Discordant results from two trials comparing conduction system pacing with biventricular pacing for cardiac resynchronization therapy (CRT) leave open the question of whether conduction system pacing is as good as, better than, or worse than biventricular pacing for improving survival in heart failure patients with left bundle branch block.

In an analysis of the two trials, Jens Cosedis Nielsen, DMSc, PhD, a professor in the Department of Cardiology at Aarhus University Hospital in Aarhus, Denmark, concluded the designs proved too different to provide a yes-or-no answer.

Nielsen, lead author of an editorial accompanying publication of the trials in JAMA Cardiology, acknowledged one trial, which employed more selective entry criteria, favored conduction system pacing over biventricular pacing for long-term survival. But he does not believe conduction system pacing should be confirmed as the new standard.

Conduction System Pacing Might Improve Outcomes

“In selected patients, conduction system pacing may turn out to be the better approach in experienced hands, but that likely will be in a minority of CRT candidates, and we need more trials to be sure,” Nielsen said in an interview with Medscape Medical News.

Both studies aimed to determine whether conduction system pacing is a viable alternative to biventricular pacing, a long-time standard for addressing left bundle branch block (LBBB)-associated ventricular dyssynchrony in patients with heart failure and reduced ejection fraction (HFrEF).

Relative to biventricular pacing, conduction system pacing, also known as LBBB pacing, acts directly on the bundle branch system. A 2023 meta-analysis comparing these pacing strategies characterized conduction system pacing as offering a more physiologic approach. In the 12-trial meta-analysis, researchers found conduction system pacing superior for certain outcomes, including a reduction in heart failure hospitalizations, but the analysis included only one randomized trial.

In the newly published multicenter studies, patients with HFrEF with LBBB were randomized 1:1 to conduction system pacing or biventricular pacing. Yet there were many important differences in the types of patients selected, trial design, and trial methodology.

In the HeartSync-LBBP trial, which employed a superiority design, researchers enrolled 200 patients with HFrEF and LBBB at six centers in China. The conduction system pacing technical success rate was 98%. After 3 years of follow-up, the composite primary outcome of death or hospitalization for heart failure occurred in significantly fewer patients treated with conduction system pacing (8% vs 28%; P < .001).

Unique to this study, LBBB was “rigorously confirmed,” meaning correction of LBBB was demonstrated at pacemaker implantation, ensuring “a substantial proportion” of patients had cardiomyopathy induced by LBBB rather than myocardial injury, according to a senior author of HeartSync-LBBB, Yangang Su, MD, PhD, of Zhongshan Hospital of Fudan University in Shanghai, China.

In the randomized PhysioSync-HF trial, which employed a noninferior design, researchers included 173 patients at 14 centers in Brazil. Unlike HeartSync-LBBB, this study did not mandate conduction system pacing technical expertise among operators. The technical success rate was 69%.

After a median 1 year of follow-up, the primary hierarchical outcome favored biventricular pacing to a degree that precluded noninferiority for conduction system pacing. The hazard ratio (HR) for the time-to-event composite primary outcome of death, HF hospitalizations, or urgent HF visits was more than doubled in the conduction system pacing arm (HR, 2.35; 95% CI, 0.99-5.61).

Biventricular Pacing Just Shy of Superiority

In the PhysioSync-HF trial, the lower limit of the 95% CI fell below the line of unity, meaning a trend for, rather than confirmation of, statistical superiority for biventricular pacing. However, noninferiority as predefined for conduction system pacing relative to biventricular pacing was missed completely (P = .99), according to a senior author Andre D’Avila, MD, PhD, director of the Cardiac Arrhythmia Service at the Thorndike EP Institute at Beth Israel Deaconess Medical Center in Boston.

The differences were “driven by a higher incidence of clinical events as well as lesser improvement in left ventricular function,” D’Avila told Medscape Medical News.

Both death (12.6% vs 4.7%) and hospitalization for heart failure (6.9% vs 4.7%) were more frequent in the conduction system pacing arm. Urgent visits for heart failure were slightly lower in the conduction system pacing arm (1.1% vs 2.3%), but the mean improvement from baseline in the left ventricular ejection fraction at the end of the study was 3.8% lower in the conduction system pacing arm. In isolation, only death among the hierarchical endpoints provided biventricular pacing with a statistical advantage over conduction system pacing (HR, 3.36; 95% CI, 1.05-10.71).

Even though these are the two largest randomized trials to compare conduction system pacing and biventricular pacing, Nielsen made clear in his editorial that the top line results are not very helpful for answering the question of relative efficacy and utility. The problem is the methodological differences.

For example, pure LBBB pacing was employed in the conduction system pacing arm of the HeartSync-LBBB trial, but conduction system pacing in the PhysioSync-HF trial included deep septal pacing, His bundle pacing, and hybrid approaches as well as LBBB pacing in a “broad conduction system pacing concept,” Nielsen wrote in the editorial.

