 34 patients and nonischemic dilated cardiomyopathy in 66 patients. Median LVAD support time was 378.3 days; 371.5 days for patients who received bridge-to-transplant therapy and 422.2 days for patients who underwent destination therapy. Left ventricular end-diastolic dimension significantly decreased at 1 month post-LVAD implantation from 71.6 ± 12.4 to 58.3 ± 13.8 mm (p < 0.001). Severity of MR also significantly decreased from 76.0% of patients having moderate or severe MR preoperatively to 8.0% with moderate or severe MR at 1 month post-LVAD (p < 0.001). These reductions were maintained at 6 months. These data demonstrate the ability of a CF-LVAD to significantly decompress the LV, leading to significant reductions in LVEDD and severity of MR. This reverse remodeling was apparent in the early postoperative period and was sustained at 6 months. Further studies are warranted to investigate whether this correlates with clinical LV recovery.

Left ventricular assist devices (LVADs) have become the standard of care for select patients with refractory end-stage heart failure. Pump technology has improved, leading to a change from pulsatile flow (PF) to continuous flow (CF) pumps.^1,2 This has contributed to improvements in short- and long-term survival, decreased incidence of associated adverse events, such as bleeding, infection, and device malfunction, as well as enhanced quality of life.^3–5

The primary objective of LVAD therapy is to improve perfusion to vital organ systems. By pumping blood directly from the left ventricle (LV) to the aorta, these pumps significantly unload the LV, which is associated with myocardial recovery, normalization of gene expression, alterations in cytokine activation, and regression of interstitial fibrosis. This process is referred to as reverse remodeling.^6–11 Although CF pumps have been associated with improved clinical outcomes compared with PF pumps, there is a paucity of data regarding their ability to unload the LV to the same degree as PF pumps.^11–16 This may have important ramifications relating to the ability of CF pumps to induce a similar degree of reverse remodeling and myocardial recovery as that described for PF pumps.^6–11

In this study, we retrospectively analyzed our institutional experience with one hundred consecutive CF-LVADs. This included HeartMate II (HM II) LVAD (Thoratec Corp., Pleasanton, CA) recipients and HeartWare LVADs (HeartWare Inc., Framingham, MA) recipients of bridge-to-transplant (BTT) therapy and destination therapy (DT). Our primary objective was to determine the magnitude and durability of reverse remodeling seen with CF pumps, assessed by the decrease in LV end-diastolic dimensions (LVEDD) and degree of mitral regurgitation (MR) when comparing pre-LVAD with post-LVAD parameters.

Methods

From March 2006 through July 2011, one hundred patients with chronic heart failure underwent implantation of a CF-LVAD at our institution. This included 93 HM II LVADs and seven HeartWare LVADs. Patients were implanted as BTT therapy in 68 cases and as DT in 32 cases. Clinical records of these patients, including echocardiograms and right heart catheterizations were retrospectively reviewed to determine how LVEDD and degree of MR were affected by CF-LVAD support at 1 and 6 months post-LVAD implantation. The procedures followed were in accordance with institutional guidelines, and this review was performed with institutional review board approval.

Device Management

Device speed was clinically adjusted to optimize flow, peripheral perfusion, organ function, and LV decompression. Patients underwent periodic echocardiograms to evaluate the degree of LV decompression, aortic ejection, residual MR, position of the interventricular septum, right ventricular (RV) function, and severity of tricuspid regurgitation.

All patients were postoperatively anticoagulated on 81 mg aspirin daily and warfarin with a target international normalized ratio range of 1.8–2.5. Heart failure medications typically included a beta-blocker, ace inhibitor, and diuretics as well as sildenafil whether there was significant residual pulmonary hypertension.

Echocardiography

Preoperative echocardiograms were reviewed and compared with echocardiograms at 1 and 6 months postoperatively. All echocardiograms were performed by staff cardiologists at Henry Ford Hospital. Left ventricular end-diastolic dimensions and severity of MR based on a graded system were recorded for each patient.

Hemodynamic Variables

Hemodynamic variables were analyzed pre-LVAD implantation and compared with 1 and 6 month post-LVAD implantation values. These variables include central venous pressure (CVP), pulmonary artery pressures (PAP), pulmonary capillary wedge pressure (PCWP), and cardiac index (CI).

Statistical Analysis

Statistical analysis was conducted by a statistician from the Department of Biostatistics at Henry Ford Hospital. Data were represented as frequency distributions and percentages. Values of continuous variables were expressed as mean ± standard deviation. Continuous variables were compared using independent samples t-tests. Categorical variables were compared by means of χ² tests. For all analyses, a p < 0.05 was considered statistically significant. Kaplan-Meier analysis was used to calculate survival along with a log-rank p value when comparing groups. Actuarial survival at 1, 3, and 5 years postimplantation was calculated by constructing life tables. All data were analyzed using SPSS 11.5 software (SPSS Inc., Chicago, IL).

