If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Restrictive cardiomyopathy (RCM) patients have poor prognosis due to progressive heart failure characterized by impaired ventricular filling of either or both ventricles. The goal of this study was to evaluate the outcome of end-stage RCM patients after left ventricular assist device (LVAD) implantation and to determine factors that may be associated with improved survival.
This investigation is a retrospective study of prospectively collected data that include 28 consecutive patients with end-stage RCM who received continuous-flow LVADs at the Mayo Clinic, Rochester, Minnesota. Outcome was assessed by survival with LVAD support until heart transplantation or all-cause mortality.
The mean follow-up time post-LVAD implantation was 448 ± 425 days. The mean hospitalization time was 29 ± 19 days and was complicated mainly by post-operative right ventricular (RV) failure requiring short-term medical support. The short-term in-hospital mortality was 14%. Ten patients underwent heart transplantation with 100% survival post-transplant during the follow-up period. One-year survival for patients with LVADs without transplantation was 64%, and was not significantly different between amyloidosis and non-amyloidosis patients. Larger left ventricle (LV) end-diastolic and end-systolic dimensions were significantly associated with improved survival rates (RR = 0.94 and 0.95, p < 0.05, respectively), and left ventricular end-diastolic diameter (LVEDD) ≤46 mm was associated with increased mortality post-LVAD implantation.
LVAD is a feasible, life-saving therapy for end-stage heart failure related to RCM, especially as a bridge to transplant and in patients with larger LV dimensions.
Restrictive cardiomyopathy (RCM) constitutes a heterogeneous group of diseases with a common physiology, characterized by impaired ventricular filling, with normal or decreased diastolic volume of either or both ventricles. Regardless of etiology, the overall prognosis is poor and is associated with progression to heart failure and increased mortality.
The continuous-flow left ventricular assist device (LVAD) has become a standard therapeutic option to improve survival and quality of life in dilated cardiomyopathy patients with advanced heart failure, whether they are awaiting transplantation or as destination therapy (DT).
However, there are almost no data describing the impact of LVAD therapy on patients with end-stage RCM.
At our center, we have used LVAD therapy in selected RCM patients whose outcome would otherwise be very poor, or those we thought could be bridged to transplant (BTT). We previously described our experience with small numbers of patients and showed that LVAD implantation may be feasible with improved survival compared with medical therapy.
Thus, the goal of this study was to present our extended experience with a larger and more heterogeneous cohort, and to look for factors that may predict better outcome after LVAD implantation for end-stage RCM patients.
We identified all consecutive patients with end-stage heart failure and restrictive physiology (including all 8 patients presented in our previous report
) who received a continuous-flow LVAD (either a HeartMate II [Thoratec, Pleasanton, CA] or an HVAD [HeartWare, Inc., Framingham, MA]) for either DT or BTT between January 2008 and August 2013 at the Mayo Clinic, Rochester, Minnesota. Restrictive heart diseases included hypertrophic cardiomyopathy, infiltrative heart disease or chemotherapy/radiation-induced cardiomyopathy. Infiltrative heart disease and hypertrophic cardiomyopathy were defined based on endomyocardial biopsy findings and by the presence of echocardiographic characteristics of restrictive physiology. Chemotherapy/radiation-induced cardiomyopathy patients were included only if they had echocardiographic findings of smaller left ventricular (LV) dimensions and significant (Grade 3-4/4) diastolic dysfunction.
Clinical, demographic and hemodynamic data
Each patient’s pre-operative data (demographic, clinical, echocardiographic, hemodynamic and laboratory data) were abstracted from the medical records. Echocardiographic examinations were performed at 19 (range 1 to 101) days and hemodynamic catheterizations were performed at 14 (range 1 to 168) days before surgery. Echocardiographic parameters recorded included: LV ejection fraction (EF, calculated using the modified Simpson’s formula); LV end-diastolic diameter (LVEDD); LV end-systolic diameter (LVESD, measured with M-mode or 2-dimensional echocardiography); tricuspid valve regurgitation grade (based on jet characteristics and/or the proximal isovelocity surface area method); and right ventricle (RV) function evaluation (quantitative grading scale = normal, mild, mild–moderate, moderate, moderate–severe, severe dysfunction). Data concerning short-term post-operative adverse events (RV failure, inotropic support, renal failure, length of hospital stay and in-hospital mortality) and long-term LVAD-related complications (thrombotic events, hemolysis, bleeding and infection) were obtained by reviewing the medical records. The primary outcome was defined as all-cause mortality or heart transplantation. Survival information was obtained from the electronic medical records. The study protocol was reviewed and approved by institutional review board at the Mayo Clinic, Rochester, Minnesota.
