CRT Outcomes Across Clinical Contexts

Evidence

Cardiac resynchronization therapy (CRT) aims to coordinate ventricular contraction in selected people with systolic heart failure and delayed electrical activation. Two recent retrospective studies examined why outcomes can still differ substantially among recipients. One measured a specific electrical interval before and after implantation. The other compared people implanted during a heart-failure hospitalization with those whose implantation was deferred until an outpatient setting. Together, the studies separate two kinds of signal: the electrical substrate CRT is intended to affect and the broader illness context surrounding implantation.

The electrical study included 415 consecutive CRT recipients treated at a Swedish tertiary center from 2015 through 2020. Median age was 72.8 years, 77.3% were male, median baseline left ventricular ejection fraction was 27.5%, and 43.1% had ischemic heart failure. Investigators measured left ventricular activation time (LVAT) on a standard electrocardiogram, defining it as the interval from QRS onset to the maximum deflection in lead V6. The combined endpoint was heart-failure hospitalization or death from any cause. Median follow-up was 2.8 years, with observation extending to 7.6 years.

Among recipients with left bundle branch block (LBBB) or intraventricular conduction delay, each 10-millisecond longer pre-implant LVAT was associated with a lower combined-event rate before adjustment: hazard ratio 0.92, with a 95% confidence interval of 0.86 to 0.99. After adjustment for clinical factors, the estimate weakened to 0.93, with a confidence interval of 0.85 to 1.01. That interval included no association. Post-implant LVAT was not associated with the endpoint, and pre-implant LVAT did not outperform conventional QRS duration in adjusted models. The finding is therefore a possible prognostic signal, not a validated replacement for established measurements.

The implantation-setting study linked institutional CRT data with the National Heart Failure Audit from 2014 through 2024. It compared 130 people who underwent CRT during a heart-failure hospitalization with 90 who received deferred outpatient CRT after hospitalization. The inpatient group entered the comparison with a different clinical profile: lower mean estimated kidney filtration, more ischemic cardiomyopathy, fewer people with LBBB, and more heart-failure hospitalizations during the preceding year.

Outcomes also differed. At least one heart-failure rehospitalization within one year occurred in 23.1% of the inpatient group and 7.8% of the deferred outpatient group. Median post-implant survival was 1,192 days for inpatients and 2,262 days for outpatients, a difference of 2.9 years. In an adjusted model, deferred outpatient implantation remained associated with lower all-cause mortality, with a hazard ratio of 0.56 and a 95% confidence interval of 0.35 to 0.88. Because the study was observational, this association cannot establish that the setting or timing itself caused the outcome difference.

Analysis — Substrate and setting mark different risks

The cross-study pattern is that an outcome observed after CRT may reflect at least two layers of selection. This is analysis, not a combined prediction model. LVAT attempts to characterize the electrical substrate that resynchronization targets, whereas inpatient status bundles acute illness, recent instability, kidney function, heart-failure cause, and other measured or unmeasured risks. The second cohort also had fewer people with LBBB in its inpatient group, while the first study found its clearest unadjusted LVAT signal among LBBB recipients. That cross-study alignment suggests a testable hypothesis: future CRT research could evaluate electrophysiologic phenotype and implantation context together rather than treating either as an isolated predictor. It does not show that one explains the other. The weakening of the LVAT association after adjustment and the persistent setting association after adjustment illustrate different statistical patterns, but residual confounding remains possible in both. The emerging direction is better prospective stratification of clinical acuity and electrical substrate, not a new rule for choosing when or for whom CRT is used.

Limitations

Both studies were retrospective and nonrandomized, so selection bias and residual confounding are central limitations. The Swedish study came from one tertiary center, was predominantly male, and included small morphology subgroups. Its primary LVAT association lost conventional statistical significance after multivariable adjustment. Post-implant electrical measurements were complicated by pacing-related changes, and detailed lead-position data were unavailable.

The implantation-setting evidence was available only at abstract depth. The source reports measured baseline differences and adjusted survival results, but the ingested text does not provide the full covariate list, missing-data handling, implantation rationale, or detailed event adjudication. People receiving inpatient CRT were already sicker on several measures, making confounding by indication especially plausible. The two cohorts were collected in different health systems and periods, and their measurements cannot be pooled directly. Neither study establishes that LVAT-guided selection or deferring implantation changes outcomes, and neither population-level association predicts an individual's course. Prospective studies with prespecified strata, comparable baseline groups, and independent validation would be needed to test the proposed interaction between electrical substrate and clinical context.