REMATCH

Published

Rose EA, et al. "Long-term use of a left ventricular assist device for end-stage heart failure". The New England Journal of Medicine. 2001. 345(20):1435-43.
PubMed • Full text • PDF

Clinical Question

In patients with end-stage heart failure, does destination LVAD therapy reduce all-cause mortality, as compared to optimal medical therapy?

Bottom Line

In patients with end-stage heart failure, destination left ventricular assist device (LVAD) therapy reduces all-cause mortality, as compared to optimal medical therapy.

Major Points

Left ventricular assist devices (LVAD) therapy was first developed as a short-term mechanical circulatory support for patients with end-stage heart failure as a bridge to heart transplant. The Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) study aimed to determine the suitability of LVAD for destination therapy.

In REMATCH, pulsatile-flow LVAD (HeartMate XVE) was compared to optimal medical therapy (OMT), with the primary endpoint being all-cause mortality. Patients with advanced heart failure (NYHA class IV, LVEF ≤25%) who were ineligible for cardiac transplantation were randomized to receive an LVAD (n=68) or continue OMT (n=61). The trial showed that LVAD implantation was associated with a 48% reduction in mortality as compared with optimal OMT (RR 0.52, 95 % CI 0.34-0.78, P=0.001). The survival at 1 year in the LVAD group was 52% as compared to 25% in OMT (P=0.002; NNT=4). However, the difference was diminished at 2 years (23% and 8% in LVAD and OMT group, respectively; P=0.09). It is also important to note that the incidence of serious adverse events was higher in the LVAD group than in the OMT group (rate ratio 2.35).

The 2007 INTREPID and 2009 HEARTMATE II studies demonstrated improvements in survival among patients with inotrope-dependent heart failure (INTREPID) and with newer devices (HEARTMATE II). Taken together, these and other studies support the use of LVAD as destination therapy in some patients with end-stage heart failure.

Guidelines

AHA/ACCF Heart Failure Guidelines (2013, adapted)

• Mechanical circulatory support (MCS) can be considered in selected patients with stage D HFrEF with planned definitive management (eg, cardiac transplantation) or cardiac recovery planned. (Class IIa, Level B)
• Nondurable MCS is reasonable as a “bridge to recovery” or “bridge to decision” for selected patients with HFrEF with acute, profound hemodynamic compromise. (Class IIa, Level B)
• Durable MCS is reasonable to prolong survival for selected patients with stage D HFrEF(672–675) (Class IIa, Level B)
  • Selection criteria for patients include: LVEF<25% and NYHA class III-IV despite optimal medical therapy, and either high predicted 1- to 2-year mortality (eg, markedly reduced peak oxygen consumption and clinical prognostic scores) or dependence on continuous parenteral inotropic support

Design

• Multicenter, prospective, randomized, controlled trial
• N=129 patients with end-stage HF ineligible for heart transplant
  • LVAD (n=68)
  • OMT (n=61)
• Setting: 20 US centers
• Enrollment: 1998-2001
• Follow-up: 2 years
• Analysis: intention-to-treat
• Primary outcome: All-cause mortality

Population

Some criteria are presented elsewhere.

Inclusion Criteria

Initial criteria:

• Age >18 years
• NYHA class IV heart failure symptoms for 90 days despite therapy with ACE inhibitors, diuretics, and digoxin
• Left ventricular EF ≤25%
• Peak oxygen consumption ≤12 ml/kg/min
• Persistent requirement for intravenous inotrope due to symptomatic hypotension, worsening renal function, or pulmonary edema
• Ineligible for cardiac transplant due to one of the following reasons:
  • Age >65 years
  • Diabetes mellitus requiring insulin and complicated by end-organ damage
  • CKD with a serum creatinine >2.5 mg/dL (221 μmol/L) for ≥90 days prior to randomization
  • Other significant conditions

The criteria were broadened 18 months into enrollment to include:

• NYHA class IV heart failure symptoms for 60 days
• Peak oxygen consumption ≤14 ml/kg/min
• NYHA class III or IV heart failure for ≥28 days and received ≥14 days of support with an intraaortic balloon pump or IV inotropes

Exclusion Criteria

• Heart failure due to or associated with uncorrected thyroid disease, obstructive cardiomyopathy, pericardial disease, amyloidosis, or active myocarditis
• High surgical risk
• INR ≥1.3 or prothrombin time >15 seconds within 24 hours prior to randomization
• Body surface area < 1.5 m²
• BMI >40 kg/m²
• COPD (FEV ≤ 1.5 L/min)
• Pregnancy
• Fixed pulmonary hypertension with pulmonary vascular resistance ≥ 8 Wood units refractory to pharmacologic intervention, within 90 days prior to randomization
• Under consideration for revascularization, therapeutic valvular repair, left ventricular reduction procedure, or cardiomyoplasty
• Cardiac transplantation, left ventricular reduction procedure, or cardiomyoplasty
• Mechanical aortic valve at the time of LVAD implant
• Other ongoing mechanical circulatory support than IABP
• AST, ALT or total bilirubin >5x upper limit of normal, or liver cirrhosis (proved by biopsy)
• Stroke ≤90 days prior to enrollment, cerebrovascular disease with significant (>80%) extracranial stenosis
• Impairment of cognitive function
• Untreated aortic aneurysm ≥5 cm 30 days prior to randomization
• Systemic infection 48 hours prior to randomization
• Platelet ≤50,000 mm³  24 hours prior to randomization
• Serum creatinine ≥3.5 mg/dL or chronic renal replacement therapy
• Severe peripheral vascular disease
• Use of calcium channel blocker (except amlodipine), or a class I or III antiarrhythmic ≤28 days prior to enrollment
• Planned abdominal surgery
• Psychiatric disease that is likely to impair with compliance to study protocol
• Survival <3 years due to a condition other than heart failure
• Participating in another clinical study
• Receiving an investigational therapy

Baseline Characteristics

From the LVAD group (n=68)

• Demographics: age 66±9.1 years, male 78%
• Clinical measurements: systolic BP 101±15 mm Hg, diastolic BP 61±10 mm Hg; heart rate 84±16 beats/min
• Cardiac measurements: LVEF 17±5.2%, PCWP 25±9.9 mm Hg, cardiac index 1.9±0.99 L/min/m² of BSA, pulmonary vascular resistance 3.4±1.8 Wood units, CVP 13±6 mm Hg
• Medical history: Heart failure due to ischemia 78 %, stroke 16%
• Treatment: ACE inhibitor 62%, ARB 10%, beta-blocker 24%, loop diuretic 96%, spironolactone 34%, IV inotrope 65%, digoxin 87%, amiodarone 45%
• Laboratory results: creatinine: 1.7±0.65 mg/dL, sodium 135.7±5.4 mmol/L
• Quality of life: Minnesota Living with Heart Failure score 75±18; SF-36-Physical function 19±19, emotional role 33±42; Beck Depression Inventory 19±9

Interventions

• Patients were randomized in a 1:1 ratio to one of two groups:
• Pulsatile-flow HeartMate XVE (Thoratec)
  • Device volume 450 ml, weight 1250 grams
  • The device was implanted in either the preperitoneal pocket or intraperitoneal cavity, depending on the surgeon's preference
• Optimal medical therapy
  • Treatment with digoxin, diuretics, and ACE-inhibitor (ARB if intolerant)
  • Beta-blocker treatment is dependent on the investigator's discretion

Outcomes

Outcomes are presented as LVAD vs. OMT.

Primary Outcome

All-cause mortality

48% reduction in the LVAD group (RR 0.52; 95% CI 0.34-0.78; P=0.001)

Survival at 1 year: 52% vs. 25% (P=0.002; NNT=4)

Survival at 2 years: 23% vs. 8% (P=0.09)

Secondary Outcomes

All serious adverse events

6.45 vs. 2.75 rate/patient-year (rate ratio 2.35, 95% CI 1.86-2.95)

Serious neurological AEs

Includes stroke, TIA, and toxic, or metabolic encephalopathy.