Baseline data are presented as percent or mean (standard deviation). Comparisons of baseline characteristics between patients with and without amyloidosis were done by chi-square test for categorical variables and t-test for continuous variables. Survival, overall and by selected subgroups, was assessed using the Kaplan–Meier method and compared (when applicable) using the log-rank test. The bivariate association between selected baseline characteristics and subsequent outcome (i.e., death or transplantation) was evaluated using Cox proportional hazards modeling. Due to the small sample size and high correlation between variables, no multivariable models were fitted. Analyses were performed using SAS version 9.3 (SAS Institute, Inc., Cary, NC) and IBM SPSS version 19 (IBM SPSS, Inc., Chicago, IL).
The study cohort consisted of 28 patients (71% males) with end-stage RCM who underwent LVAD implantation (26 [93%] patients with a HeartMate II and 2 [7%] with an HVAD), either as BTT (61%) or DT (39%). The mean age at the time of implant was 57 ± 13 years. The etiology of restrictive cardiomyopathy was amyloidosis (36%), hypertrophic cardiomyopathy (28.5%), sarcoidosis (18%), chemotherapy/radiation-induced cardiomyopathy (14%) and Fabry disease (3.5%). All patients were severely symptomatic and classified as Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) Profile 1 (25%), 2 (68%) or 3 (7%) (Table 1).
Table 1Baseline Characteristics of Study Patients
Mean ± SD
Age at implant (years)
57 ± 13
26.9 ± 3.9
1.9 ± 0.2
Ejection fraction (%)
27 ± 16
53.7 ± 11.3
46.8 ± 12.3
24 ± 11
Deceleration time (ms)
153 ± 41
RVWSI (mm Hg/ml/m²)
0.45 ± 0.35
Right ventricle function:
Tricuspid regurgitation grade
Right ventricular systolic pressure (mm Hg)
46 ± 13
Mean right atrial pressure (mm Hg)
16 ± 6
Mean pulmonary arterial pressure (mm Hg)
33 ± 8
Mean wedge pressure (mm Hg)
23 ± 7
Cardiac index (liters/min/m²)
1.7 ± 0.4
Pulmonary arteriolar resistance
5.8 ± 3.9
11 ± 1.8
1.4 ± 0.6
4,930 ± 4,190
2 ± 1.2
Length of hospitalization (days)
24 ± 18
Intra-aortic balloon pump support before LVAD
Bridge to transplant
Tricuspid valve repair/replacement during surgery
BMI, body mass index; BSA, body surface area; LVAD, left ventricular assist device; LVEDD, left ventricle end-diastolic dimension; LVESD, left ventricle end-systolic dimension; NT-pro-BNP, N-terminal pro‒brain natriuretic protein; RCM, restrictive cardiomyopathy; RVSWI, right ventricular stroke work index.
Baseline echocardiographic examination showed the following mean values: LVEF 27 ± 16%; LVEDD 53.7 ± 11.4 mm; and LVESD 46.8 ± 12.3 mm. Most patients were considered to have at least moderate RV dysfunction and greater-than-mild tricuspid valve regurgitation (81% and 61%, respectively). Hemodynamic data showed right atrial pressure, capillary wedge pressure and mean pulmonary pressure to be elevated in most patients.
Short-term post-operative outcomes
The mean hospitalization time was 24 ± 18 days. The most common post-operative complication was RV failure (11 patients, 39%), presented as volume overload, renal failure, liver congestion and elevated right heart and pulmonary pressures. All these patients required inotropic support of the RV (isoproterenol or milrinone), and the mean duration of inotropic support post-surgery was 17 days. Two patients were discharged from the hospital with a continuous intravenous milrinone pump and were able to wean of inotropic support at 6 weeks post-implant. Four patients were treated with oral phosphodiesterase-5 inhibitor medications to lower pulmonary pressure during the hospitalization post-LVAD implant. Seven patients had severe renal failure that required renal replacement therapy post-surgery, 3 patients’ renal function recovered before discharge from the hospital, and 4 patients had fatal complications during the post-operative hospitalization. Only 1 patient, who was in cardiogenic shock before LVAD implant surgery, required right-sided VAD support for 5 days along with inotropic therapy, anti-pulmonary hypertension medications and dialysis, and was eventually discharged from the hospital (after 62 days) without any significant RV failure or need for renal replacement therapy.