Nielsen also pointed out that all conduction system pacing was performed in HeartSync-LBBB by operators who had performed more than 250 prior cases, whereas 45% of the conduction system pacing implants in PhysioSync-HF were performed by operators with less than 40 previous cases.

Based on prior studies cited by Nielsen, there is a steep learning curve for conduction system pacing, a variable that might explain the differences in technical success in the HeartSync-LBBB and PhysioSync-HF studies. But Nielsen also emphasized that HeartSync-LBBB was designed specifically to enroll those most likely to have LBBB reversible by pacing.

‘Super Responder’ Target Population?

Owing to the HeartSync-LBBB entry criteria, the study was enriched for “the subgroup most likely to benefit from physiological resynchronization,” Nielsen wrote in his editorial. In distinguishing between conduction system pacing and biventricular pacing, he suggested there might be a “super responder” population for which conduction system pacing is the best choice, a possibility supported but not confirmed by HeartSync-LBBB.

The enrolled populations of the two studies also differed in other ways.

By using less rigorous screening for “true LBBB,” PhysioSync-HF was left with a population that “more likely resembled a broad population of patients with clinical heart failure,” which is likely to have more pathology driven by non-LBBB factors, Nielsen said.

The PhysioSync-HF population was more heterogenous in other ways. For example, relative to the homogenous population in HeartSync-LBBB, the PhysioSync population included substantial racial differences: 39% of participants were multiracial and 20% were non-White. PhysioSync also included more women than HeartSync-LBBB (50% vs 32%).

Even though the mean New York Heart Association (NYHA) heart class was approximately III in HeartSync-LBBB, only 7% of those enrolled in PhysioSync-HF were in NYHA class III. The remainder were class I or II.

Yet mortality rates were only about 1% per year (< 3% at the end of 3 years) in HeartSync-LBBB but 8.4% at the end of 1 year in PhysioSync-HF, a potential reflection of a greater burden of morbidities.

These and other differences not only complicate the comparison of conduction system pacing and biventricular pacing but also highlight the need for studies allowing an apples-to-apples comparison regarding patient characteristics and operator experience.

If restricted to these data, one possible interpretation, according to Nielsen, is that conduction system pacing is better than biventricular pacing when performed by experienced operators in patients with “truly correctable LBBB.” Biventricular pacing might be a better option in a less carefully selected population and when treatment is delivered at less experienced centers.

However, the cross-trial comparison only suggests, not proves, this interpretation, Nielsen said. He further noted these studies, even if they are the largest so far conducted, are still only of moderate size.

Nielsen played down any expectation that conduction system pacing could replace the biventricular pacing standard.

“I think that biventricular pacing will stay as a cornerstone in treating patients with heart failure and bundle branch block,” Nielsen told Medscape Medical News. He awaits further evidence that conduction system pacing should replace biventricular pacing in selected patients treated at experienced centers.

Trials Are Different, Not Conflicting

D’Avila, the PhysioSync-HF author, reached the same conclusion. On the surface, the HeartSync-LBBB and PhysioSync-HF trials may appear to demonstrate conflicting results, but he said these studies “were testing fundamentally different questions under vastly diverse conditions.”

“PhysioSync-HF tested the real-world effectiveness of a broader conduction system pacing approach,” he said, acknowledging and reemphasizing all of the differences Nielsen described.

“Ultimately, these trials do not contradict each other. Rather, they collectively suggest that while conduction system pacing can yield superior outcomes in carefully selected patients at expert centers, it is not yet ready to universally replace biventricular pacing as the first-line standard.”

He believes the editorial by Nielsen “perfectly synthesizes” this message.

Further Trials Need Narrow Focus

Each approach might have potential advantages for specific patients and settings, but this has to be proven, Nielsen said. He thinks the next step is “more trials to shed light on which patients should be considered for conduction system pacing relative to biventricular pacing” and what level of operator experience should be required if conduction system pacing is selected, he said.

Michael H. Kim, MD, an electrophysiologist and the chair of the Department of Medicine at Creighton University School of Medicine in Omaha, Nebraska, was the senior author of the 2023 meta-analysis. Even though conduction system pacing was favored over biventricular pacing in that study, he cautioned that it was “exploratory.”

“I agree that larger, more definitive randomized controlled clinical trials are needed on this topic,” he said.

In an interview with Medscape Medical News, Kim expressed concern about cross-trial comparisons, particularly for those conducted in diverse settings, such as China vs Brazil.

“The substantial differences in how healthcare is funded and delivered in these two countries, which may have impacted outcomes and access to specialized care, is another consideration,” he said.

Nielsen, Su, and D’Avila reported having no potential conflicts of interest. Kim reported a financial relationship with Sanofi.