Reviews/Heart Failure Management/Cardiac resynchronisation therapy with defibrillator and with pacemaker have similar all-cause mortality
CRT · HFM-CRT-02 · v2
The claim says that in non-ischaemic heart failure, adding a defibrillator to resynchronisation therapy makes no difference to how long patients live. The two analyses in this corpus that actually look only at non-ischaemic patients both found the defibrillator version doing better, not the same: a COMPANION analysis reported a reduced risk of death for CRT-D in this group [3], and a 50,084-patient national registry found the same direction in its non-ischaemic subgroup [2]. The studies that reported no difference [1], [12], [13] all mixed ischaemic and non-ischaemic patients together and never reported a non-ischaemic-only result, so they do not speak directly to this claim. The one trial run purely in non-ischaemic patients [6] tested a defibrillator against usual care rather than comparing the two device types, so it cannot settle the head-to-head question either. No randomised trial directly comparing CRT-D with CRT-P exists in this corpus. The panel therefore judged the no-difference claim probably not true as stated, while noting the mixed and indirect data leave it short of being ruled out.
The only two analyses that isolate the non-ischaemic subgroup both find CRT-D superior for all-cause mortality: [3] reports an adjusted HR of 0.54 (95% CI 0.34 to 0.86) in NICM patients (n=555), and [2] reports an adjusted HR of 0.82 (95% CI 0.74-0.91) in its subgroup without ischaemic heart disease out of 50,084 implantations. The estimates compatible with no difference all come from populations mixing ischaemic and non-ischaemic patients: [1] adjusted HR 0.99 (95% CI 0.79-1.22) in 1775 late survivors, [12] entropy-balanced HR 0.99 (95% CI 0.81-1.20), and [13] propensity-matched HR 0.95 (95% CI 0.77 to 1.16). The dedicated non-ischaemic randomised trial [6] found no significant all-cause mortality benefit (HR 0.87, 95% CI 0.68 to 1.12, P=0.28) but randomised ICD against usual care, with 58% of patients in both arms receiving CRT. No trial in this corpus randomises CRT-D against CRT-P directly. The contested point: The division is one of population scope: whether 'no difference' estimates drawn from mixed ischaemic and non-ischaemic cohorts ([1], [12], [13]) can be read onto a strictly non-ischaemic claim. When the non-ischaemic subgroup is isolated ([2], [3]), the signal reverses toward a CRT-D advantage, and [3]'s significant device-etiology interaction (pinteraction=0.015) implies the two etiologies should not be pooled.
All 18 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Heart Failure Management, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The claim is specific to non-ischaemic HF, yet the two analyses that actually isolate that subgroup both show CRT-D significantly superior ([3] HR 0.54, 0.34-0.86; [2] aHR 0.82, 0.74-0.91) — the opposite of 'no significant difference'.
The null-supporting estimates ([1] HR 0.99, [12] HR 0.99, [13] HR 0.95) all come from mixed ischaemic/non-ischaemic populations and report no non-ischaemic-only result; applying them to a non-ischaemic claim is a scope error.
Where it remains uncertain. The null estimates ([1] HR 0.99, [12] HR 0.99, [13] HR 0.95) come only from mixed-etiology cohorts, and no randomised CRT-D versus CRT-P comparison exists in the corpus to adjudicate the confounding-by-indication concern.
Aleatory
The direct non-ischaemic estimate rests on 555 randomised NICM patients in [3], a sample small enough that the point estimate carries wide sampling variability around the true effect.
Epistemic
The corpus contains no trial randomising CRT-D against CRT-P; [4] notes no such trial has been designed, so the device-type comparison is unobserved under randomisation.
The load-bearing papers from a corpus of 18, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Supports no difference, but in a mixed ischaemic/non-ischaemic late-survivor cohort — not non-ischaemic-specific.
“The adjusted hazard ratio (HR) for all-cause mortality was 0.99 (95% CI 0.79-1.22).”
Panel auditInternal validity: medium · scope of inference: Long-term (beyond 5 years) survival of CRT recipients, mixed etiology, among those already alive at 5 years.Key limitations: observational · mixed etiology not non-ischaemic-specific · survivor selection · CI crosses 1
Dedicated non-ischaemic RCT; adding a defibrillator gave no all-cause mortality benefit — indirect support, but tests ICD-vs-control, not CRT-D vs CRT-P.
“the primary outcome had occurred in 120 patients (21.6%) in the ICD group and in 131 patients (23.4%) in the control group (hazard ratio, 0.87; 95% confidence interval [CI], 0.68 to 1.12; P=0.28).”
Panel auditInternal validity: high · scope of inference: Prophylactic ICD versus usual care in non-ischaemic systolic heart failure.Key limitations: contrast is ICD vs control not CRT-D vs CRT-P · CRT received by only 58% · not a direct device-type comparison
Supports CRT-P non-inferiority, but mixed-etiology, so does not isolate the claim's population.
“After adjustment for age, HR was 1.13 (95% CI: 0.95-1.35) and after entropy balancing 0.99 (95% CI: 0.81-1.20).”
Panel auditInternal validity: medium · scope of inference: Contemporary CRT recipients without secondary-prevention ICD indication; survival by device type.Key limitations: observational/claims data · not stratified by non-ischaemic etiology · residual confounding
Large registry; the non-ischaemic subgroup (aHR 0.82, CI excludes 1.0) also favours CRT-D, contradicting the claim.
“excess mortality (analogue of RS) was lower after CRT-D than after CRT-P in all patients [adjusted hazard ratio (aHR) 0.80, 95% confidence interval (CI) 0.76-0.84] as well as in subgroups with (aHR 0.79, 95% CI 0.74-0.84) or without (aHR 0.82, 95% CI 0.74-0.91) ischaemic heart disease.”
Panel auditInternal validity: medium · scope of inference: Nationwide real-world CRT recipients, relative survival by device type including a non-ischaemic subgroup.Key limitations: observational/administrative · relative survival not crude ACM · residual confounding by indication
4 of 18 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
With no head-to-head RCT and the two non-ischaemic-specific analyses both favouring CRT-D, the null claim is probably not true as stated, though mixed-cohort and indirect data keep it from being fully refuted.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
No randomised CRT-D vs CRT-P trial in non-ischaemic HF. Every estimate bearing on the claim is observational or a non-randomised subgroup, so confounding by indication cannot be excluded in either direction; [4] confirms no such trial has been designed. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A randomised trial allocating non-ischaemic CRT-eligible patients to CRT-D or CRT-P with all-cause mortality as the primary endpoint.
Open question: In non-ischaemic heart failure, does CRT-D reduce all-cause mortality relative to CRT-P when the two devices are directly randomised? Why it is still open: The corpus contains no head-to-head randomised comparison; [2] and [3] favour CRT-D but are a registry and a non-randomised subgroup, while [6] tests ICD against usual care rather than device type.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19