Another common post-operative complication was bleeding in 7 patients (25%) (5 patients had gastrointestinal bleeding that was controlled before discharge and 2 patients had fatal endotracheal bleeding and hemothorax). Two patients had hemodynamically stable monomorphic ventricular tachycardia that was controlled with amiodarone therapy, and 3 other patients had rapid atrial tachycardia treated with β-blocker medications. Two patients had documented thrombotic events post-surgery: 1 patient had a fatal ischemic stroke and 1 was diagnosed with pulmonary emboli that did not require any intervention and was discharged from the hospital with anti-coagulation therapy.
Short-term in-hospital mortality was 14% (2 patients [with amyloid and chemotherapy/radiation-induced cardiomyopathy] developed complications secondary to lung bleeding, 1 amyloid patient had ischemic stroke, and 1 patient with chemotherapy/radiation-induced cardiomyopathy had septic shock). Eighteen patients (64%) were discharged from the hospital to a rehabilitation facility for a mean duration of 10 ± 5 days.
The mean follow-up time post-LVAD implantation was 448 ± 425 days; the median follow-up was 230 days. Among 24 patients who survived the short-term post-LVAD implant period, 13 (54%) had no LVAD-related complications during the follow-up period. The most common LVAD-related complication was bleeding: 7 patients (29%) had gastrointestinal bleeding and 1 had hemorrhagic stroke that was presumably related to supratherapeutic anti-coagulation therapy. Three patients with sarcoidosis RCM had thrombotic events: 1 had an ischemic stroke while recovering from a bleeding event with reduced anti-coagulation therapy, and 2 others had evidence of hemolysis due to suspected LVAD pump thrombosis. One of the patients was treated with continuous intravenous anti-coagulation therapy until successful heart transplant and the second was treated with a combination of anti-coagulation therapy and dual anti-platelet therapy (aspirin and clopidogrel) and remains stable while on the waiting list for heart transplantation. One patient had staphylococcal bacteremia shortly after discharge from the hospital and died within 90 days post-LVAD implant. None of the other patients had any significant LVAD or drive-line‒related infections during the follow-up period. None of the cohort patients had recovery of LV function, and no patient underwent pump exchange due to LVAD complications.
Ten patients underwent heart transplantation (4 hypertrophic cardiomyopathy, 3 sarcoidosis, 2 amyloidosis and 1 chemotherapy/radiation-induced). The mean follow-up time from LVAD implantation to heart transplantation was 472 (range 273 to 671) days, with 100% survival post-transplant during the follow-up period. Seven patients who received an LVAD as a BTT did not eventually proceed to transplantation: 5 patients remained clinically stable and their status on the transplant waiting list did not require an upgrade to a higher level, and 2 patients had fatal complication before transplantation.
Eighteen patients underwent LVAD implantation without heart transplantation. The mean and median survival times for this group of patients were 651 (range 358 to 945) days and 507 (range 174 to 840) days, respectively, with a 1-year survival of 64%. There was no significant difference in survival time with LVAD between the groups of patients with and without transplantation (p = 0.43) (Figure 1). Additional subgroup survival analysis showed no difference in patients who received an LVAD as a BTT or as DT (p = 0.76) (Figure 2).
Comparison between amyloidosis and non-amyloidosis patients
Among the RCM patients were 10 patients with cardiac amyloidosis. Nine patients had transthyretin amyloidosis (3 patients had familial type and 6 had senile wild-type amyloid) and did not receive any additional therapy for the disease. One patient had light-chain amyloid, which primarily involved the heart. He was treated with melphalan, lenalinomide and prednisone for 9 months before LVAD implantation, with complete hematologic remission (defined as negative immunofixation on serum and urine and normal free light-chain ratio).
We divided the restrictive cardiomyopathy patients into 2 groups according to amyloid (n = 10) vs non-amyloid (n = 18) etiology. The amyloid patients were significantly older at implant (67 ± 10 vs 52 ± 12 years, p < 0.05) and were predominantly male (10 [100%] vs 10 [56%], p < 0.05) and with smaller LVEDDs (46 ± 10 vs 58 ± 10 mm, p < 0.05). There was also a significant difference in the DT/BTT ratio (n = 8/2 [80%/20%] vs n = 3/15 (17%/83%), p < 0.05) (Table 2). Survival analysis showed no significant difference in mean survival time between the amyloid and non-amyloid groups (536 [range 213 to 859] vs 535 [range 317 to 752] days, p = 0.94) (Figure 3).