Results

Demographics

There were 73 male and 27 female patients with LVAD with a mean age of 52.8 ± 11.9 years. Etiology of heart failure was coronary artery disease (ischemic cardiomyopathy) in 34 patients and nonischemic dilated cardiomyopathy in 66 patients. Median LVAD support time was 378.3 days; 371.5 days for patients who received BTT therapy and 422.2 days for patients who underwent DT. Additional preoperative clinical characteristics of the patients are listed in Table 1.

Table 1: Baseline Clinical Characteristics of Patients With LVAD

Echocardiographic Data

Left ventricular end-diastolic dimensions significantly decreased at 1 month post-LVAD implantation from 71.6 ± 12.4 to 58.3 ± 13.8 mm (p < 0.001). The severity of MR was moderate or severe in 76.0% of patients preoperatively. At 1 month, only 8.0% had moderate or severe MR (p < 0.001), and this was maintained at 6 months post-LVAD implantation (Table 2).

Table 2: LVEDD and Severity of MR Pre-LVAD Compared With 1 Month and 6 Months Post-LVAD

Hemodynamic Data

At 1 month post-LVAD implantation, CVP decreased from 11.7 ± 6.2 to 9.1 ± 4.7 mm Hg (p < 0.001), systolic PAP decreased from 51.7 ± 15.3 to 35.8 ± 12.4 mm Hg (p < 0.001), and PCWP decreased from 23.0 ± 10.2 to 12.4 ± 8.1 mm Hg (p < 0.001). Additionally, CI increased from 1.9 ± 0.5 to 2.5 ± 0.6 L/min/m² (p < 0.001). Transpulmonary gradient did not change significantly (p = nonsignificant).

Discussion

Despite significant improvements in pump design and better short- and long-term outcomes with CF pumps, LVAD therapy remains confined to patients with advanced, end-stage heart failure and with severe functional limitations. Extension of this therapy to patients at earlier stages of disease is dependent on the continued improvement in long-term outcomes as well as the potential for significant myocardial recovery with the option for device explantation. Left ventricular reverse remodeling is crucial to the concept of myocardial recovery. The clinical and pathologic sequelae of LV remodeling in heart failure are well defined, and various interventions have been tried to slow or reverse the progression of the remodeling process. Passive constraint devices including the Acorn CorCap and the Myosplint (Myocor, Inc., Maple Grove, MN) have had experimental success in inducing LV reverse remodeling in patients with heart failure but have not led to significant improvements in clinical outcomes.^17,18 Left ventricular assist device therapy, on the other hand, has significantly improved survival and quality of life in patients with end-stage heart failure. However, data concerning LV reverse remodeling with CF device therapy are sparse. The ultimate goal of LV recovery, namely device explantation, remains a rare clinical event. Understanding LV reverse remodeling is an important precursor in being able to promote this recovery process.

Reductions in CVP and PAP with concomitant RV unloading are important goals of LVAD therapy because sustained RV function is crucial to good long-term outcomes. Our patients manifested considerable unloading of the RV demonstrated by significant reductions in CVP and PAP at 1 and 6 months postimplantation. There is potential for these devices to induce RV dysfunction by causing leftward shifting of the interventricular septum.¹⁹ Our practice is therefore to obtain serial echocardiograms during the initial days and weeks of support to determine the LVAD speed and ventricular loading conditions that will maximally decompress the LV without excessively shifting the interventricular septum and exacerbating RV dysfunction. These studies are performed at rest, although there may be additional benefit to performing them with activity as well. Device speed is increased until there is optimal decompression of the LV with no bowing of the interventricular septum and either no residual or trivial MR. Additionally, we aim to have the aortic valve open several times per minute.

In conclusion, our results demonstrated that CF-LVAD support induced immediate and sustained reverse remodeling of the LV measured by significant reductions in LVEDD and severity of MR. Our data are consistent with similar studies with PF pumps and confirm the findings of smaller series with CF pumps.^11–16,20 Given the substantial and consistent reduction in the severity of MR post-LVAD implantation, we do not believe that concomitant mitral valve repair is necessary during the initial operation. To date, we have not performed any concomitant mitral valve procedures on patients undergoing LVAD implantation. Residual MR can generally be reduced by increasing the revolutions per minute, which yield greater LV decompression. For the rare patient with symptomatic, refractory, severe residual MR, mitral valve repair may be warranted.

The engineering paradigm of CF will likely be the platform for future LVAD device development because of the associated advantages of CF pumps, namely improved survival, quality of life, and decreased pump-related adverse events. Although we have observed LV recovery leading to prolonged device explantation in only one of these patients, a more detailed examination of the reverse remodeling process, such as analysis of myocyte size, regression of fibrosis, cellular hypertrophy, collagen content, gene expression, and various biomarkers of recovery, such as cardiac tumor necrosis factor-alpha, will extend our understanding of the remodeling process beyond just volume unloading of the LV, and may allow for more focused attempts at long-term LV recovery. Additionally, it is possible that partial unloading of the LV may induce substantial reverse remodeling and promote myocardial recovery, although this hypothesis requires further preclinical and clinical analyses.

Study Limitations

This is a retrospective review, and limitations include potential inaccuracy of data retrieved from medical records. Additionally, the number of patients in this study was relatively small, thus limiting the statistical power of the analysis and conclusions. Further studies with more patients and longer follow-up will be useful.