The Effect of Cardiac Resynchronization on Morbidity and Mortality in Heart Failure (CARE-HF)

Authors, Journal, Affiliations, Type, DOI

• Authors: John G.F. Cleland, Jean-Claude Daubert, Erland Erdmann, Nick Freemantle, Daniel Gras, Lukas Kappenberger, Luigi Tavazzi; for the CARE-HF Study Investigators
• Journal: New England Journal of Medicine, 2005;352:1539–1549
• Published online: March 7, 2004; print April 14, 2005
• Type: Multicenter, international, randomised, open-label controlled trial
• Funding: Medtronic (device manufacturer); Medtronic provided study manager and commented on manuscript pre-submission; had no access to database; sponsor did not participate in analysis or writing
• DOI: 10.1056/NEJMoa050496
• Sites: 82 European centres; enrolled January 2001 – March 2003

Overview

CARE-HF randomised 813 patients with NYHA class III/IV, LVEF ≤35%, and cardiac dyssynchrony (QRS ≥120 ms with echocardiographic dyssynchrony criteria for QRS 120–149 ms) to CRT-P (biventricular pacing, no defibrillator) or medical therapy. Over a mean 29.4 months, CRT-P substantially reduced the primary composite of death or major CV hospitalisation (HR 0.63; P<0.001) and — critically — reduced all-cause mortality (HR 0.64; P<0.002; 20% vs 30%). This is the definitive CRT mortality trial: prior COMPANION showed a CRT-P mortality trend (P=0.06) at 16 months; CARE-HF confirmed a significant mortality benefit at 29.4 months. CRT also reversed LV remodeling, reduced mitral regurgitation, and improved QoL. CRT-P was used without an ICD — 7% of CRT patients still died suddenly.

Keywords

Cardiac resynchronization therapy, CRT-P, biventricular pacing, cardiac dyssynchrony, QRS duration, LBBB, heart failure, CARE-HF, all-cause mortality, reverse remodeling, interventricular mechanical delay

Key Takeaways

Background

• Despite pharmacological advances, many NYHA III/IV HF patients had persistent severe symptoms and poor prognosis
• Cardiac dyssynchrony (delayed regional activation) was common and associated with worse outcomes
• Prior short-term trials showed CRT improved symptoms, exercise capacity, QoL, and ventricular function
• COMPANION trial (NEJM 2004) showed CRT-P reduced the composite of death/hospitalisation (P<0.001) but all-cause mortality trend with CRT-P was not significant (P=0.06) at 16 months; CARE-HF was designed to definitively test mortality

Population

• n=813; 404 medical therapy alone, 409 CRT (+ medical therapy)
• Age ≥18; NYHA class III or IV for ≥6 weeks despite standard pharmacological therapy
• LVEF ≤35%
• LV end-diastolic dimension ≥30 mm indexed to height
• QRS ≥120 ms on ECG; for QRS 120–149 ms: required ≥2 of 3 additional dyssynchrony criteria: aortic pre-ejection delay >140 ms, interventricular mechanical delay >40 ms, or delayed posterolateral LV activation
• Excluded: AF or atrial flutter (cannot benefit from atrial component of resynchronisation); major CV event within 6 weeks; conventional pacemaker or ICD indication; HF requiring continuous IV therapy
• Background at baseline: ~85% beta-blocker (at some point during study); 43% high-dose diuretics; pre-ARNi, pre-SGLT2i era
• Mean follow-up 29.4 months (range 18.0–44.7)

CRT Device

• CRT-P only — no defibrillator function
• Medtronic InSync or InSync III device; biventricular pacing via standard RV + Attain coronary sinus LV lead (lateral/posterolateral LV wall)
• AV delay echocardiographically optimised; interventricular delay set to zero
• Implantation attempted in 404/409 assigned patients; successful implantation and activation in 390 (95%); median 4 days post-randomisation

Primary Endpoint — Death or Unplanned CV Hospitalisation

• Composite of death from any cause or unplanned hospitalisation for a major cardiovascular event (first event counted)
• CRT 39% (159/409) vs medical therapy 55% (224/404)
• HR 0.63 (95% CI 0.51–0.77; P<0.001)
• Unplanned CV hospitalisations: 222 (CRT) vs 384 (medical)
• No heterogeneity across prespecified subgroups (ischaemic vs non-ischaemic, NYHA class)