Table 2Baseline Characteristics for Amyloidosis vs Non-amyloidosis
Cox regression analysis was performed to look for an association between a list of clinical, echocardiographic and hemodynamic parameters and patient outcome. The analysis results reveal a significant association between LV end-diastolic and end-systolic diameter and survival rates, such that the larger the LV end-diastolic and end-systolic dimensions, the better the survival estimates (relative risk [RR] = 0.94, p < 0.05, and RR = 0.95, p < 0.05, respectively). There was no significant association between other parameters tested and mortality (Table 3).
Table 3Cox Regression—Association of Selected Variables With Mortality
Patients were then stratified into 2 groups based on LVEDD of ≤46 mm (n = 8) and LVEDD >46 mm (n = 20). By Kaplan–Meier analysis, the mean and median for survival times were significantly shorter among patients with LVEDDs ≤46 mm (112 [range 51 to 172] days vs 678 [range 475 to 882] days and 83 vs 625 days, respectively, p < 0.01) (Figure 4).
This is the largest study thus far to evaluate LVAD therapy in patients with end-stage RCM. The main study findings are: (1) LVAD implantation is technically feasible in patients with advanced RCM; (2) LVAD therapy is associated with improved survival compared with medical therapy and regardless of RCM etiology; and (3) LV dimensions are an important predictor for outcome and LVEDD <46 mm is significantly associated with increased mortality. These results have major implications for the approach to the population of end-stage RCM patients, and the findings support our previous observations on a smaller cohort of RCM patients showing that LVAD is a feasible and lifesaving therapy.
Cardiac transplantation is a widely accepted treatment to improve long-term survival in those with advanced symptoms; however, due to continued donor shortages, many of these patients face a long waiting period for transplant, and are prone to develop irreversible pulmonary hypertension and die from heart failure complications. Continuous-flow LVADs have been recognized to improve outcomes in patients with advanced heart failure who are failing maximal medical treatment either as BTT or DT; however, most LVAD trials have focused on subjects with end-stage dilated and ischemic cardiomyopathy, and patients with advanced restrictive cardiomyopathy were not represented in those trials.
RCM patients present unique challenges with regard to LVAD implantation. They may have persistent right heart failure likely due to primary involvement of the RV from the underlying cardiomyopathy process. There are also technical challenges with LVAD implantation in patients with thickened LV walls and small LV chamber sizes. These unique characteristics could make the recovery period after LVAD implantation challenging, with a high incidence of RV failure requiring short-term inotropic support and renal failure and volume overload necessitating renal replacement therapy.
Our results showing a 1-year survival of 64% demonstrate that LVAD implantation has a survival benefit compared with the reported natural history of end-stage RCM
Our findings are also comparable to the survival data published on end-stage dilated or ischemic cardiomyopathy patients with continuous-flow LVADs in the pivotal multicenter trials showing 68% 1-year survival in BTT and DT patients.
Although survival rates for LVAD patients have continued to improve with better patient selection and clinical experience, they remain not significantly different from our data. More recently, Jorde et al reported outcomes of the first 247 patients who underwent implantation of HeartMate II devices for DT after U.S. Food and Drug Administration approval, with Kaplan‒Meier survival at 12 months of 74% for all patients and 71% for patients with INTERMACS Profiles 1 to 3.
We observed that the survival benefit from LVAD was not affected by RCM etiology when comparing amyloidosis to non-amyloidosis patients. Cardiac amyloidosis is one of the forms of infiltrative RCM resulting from deposition of abnormal protein (amyloid) within the heart. This entity is increasingly being recognized, especially in the elderly heart failure population, and, as with other forms of RCM, carries a poor prognosis with few therapeutic options. Mechanical assist devices have not been used widely in patients with cardiac amyloidosis for the same reasons as those just described. From our previously published data on 9 cardiac amyloidosis patients after LVAD implantation it appears that the outcome can be varied.
The comparison of cardiac amyloid patients to other RCM etiologies suggests that these patients are less suitable for heart transplantation, either due to their advanced age or to significant co-morbidities that may be related to the systemic involvement of their disease. Nevertheless, our data show that survival rates were similar among amyloid and non-amyloid patients.
