Results
Clinical Characteristics of the Patients with MA-PVCs on Admission: The frequency of PVCs (%PVC) was calculated as: [number of PVC / number of total heart beats per 24 hours] x 100. The mean %PVC of the 22 patients (14 males and 8 females with a mean age of 61 ± 19 years, body mass index [BMI] of 25.3 ± 4.9 kg/m2, and serum creatinine of 0.95 ± 0.78 mg/dl) in this study was 20.1 ± 8.6%. The relationships between the LVDd and %PVC by 24-h Holter monitoring in the 22 patients before the RFCA were examined (Figure 2A). A significant correlation was found between the LVDd and %PVC (y = 29.43 + 15.76x, r = 0.499, p = 0.018), indicating that frequent PVCs might cause LV dilation. Thus, the 22 patients were divided into two groups (the lower [≤ 20%; n = 14] and upper [20% <; n = 8] groups) according to the quantile of the %PVC. The baseline characteristics of the patients are shown in Table 1. There were no statistical differences among all the factors except no medication including the age (55 ± 19 versus 71 ± 14; p = 0.062), BMI (24.9 ± 4.7 versus 26.1 ± 5.4 kg/m2; p = 0.607), serum creatinine (0.76 ± 0.12 versus 1.30 ± 1.27; p = 0.121), prevalence of an inferior (93% versus 100%; p = 0.463) or superior axis (7% versus 0%; p = 0.174) in the frontal plane of the 12-ECG of the MA-PVCs, being male (64% versus 63%; p = 0.683), co-existence of hypertension (36% versus 75%; p = 0.083), dyslipidemia (21% versus 38%; p = 0.440), diabetes mellitus (14% versus 25%; p = 0.553), ex- or current smoking (50% versus 38%; p = 0.592), prevalence of symptoms including palpitations (93% versus 100%; p = 0.463), general fatigue (43% versus 50%; p = 0.760), chest discomfort / pain (21% versus 50%; p = 0.182), fainting (14% versus 13%; p = 0.912), and medications including beta-blockers (43% versus 75%; p = 0.160), calcium channel blockers (21% versus 13%; p = 0.622), and class I agents (21% versus 35%;p = 0.856) between the Lower and Higher group. The prevalence of no medications (43% versus 0%; p = 0.030) in the Lower group was higher than that in the Higher group. No patients were on amiodarone or sotalol. Although those agents were not sufficiently effective in eliminating the PVC-associated symptoms before the RFCA, all of the patients with a successful procedure reported the absence of any PVC-associated symptoms and could discontinue the antiarrhythmic agents after the RFCA. The mean age of the patient, serum creatinine, and prevalence of hypertension were slightly older and higher in the Higher group than Lower group, but the differences were not significant between the two groups.
Analysis of the 24-h Holter monitoring, New York Heart Association (NYHA) functional class, and serum brain natriuretic peptide (BNP) concentration: Table 2 summarizes the analysis of the 24-h Holter monitoring, NYHA functional class by the specific activity scale, and serum BNP concentration. In 20 patients who achieved procedural success, the number of total heart beats did not statistically differ before (106,911 ± 13,950 versus 110,788 ± 9,666 beats per day; p = 0.504) and after (106,506 ± 12,689 versus 106,604 ± 7,623 beats per day; p = 0.985) the RFCA between the Lower and Higher groups. The RFCA significantly reduced the number of total PVCs (16,046 ± 3,804 versus 54 ± 31 beats per day; p < 0.001 and 32,927 ± 10,439 versus 147 ± 161 beats per day; p < 0.001) and %PVC (14.9 ± 2.3 versus 0.05 ± 0.03%; p < 0.001 and 29.3 ± 7.6 versus 0.14 ± 0.14%; p < 0.001) with no evidence of a recurrence of the culprit PVC in the Lower and Higher groups. There was no statistical difference in the number of total PVCs (54 ± 31 versus 147 ± 161 beats per day; p = 0.065) and %PVC (0.05 ± 0.03 versus 0.14 ± 0.14%; p = 0.064) after the RFCA between the two groups. There was no statistical difference in the frequency of premature atrial complexes before and after the RFCA between the three groups (data not shown). The NYHA functional class and serum BNP concentration were evaluated in all patients on admission for the RFCA. Before the RFCA, the NYHA functional class and serum BNP concentration were demonstrated to be significantly worse and higher in the Higher group than in the Lower group (1.67 ± 0.49 versus 2.50 ± 0.53; p = 0.002 and 44 ± 28 versus 145 ± 94; p = 0.003). RFCA significantly improved the NYHA functional class and serum BNP concentration in both the Lower (1.67 ± 0.49 versus 1.00 ± 0.00; p < 0.001 and 44 ± 28 versus 20 ± 12; p = 0.014) and Higher (2.50 ± 0.53 versus 1.00 ± 0.00; p < 0.001 and 145 ± 94 versus 40 ± 26; p = 0.015) groups as compared to that before the intervention.