0.39 vs. 0.09 rate/pt-yr (rate ratio 4.35, 95% CI 1.31-14.50)

Bleeding (non-neurologic)

0.56 vs. 0.06 rate/pt-yr (rate ratio 9.47, 95% CI 2.30-38.90)

Infection

Local: 0.39 vs. 0.24 rate/pt-yr (rate ratio 1.63, 95% CI 0.72-3.70)

Sepsis: 0.6 vs. 0.3 rate/pt-yr (rate ratio 2.03, 95% CI 0.99-4.13)

Other organ dysfunction

Renal failure: 0.25 vs. 0.18 rate/pt-yr (rate ratio 1.42, 95% CI 0.54-3.71)

Hepatic failure: 0.02 vs. 0 rate/pt-yr

Cardiovascular

Cardiac arrest: 0.12 vs. 0.18 rate/pt-yr (rate ratio 0.65, 95% CI 0.21-2.00)

Supraventricular arrhythmia: 0.12 vs. 0.03 rate/pt-yr (rate ratio 3.92, 95% CI 0.47-32.40)

Ventricular arrhythmia: 0.25 vs. 0.56 rate/pt-yr (rate ratio 0.45, 95% CI 0.22-0.9)

Non-perioperative myocardial infarction: 0.02 vs. 0.03 rate/pt-yr (rate ratio 0.65, 95% CI 0.04-10.30)

Peripheral thromboembolism: 0.14 vs. 0.06 rate/pt-yr (rate ratio 2.29, 95% CI 0.48-10.80)

Syncope: 0.04 vs. 0.03 rate/pt-yr (rate ratio 1.31, 95% CI 0.12-14.40)

LVAD-related events

Suspected malfunction: 0.75 rate/pt-yr

LVAD failure: 0.08 rate/pt-yr

Perioperative bleeding: 0.46 rate/pt-yr

Drive-line tract or pocket infection: 0.41 rate/pt-yr

Pump interior, inflow, or outflow tract infection: 0.23 rate/pt-yr

Right heart failure: 0.17 rate/pt-yr

Thrombosis in LVAD: 0.06 rate/pt-yr

Perioperative MI: 0 rate/pt-yr

Days spent in the hospital

88 vs. 24 days (P-value not reported)

Quality of life and functional status at 1-year, assessed by

Minnesota Living with Heart Failure questionnaire: 41±22 vs. 58±21 (P=0.11)

SF-36 physical function: 46±19 vs. 21±21 (P=0.01)

SF-36 emotional role: 64±45 vs. 17±28 (P=0.03)

Beck Depression Inventory: 8±7 vs. 13±7 (P=0.04)

Other outcomes

Causes of death

There were 41 and 54 deaths in the LVAD and OMT group, respectively

In the LVAD group, the most common cause (41.5%) was sepsis. Other causes include LVAD failure (17.1%), cerebrovascular disease (9.8%), and pulmonary embolism (4.9%).

In the OMT group, the most common cause (92.6%) was left ventricular dysfunction.

Subgroup analysis

The trial was not powered for subgroup analyses, but the following prespecified subgroup analyses of the primary outcome were reported.

Age

18-59 years: No significant difference

60-69 years: Favors LVAD (relative risk 0.49; 95% CI, 0.25-0.95)

≥70 years: No significant difference

Criticisms

• Low survival at 2 years after LVAD. Therefore the effect of LVAD on quality of life needs to be carefully measured as it is an important factor supporting the use of mechanical circulatory support.
• The cost for LVAD therapy needs to be considered before deciding that it is a suitable alternative to medical therapy.
• The causes of death need to be considered. While 93% of deaths in the OMT group were due to heart failure, common causes of death in the LVAD group included sepsis, pulmonary embolism, and bleeding all of which may occur immediately post-operation.

Funding

• The National Heart, Lung, and Blood Institute
• Thoratec