One of the most striking observations in our study is the excellent long-term survival of BTT patients who underwent heart transplantation. Two studies analyzed the United Network of Organ Sharing (UNOS) database of post-transplant outcomes in patients with RCM compared with all other patients with a non-restrictive phenotype.
Both studies demonstrated that patients with RCM, due to various etiologies, have comparable short- and long-term survival after cardiac transplantation to other patients with a non-restrictive phenotype. However, RCM transplant patients had worse survival when compared specifically with dilated cardiomyopathy patients. Among patients included in the RCM cohort, those with amyloidosis and radiotherapy/chemotherapy-induced cardiomyopathies had a significantly worse outcome than other RCM patients.
; in addition, given the decreased utility of inotropic therapy for these patients, prolonged hospitalization until transplantation is often required and there is a disproportionate risk of death while waiting,
For all of those reasons, it can be postulated that these patients are particularly vulnerable, transition to transplantation in a “sicker” condition, and thus may exhibit worse outcomes. In our study, 10 RCM patients with diverse etiologies underwent heart transplantation after LVAD implantation as a bridge therapy. Although this is a small cohort of patients and should be interpreted cautiously, the survival rate through post-transplant follow-up was 100%. These findings may suggest greater benefits from LVAD therapy as a BTT to improve post-transplant outcome among this group of patients.
Herein we have described for the first time parameters associated with better outcomes after LVAD implantation exclusively among RCM patients. Several studies have sought to identify risk factors associated with post-LVAD survival and showed associations between pre-operative renal failure,
and mortality. Those studies included all LVAD patients, most of whom were dilated or ischemic cardiomyopathy patients, but our data do not support any of their findings.
We observed an interesting association between greater baseline LVESD and LVEDD with better post-LVAD outcome. Several previous studies had reported associations between smaller LV and worse post-LVAD outcome.
Those studies evaluated mostly dilated cardiomyopathy patients and identified small LVEDD as being <62 to 64 mm. The RCM patients represent a unique group with particularly small LVs (mean LVEDD in our cohort was 53.7 ± 11.3 mm), and we found that an LVEDD of ≤46 mm was significantly associated with increased mortality.
Several possibilities may account for the association between smaller LV dimensions and poor outcome after LVAD implantation. First, these patients are at increased risk for “suck-down” events as the septum encroaches upon the inflow cannula, resulting in ventricular arrhythmias or drastically decreasing LVAD pre-load. Second, off-loading the small LV by the mechanical assist device may cause shifting of the interventricular septum to the left, which in turn impairs RV output and may exacerbate RV failure. There may be other possible explanations, specifically for patients with RCM; smaller LV dimensions may represent an advanced stage of the cardiomyopathy process that is less responsive to assist device support. Alternatively, smaller LV dimensions present technical challenges for LVAD inflow cannula implantation and may result in sub-optimal device placement and inadequate mechanical support.
Our results should encourage cardiologists to consider LVAD as a therapeutic option for end-stage RCM patients who would otherwise not survive, and especially for patients with larger LV dimensions who are eligible for heart transplantation.
We acknowledge some limitations to our study. First, this was an observational study with a retrospective analysis of prospectively collected data. Second, the study was conducted in a single tertiary medical center and there may have been patient selection bias. Thus, the data cannot necessarily be extrapolated to other centers. Third, the study cohort consisted of patients with a heterogeneous group of diseases with different etiologies resulting in end-stage heart failure. Nevertheless, regardless of etiology, all these patients shared a common physiology, characterized by impaired ventricular filling and diastolic dysfunction. Fourth, the total artificial heart device is another potential therapy as a BTT for these patients, yet we did not have an RCM cohort with this device to compare long-term outcomes between the 2 modalities. Fifth, the small patient cohort may lack the statistical power to identify other factors associated with better outcome or to identify significant differences in survival between amyloid and non-amyloid patients.
In conclusion, in this study we have assessed the largest cohort of end-stage heart failure patients with RCM who underwent LVAD implantation. We confirmed that LVAD is a feasible, lifesaving therapy, especially as a BTT. Larger LV dimensions may predict improved outcomes after implantation and allow for better patient selection, which, along with technological developments and further clinical implementation, will improve survival in RCM patients supported with LVADs.
The authors have no conflicts of interest to disclose.
Clinical profile and outcome of idiopathic restrictive cardiomyopathy.