Baseline characteristics of the study patients
Baseline characteristics of the study participants are listed in Table 1. During a median 4.9 (interquartile range: 4.7-5.1) years follow-up, 198 (18.4 %) patients (group 1, mean age 72.1 ± 9 years; 59.6% female) experienced progressive post-implant TR, whereas 877 patients (group 2, mean age 71.9 ± 12 years; 49.8% female) did not have progressive post-implant TR. The percentage of female individuals was higher in group 1 than group 2. Additionally, the prevalence of atrial fibrillation (paroxysmal or non-paroxysmal) was also higher in group 1. A higher percentage of sick sinus syndrome for PPM was noted in group 1 (group 1 vs. group 2; 63.1% vs. 53.9%; p=0.022). There was no difference in the distribution of ventricular lead position, pacing QRS duration, ventricular pacing percentage, serum creatinine level and medication used between the 2 groups.
Pre-implant and post-implant echocardiographic parameters of study patients
At pre-implant, group 1 had significantly larger LA dimension (group 1 vs. group 2; 37.8 ± 6.6 mm vs. 36.7 ± 6.3 mm; p = 0.063) and significantly higher average TRPG (group 1 vs. group 2; 23.1 ± 4.9 mmHg vs. 20.7 ± 6.1 mmHg; p < 0.001) than group 1 (Table 2). The two groups did not differ in LVEDV and LVEF.
The median follow-up period was similar between the two groups (group 1 vs. group 2; 4.7 (4.4-5.4) years vs. 4.5 (4.2-4.8) years; p = 0.610). At post-implant, group 1 had significantly larger LA dimension, lower LVEF and more severe TR grade than group 2. Additionally, group 1 had significantly higher post-implant TRPG (group 1 vs. group 2; 35.9 ± 9.1 mmHg vs. 21.8 ± 5.4 mmHg; p < 0.001) and larger changes in post-implant TRPG (group 1 vs. group 2; 12.8 ± 9.6 mmHg vs. 1.1 ± 7.6 mmHg; p < 0.001) than group 2. Figure 2 showed the changes in pre-implant and post-implant TRPG in group 1 (p < 0.001).
Figure 3 shows the cumulative incident rate of progressive TR grade and/or abnormal TRPG from 1.3% in the first year to 18.4% in the sixth year in the study cohort.
Univariable and multivariable Cox regression analyses of predictors of progressive post-implant TR
Female gender, end stage renal disease, atrial fibrillation, larger pre-implant LA dimension, and higher pre-implant TRPG, were included for multivariable Cox regression analyses of progressive post-implant TR (Table 3). However, only pre-implant TRPG (HR: 1.075; 95% confidence interval (CI): 1.035-1.117; p = 0.001) was independent predictor of progressive post-implant TR (Table 3).
Compared to patients with pre-implant TRPG ≤25 mmHg, patients with pre-implant TRPG between >25 and ≤30 mmHg had 1.825 times (CI: 1.202-2.770; p=0.006) of relative risk of developing progressive post-implant TR during follow-up period.
Clinical outcomes of the study patients
During the follow-up period, group 1 had a significantly higher incidence of HF hospitalization compared to group 2 (13.6% vs. 4.7%; p < 0.001) (Table 4 and Figure 4). However, the incidence of late-onset atrial fibrillation, sudden death or ventricular tachyarrhythmias, cardiovascular mortality (group 1 vs. group 2; 5.7% vs. 4.1%; p=0.413), and all-cause mortality (group 1 vs. group 2; 16.2% vs. 14.1%; p=0.503) did not differ between the two groups during follow-up period (Table 4).
Univariable and multivariable Cox regression analyses of predictors of HF hospitalization
By univariable Cox regression analyses, older age, high body mass index, diabetes mellitus (DM), coronary artery disease (CAD), longer pacing QRS length, ventricular lead position at the lower septum and apex, larger pre-implant LA dimension and larger pre-implant LVEDV, larger post-implant LVEDV, lower post-implant LVEF, and progressive post-implant TR were significant predictors of HF hospitalization (Table 5). However, by multivariable Cox regression analyses, only older age (HR: 1.073; 95% CI: 1.037-1.110; p < 0.001) and CKD stage of >3 (moderate to severe CKD) (HR: 1.865; 95% CI: 1.008-3.450; p = 0.047) were independent predictors of HF hospitalization. Post-implant LVEF (HR: 0.957; 95% CI: 0.934-0.980; p < 0.001) was independently inversely associated with HF hospitalization in multivariable analysis. Progressive post-implant TR was significantly associated with HF hospitalization in univariable analysis (HR: 2.459; 95% CT: 1.511-4.000; p < 0.001) and had a non-significant trend toward HF hospitalization in multivariable analysis (HR: 1.694; 95% CI: 0.959-2.994; p = 0.070).