In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

PACE: Pacing to Avoid Cardiac Enlargement (PACE) study

Arrhythmias and Device Therapy

Cheuk-Man Yu
Presenter | see Discussant report Webcasts become available 24h after the presentation
Yu, Cheuk-Man
(Hong Kong SAR, People's Republic of China)
Open presentation slides

List of Authors:

Cheuk-Man Yu, MD, FRCP; Joseph Yat-Sun Chan, FHKAM(Medicine); Fang Fang, PhD; Qing Zhang, MM, PhD; Omar Razali, MD; Gabriel Wai-Kwok Yip, MD, FACC; Hussin Azlan, MD; Hamish Chi-Kin Chan, FRCP; Jeffrey Wing-Hong Fung, MD, FRCP


Background: The recently published Pacing to Avoid Cardiac Enlargement (PACE) study is a prospective, double-blinded, randomized, multicenter study that has confirmed the superiority of biventricular (BiV) pacing to right ventricular apical (RVA) pacing in prevention of left ventricular (LV) remodeling at 12 months follow up, in patients with bradycardia and preserved LV systolic function [ejection fraction (EF) ≥45%]. However, it is unclear whether this is related to pacing-induced systolic dyssynchrony. We hypothesized that patients who developed dyssynchrony at early stage is a major determinant of medium-term remodeling and deterioration of LV systolic function.

Methods: In the whole PACE study population (n=177) which included 89 patients randomized to BiV pacing and 88 to RVA pacing, serial echocardiography was performed at baseline, 1 month and 12 months. Systolic dyssynchrony was measured by tissue Doppler imaging using the LV 6-basal, 6-mid segmental model (i.e., Ts-SD >33ms) at 1-month follow up. Significant deterioration of LV systolic function was defined as reduction of EF ≥5% at 12 months.

Result: At 1 month, systolic dyssynchrony was present in 52% (n=46) of patients in RVA pacing group and 15% (n=13) in the BiV pacing group (2=28.25, p<0.001). At 12 months, increase in LV volume and decrease in EF was only observed in patients who developed systolic dyssynchrony at 1 month (Table). Reduction of EF ≥5% occurred in 67% (n=39) of patients who had dyssynchrony at 1 month, but only in 18% (n=21) who did not (χ2=40.83, p<0.001); and is more prevalent in the RVA than BiV pacing group (50% vs. 20%, χ2=17.71, p<0.001). By multivariate analysis, both systolic dyssynchrony at 1 month [odds ratios (OR): 7.081, 95% CI: 3.164-15.848, p<0.001) and RVA pacing mode (OR: 2.373, 95% CI: 1.070-5.264, p=0.033) were the two independent predictors for reduction of EF at 12 months. Furthermore, RVA pacing mode had incremental predictive value on top of pacing-induced systolic dyssynchrony at 1 month (χ2=4.324, p=0.037).

Conclusion: The PACE study shows that in patients received RVA pacing, medium-term LV remodeling and deterioration of EF occurred in those who developed early systolic dyssynchrony. These adverse consequences could be prevented effectively by BiV pacing.

Pace Abstract table

Discussant | see Presenter abstract Webcasts become available 24h after the presentation
Abraham, William
(United States of America)
See the presentation slides


This pre-specified analysis evaluated the potential mechanism of deterioration in left ventricular ejection fraction (LVEF) and increases in left ventricular volumes seen in response to right ventricular versus biventricular pacing in the PACE study. The main results of this study have been published (Yu CM et al., N Engl J Med 2009; 361:2123-2134). In 177 patients with a bradycardia pacing indication and normal LV systolic function randomized to receive right ventricular apical pacing or biventricular pacing, LV ejection fraction was significantly lower in the right ventricular pacing group than in the biventricular pacing group (54.8±9.1% vs. 62.2±7.0%, P<0.001) and LV end-systolic volume was significantly higher in the right ventricular pacing group than in the biventricular pacing group (35.7±16.3 ml vs. 27.6±10.4 ml, P<0.001), following 12 months of follow-up.
Clinical outcomes were not different between the two groups; however, the study was not powered to detect such differences. The findings suggest that conventional right ventricular apical pacing results in adverse LV remodeling and that biventricular pacing prevents these deleterious effects. The implication is that perhaps all patients with a bradypacing indication and normal LV systolic function should receive biventricular rather than right ventricular apical pacing.

The present PACE analysis was undertaken to explore the mechanism of pathological remodeling associated with pacing. Specifically, the investigators set out to determine whether pacing-induced systolic dysynchrony was the major determinant of deterioration of LV ejection fraction and increases in LV volumes. Systolic dysynchrony at one month (Dysynchrony Group) occurred in 52% of patients with right ventricular apical pacing, but only 15% with biventricular pacing (p<0 0.001).
The Dysynchrony Group had significantly lower LV ejection fraction with an absolute difference of 7.1% (p<0.001). LV end-systolic volume increased significantly in the Dysynchrony Group with a relative difference of 30.7% (p<0.001). Reduction of ejection fraction ≥5% occurred in 67% of patients in the Dysynchrony Group, but only in 18% in the No Dysynchrony Group (p<0 0.001). By univariate and multivariate analyses, right ventricular pacing mode was only weakly predictive of the fall in LVEF, while dysynchrony at 1 month was most strongly predictive (odds ratio [95% confidence interval] 7.081 [3.164 3.164-15.848], <0.001). No clinical characteristic at baseline was predictive of the decline in LVEF.

The results of this analysis speak to the mechanism of pathological remodeling with pacing; specifically, these results support the notion that pacing-induced dysynchrony causes the pathological LV remodeling seen. However, the practical application of these observations is unclear, since there was no clear baseline predictor of the decline in LVEF or the development of mechanical dyssynchrony that could be determined prior to device implantation and pacing. Moreover, only half of the right ventricular apical pacing patients demonstrated dysynchrony at 1 month, the strongest predictor of LV ejection fraction change, and 15% of the biventricular pacing group developed such dysynchrony.
This leaves us with a clinical dilemma: Should all patients with a bradypacing indication and normal LV systolic function receive a biventricular pacing device or should they receive a right ventricular pacing device and be upgraded later if dysynchrony is demonstrated at 1 month or later when the LVEF falls? The former approach may result in half of these patients receiving biventricular pacing unnecessarily. The latter approach potentially subjects half of the patients to a second implant procedure 1 month following the first.

In the future, the clinical decision will be guided by the results of large randomized controlled outcomes trials comparing right ventricular versus biventricular pacing, particularly by the BIOPACE and BLOCK-HF studies.
The first of these is evaluating right ventricular versus biventricular pacing in 1,800 patients enrolled regardless of ejection fraction. Many enrolled patients will likely be similar to those studied by Yu and colleagues, i.e. have normal LV systolic function.
The second will randomize approximately 1,600 patients with baseline LVEF less than or equal to 50% to right ventricular versus biventricular pacing. Taken together, these trials should inform clinical practice and expand upon Yu’s fine work.


711001 - 711002


Clinical Trial Update III
The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.