Analysis of the Echocardiogram: Figure 2 summarizes the results of the analysis of the echocardiograms. Before the RFCA, a subgroup of patients in the Higher group demonstrated a significantly reduced LVEF (73 ± 8 versus 61 ± 8%; p = 0.005), enlarged LVDd (48 ± 5 versus 53 ± 5 mm; p = 0.042) and LVDs (31 ± 5 versus 38 ± 4 mm; p = 0.002), as compared to the Lower groups (Figure 2BCD). However, at the 6 month examination after the RFCA, all of those abnormalities seen in the Higher group before the RCFA had reversed (Figure 2BCD), and there was no statistical difference in these parameters between the two groups. There was no statistically differences in the RVD and MR before and after the RFCA between the two groups (Figure 2EF).
Electrophysiological Findings and Approach Sites with a Successful RFCA: Table 3 summarizes the results of the electrophysiological findings and successful ablation sites. An RFCA procedure for MA-PVCs was performed in 22 patients. Procedural success was achieved in 20 (91%) of 22 patients. No patients suffered from any procedure-related complications. During the follow-up, recurrence of the culprit MA-PVCs was observed in one patient (5%), who underwent a repeat RFCA with a successful result. The mean QRS duration and PDI of the PVCs in the 22 patients were 142 ± 28 ms and 0.60 ± 0.07, respectively. There were no statistical differences in the QRS duration (145 ± 27 versus 135 ± 29 ms; p = 0.235) and PDI (0.61 ± 0.08 versus 0.57 ± 0.06; p = 0.438) of the PVCs between the Lower and Higher groups.
The approach sites of the MA-PVCs for the RFCA are demonstrated in Figure 3B-F. Of the 22 patients, in 13 (59%), 2 (9%), and 1 (5%) patient a successful RFCA on the ventricular side of the MA from the supra-mitral valve (Figure 3B) and infra-mitral valve (Figure 3C) using the trans-interatrial septal approach and trans-coronary sinus approach (Figure 3E) was achieved. Interestingly, in 4 (18%) patients a successful RFCA on the LA side of the MA using a trans-interatrial septal approach was achieved (Figure 3D). A successful RFCA of the MA-PVCs could be successful RFCA of MA-PVCs in 20 patients with these steps. The notable point is that, of the 20 patients with a successful RFCA, in 19 (95%) patients a successful RFCA using a trans-interatrial septal approach could be achieved. However, in the remaining 2 patients successful results could not be achieved even though additional RFCA using the trans-aortic approach was performed (Figure 3F).
Figure 1B-G demonstrated that the EnSiteTM 3D-mapping images of the earliest activation sites during the culprit PVC (Figure 1BE) and intra-cardiac electrocardiograms (Figure 1CF) and 12-ECGs of the pace map (Figure 1DG) at the successful ablation sites. The EnSiteTM 3D-mapping images demonstrated that the earliest activation sites during the culprit PVCs were confirmed in the white area on the ventricular (Figure 1B) and LA (Figure 1E) side of the MA. The yellow points were the successful ablation sites. The findings from the unipolar potentials recorded by the ablation catheter demonstrated a QS pattern, bipolar potentials were recorded by the tip of the ablation catheter that proceeded the QRS in the 12-ECG by 32 ms (Figure 1C) during the culprit PVC, and a suboptimal pace map during pace mapping at 9.9 V (Figure 1D) was obtained by the ablation catheter placed at the successful ablation site on the ventricular side of the MA using a supra-mitral valve approach (Figure 3B). Of course, the atrial potentials were not confirmed by the tip of the ablation catheter. On the other hand, the findings from the unipolar potentials recorded by the ablation catheter demonstrated a QS pattern and bipolar potentials recorded by the tip of the ablation catheter that proceeded the QRS in the 12-ECG by 48 ms (Figure 1F) during the culprit PVC were obtained by the ablation catheter placed at successful ablation site on the LA side of the MA (Figure 1E). Of course, the atrial potential was confirmed by the tip of the ablation catheter, and pace mapping at 9.9 V in the LA side of MA (Figure 1E) captured the LA but not the LV (Figure 1G). After radiofrequency energy was delivered at those points, the PVCs were steadily terminated.
Figure 3G summarizes the earliest activation sites of the MA-PVCs that were successful (red circle; n = 20) or unsuccessful (black circle; n = 2) RFCA sites in the left anterior oblique view of the MA. They were crowded in the direction from 0:00 to 3:00 in the left anterior oblique view of the MA.
Outcome of patients with an unsuccessful RFCA. Of the 22 patients, in 2 patients in the Lower group, unfortunately the procedure was unsuccessful during this study. One patient received a repeat RFCA with a successful result. Because the other did not want to receive a repeat RFCA, medical therapies were continued. Tables 1, 2, and 3 and Figure 2A-D summarizes the clinical characteristics (Table 1), results of the analysis of the 24-h Holter monitoring, NYHA functional class, serum BNP concentration (Table 2), echocardiograms (Figure 2B-F), and electrophysiological findings (Table 3) of those 2 patients who looked like men with a slender build (Table 1). All of those factors including the %PVC, NYHA functional class, serum BNP concentration, and LVDd, LVDs, and LVEF obtained by the echocardiogram, in the 2 patients with an unsuccessful RFCA still deteriorated further during the follow-up. The mean QRS duration and PDI were 153 ms and 0.57, respectively.