Discussion
Our report focuses on comparing the procedural outcomes of two current approaches to “physiological” pacing during the initial learning curve phase of implanting physicians at a single center. By describing this early experience with both forms of pacing, we hope to better inform other adopters as they consider both these options.
In this report, we observed the following: 1) His-bundle and left bundle branch area pacing result in similarly narrow paced QRS widths ≤ 120 ms (61.9% vs 65.9%, p = 0.62) and can be achieved with a low risk of acute complications. 2) Significant deterioration in pacing parameters emerged as soon as 4 weeks after implant in the HBP group and persisted/worsened over follow up. This pattern was not seen in the LBBAP group. 3) A significantly greater rate of adverse events (lead revisions and premature generator change) occurred in the HBP group (13%) compared to the LBBAP group (0%).
Immediately following implantation, the acceptable pacing endpoint (APE) was met by 18% fewer patients in the HBP group compared to the LBBAP group. This measure of lead safety and efficiency progressively worsened over time. By first follow-up, 40.4% fewer patients continued to meet the APE in the HBP group compared to LBBAP. This marked difference persisted to the most-recent follow-up. The deterioration of pacing parameters was driven primarily by worsening pacing thresholds, culminating in 6 of 45 HBP patients (13.3%) requiring lead revision and generator replacements at follow-up. R-wave amplitudes and lead impedance remained stable with additional follow-up in both groups.
Experienced implanters place leads into the RV apex with relative ease. Despite early procedural complications like lead dislodgement and perforation, progressive threshold rises rarely occur and apical pacing remains the traditional RV pacing site of choice.21 In contrast, the 13% of patients in the HBP group who required lead revisions in this report highlights a distinct shortcoming of choosing the His-bundle as the site for physiological pacing as opposed to LBBAP.
The incidence of progressive rise in thresholds and lead revision rates amongst early adopters of HBP has been variable in the literature.22-23 Bhatt et al, Keene et al and Teigeler et al in their respective single center reports described 8%, 7.5% and 11% rates of lead revision/intervention, respectively.11,14,24 On the other hand, Chaumont et al. reported a much lower incidence of lead revisions in their multicenter experience (3.4%) and Qian et al. reported no lead revisions.25-26 The threshold rises have been postulated to be the result of several mechanisms: 1) a relative lack of muscle in the underling region of the His-bundle 2) progressive fibrotic changes that occur over time after lead fixation, and 3) progressive degrees of micro-dislodgement.16,27 On the other hand, while published reports of LBBAP pacing have been primarily restricted to centers with significant technical expertise, reports of significant threshold changes and lead revisions are distinctly uncommon (<1%) in patients undergoing LBBAP.20,28-29
In their single center comparative study of HBP and LBBAP, Qian et al did not report any lead revisions, while noting an increase in capture threshold in 12.5% of patients in the HBP group.26 In contrast, our single center comparative report describes a 13% rate of lead revision/generator change in the HBP group. Importantly, our events all occurred beyond the one-year of follow-up period, beyond the time studied by Qian et al. Our study underscores how, with continued follow-up of patients with HBP, progressive worsening of pacing parameters over time clinically impacts the lives of our patients.
Our 13% event rate of lead revisions is higher than many of the above-mentioned reports and may reflect differences in our implant technique: for example, we did not consistently document His-bundle injury during the procedure, which has been described as predictive of lower chronic thresholds. While the early nature of our experience may partly explain these rates, they are in stark contrast to our LBBAP group where no patients experienced any additional procedures during follow-up. The comparatively shorter period of follow-up of our LBBAP group is a limitation. But at most recent follow-up, the LBBAP group has not shown any deterioration in pacing parameters, whereas the HBP group had already demonstrated a deterioration in APE at the same comparable point of follow-up.
While all LBBAP implanters in this study benefited from the learning curve of HBP, the differences in outcomes between both groups suggest distinct advantages of placing the lead deep within the septum. In fact, while early lead dislodgement and, rarely, lead perforations in the LV cavity can occur in LBBAP, threshold rises have not been reported using this technique and were not seen in our experience. Though it is likely that inexperienced operators will fail at consistently achieving selective or non-selective LBB capture (i.e., only left ventricular septal capture), the necessity of actual LBB capture is unclear and the almost obligate increases in pacing thresholds are not seen.30 LBBAP leads to very low lead revision/premature generator replacement rates with no incremental procedural risk for dependent patients. These should be requisite characteristics for the widespread adoption of any alternative to RV apical pacing.
Study Limitations :
This report describes the experience of a select few operators at a single institution. During this study, operators transitioned from pursuing HBP to LBBAP in an almost sequential fashion. The initial experience with HBP undoubtedly had a positive impact on certain procedural aspects of LBBAP such as fluoroscopy and procedural times. Importantly, this study did not systematically evaluate for the presence of LBB capture in the LBBAP group.