Local electrogram delay recorded from left ventricular lead at implant predicts response to cardiac resynchronization therapy: retrospective study with 1 year follow up

Abstract

Background: Considerable proportion of patients does not respond to the cardiac resynchronization therapy (CRT). This study investigated clinical relevance of left ventricular electrode local electrogram delay from the beginning of QRS (QLV). We hypothesized that longer QLV indicating more optimal lead placement in the late activated regions is associated with the higher probability of positive CRT response.

Methods: We conducted a retrospective, single-centre analysis of 161 consecutive patients with heart failure and LBBB or nonspecific intraventricular conduction delay (IVCD) treated with CRT. We routinely intend to implant the LV lead in a region with long QLV. Clinical response to CRT, left ventricular (LV) reverse remodelling (i.e. decrease in LV end-systolic diameter - LVESD ≥10%) and reduction in plasma level of NT-proBNP >30% at 12-month post-implant were the study endpoints. We analyzed association between pre-implant variables and the study endpoints.

Results: Clinical CRT response rate reached 58%, 84% and 92% in the lowest (≤105 ms), middle (106-130 ms) and the highest (>130 ms) QLV tertile (p < 0.0001), respectively. Longer QRS duration (p = 0.002), smaller LVESD and a non-ischemic cardiomyopathy (both p = 0.02) were also univariately associated with positive clinical CRT response. In a multivariate analysis, QLV remained the strongest predictor of clinical CRT response (p < 0.00001), followed by LVESD (p = 0.01) and etiology of LV dysfunction (p = 0.04). Comparable predictive power of QLV for LV reverse remodelling and NT-proBNP response rates was observed.

Conclusion: LV lead position assessed by duration of the QLV interval was found the strongest independent predictor of beneficial clinical response to CRT.

Figure 1

Measurement of the QLV. Printout of the electrophysiological recording system at 200 mm/s paper speed showing the interval from the beginning of the native QRS complex to the local electrogram from the LV lead. Labels: Lead II, III, aVR of the surface ECG; RVA-1/2 - right ventricular electrogram; LV-1/2 - LV electrogram; HRA 1/2 - right atrial electrogram.

Figure 2

Relationship of the QLV and CRT effects. The greater QLV at implantation of CRT system correlates with an increase in LVEF, decrease in LVESD, shortening of QRSd, and reduction in NT-proBNP at 12-month follow-up. Pearsons’s correlation coefficients (r) with p-values are provided. Abbreviations as in Table 1.

Figure 3

CRT responder rates in subgroups defined by tertiles of baseline variables. Response rates in percentages when population was categorized by tertiles of the QLV, QLV ratio, QRSd, and LVESD. Grey bars indicate clinical response to CRT and black bars proportion of patients who showed reverse LV remodelling. Abbreviations as in Table 1.