All-Cause Mortality (Principal Secondary Endpoint)

• CRT: 82 deaths (20%) vs medical: 120 deaths (30%)
• HR 0.64 (95% CI 0.48–0.85; P<0.002)
• 1-year mortality: 9.7% CRT vs 12.6% medical; 2-year: 18.0% CRT vs 25.1% medical
• NNT ≈ 9 (9 devices implanted to prevent 1 death and 3 HF hospitalisations)
• Mode of death: CV 83%, non-CV 17%; worsening HF 40% (CRT) vs 47% (medical); sudden death 35% (CRT) = 29 patients (7% of CRT group) vs 32% (medical) = 38 patients — residual SCD despite CRT

Other Secondary Endpoints

• Composite of death or HF hospitalisation: HR 0.54 (95% CI 0.43–0.68; P<0.001)
• HF hospitalisations: 18% of CRT patients (72) vs 33% of medical patients (133)
• NYHA class at 90 days: significantly improved (P<0.001); at 18 months: I/II/III–IV = 105/150/80 (CRT) vs 39/112/152 (medical)
• QoL at 90 days: Minnesota Living with Heart Failure score improved P<0.001; EuroQoL EQ-5D improved P<0.001
• Benefit consistent across ischaemic and non-ischaemic aetiology

Structural and Biomarker Effects (Echocardiography + NT-proBNP Core Laboratories)

At both 3 months and 18 months, CRT group showed:

• ↑ LVEF (significant vs medical)
• ↓ LV end-systolic volume index
• ↓ Mitral regurgitation area (colour-flow Doppler jet/LA area ratio)
• ↓ Interventricular mechanical delay
• ↓ NT-proBNP (by 18 months)
• ↑ Systolic blood pressure

Serious Adverse Events

• Lead displacement: 24 patients
• Coronary sinus dissection: 10
• Pocket erosion: 8
• Pneumothorax: 6
• Device-related infection: 3
• Atrial arrhythmias/ectopy: 64 (CRT) vs 41 (medical; P=0.02) — more frequent in CRT group
• Ventricular arrhythmias, renal dysfunction, hypotension, neurologic events: similar between groups
• Worsening heart failure as adverse event: 191 (CRT) vs 263 (medical; P<0.001) — less in CRT

Limitations of the Document

• Open-label (not blinded) — investigators and patients aware of treatment assignment; potential performance and ascertainment bias
• CRT-P only, no ICD: 29 patients (7%) in CRT group still died suddenly — CRT-D would likely prevent additional deaths; the CARE-HF mortality result may underestimate benefit of CRT-D
• NYHA III/IV only: benefit in NYHA I/II established later (MADIT-CRT 2009, REVERSE) — CARE-HF does not support Class I CRT for NYHA II
• QRS ≥120 ms, not restricted to LBBB: modern evidence suggests non-LBBB benefit is attenuated; current Class I guidelines require LBBB; the CARE-HF dyssynchrony criteria (2/3 echo criteria for QRS 120–149 ms) did not map directly to contemporary LBBB morphology classification
• AF excluded — not generalisable to AF population (a common HF comorbidity)
• Pre-ARNi/SGLT2i era: background therapy did not include sacubitril-valsartan or SGLT2i; modern quadruple GDMT independently promotes reverse LV remodeling, potentially reducing dyssynchrony substrate and narrowing the absolute CRT benefit
• Industry-sponsored (Medtronic); Medtronic provided study manager and commented on manuscript
• Hospitalizations within 10 days of randomisation excluded from primary endpoint — may slightly undercount events in both groups
• Small crossover: 19 patients in medical group received activated CRT before primary endpoint (5%); additional 31 after endpoint

Key Concepts Mentioned

• concepts/Cardiac-Dyssynchrony — mechanistic entry criterion; interventricular mechanical delay target
• entities/CRT — device and therapy described

Key Entities Mentioned

• entities/CRT — CRT-P device; primary subject of trial
• entities/HFrEF — population; management context; device therapy section

Wiki Pages Updated

• Created wiki/sources/crt-carehf-nejm-2005.md
• Updated wiki/entities/CRT.md — CARE-HF added as foundational mortality trial; source_count 1→2
• Updated wiki/entities/HFrEF.md — CRT section enriched with CARE-HF data; contradiction added
• Updated wiki/sourceindex.md
• Updated wiki/wikiindex.md