- Email: [email protected]
Implications of right ventricular septal pacing for medium-term prognosis: Propensity-matched analysis
International Journal of Cardiology 220 (2016) 214–218
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Implications of right ventricular septal pacing for medium-term prognosis: Propensity-matched analysis Akira Mizukami a,⁎,1, Yuya Matsue a,1, Yoshihisa Naruse b,1, Shinya Kowase c,1, Kenji Kurosaki c,1, Makoto Suzuki a,1, Akihiko Matsumura a,1, Akihiko Nogami b,1, Kazutaka Aonuma b,1, Yuji Hashimoto a,1 a b c
Department of Cardiology, Kameda Medical Center, 929 Higashi-cho, Kamogawa, Chiba 296-8602, Japan Cardiovascular Division, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan Department of Cardiovascular Medicine, Yokohama Rosai Hospital, 3211 Kozukue-cho, Kohoku-Ku, Yokohama, Kanagawa 222-0036, Japan
a r t i c l e
i n f o
Article history: Received 24 January 2016 Received in revised form 15 May 2016 Accepted 26 June 2016 Available online 28 June 2016 Keywords: RV septal pacing RV non-apical pacing Prognosis Heart failure Mortality Propensity matching
a b s t r a c t Background: The effect of right ventricular (RV) septal pacing as opposed to RV apical pacing on prognosis in patients undergoing pacemaker implantation remains controversial. This study was performed to examine the clinical efficacy of RV septal pacing in a large cohort with medium-term follow-up and propensitymatched analysis. Methods: A total of 982 consecutive patients with first pacemaker implantation between 2008 and 2013 at two centers in Japan (51.4% male, age 76.1 ± 10.6 years, 64.3% septal pacing, 94% preserved ejection fraction [EF]) were enrolled. Propensity matching successfully matched 446 patients into RV septal and apical pacing groups. The primary endpoint, a combination of all-cause death and hospitalization due to heart failure, was compared between the two groups. Results: In the propensity-matched cohort, the primary endpoint was observed in 61 patients (13.7%) over a median follow-up period of 2.1 years (interquartile range, 1.1–3.5 years). The effects of septal pacing on prognosis were not statistically significant (hazard ratio [HR] = 1.10, 95% confidence interval [CI] = 0.60–2.04, P = 0.752). No significant benefit of septal pacing was observed on all-cause death (HR = 1.86, 95%CI = 0.74–4.66, P = 0.187) and heart failure hospitalization (HR = 0.93, 95%CI = 0.44–1.98, P = 0.847) when assessed separately. Conclusion: Septal pacing did not show medium-term advantages in prognosis in this large-scale retrospective cohort study with propensity matching of patients with predominantly preserved EF. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The right ventricular (RV) apex has long been the preferred ventricular pacing site for transvenous cardiac pacemakers due to technical ease of implantation, good stability with both passive and active fixation leads, and reliability based on its history with substantial clinical data [1–3]. However, the non-physiological ventricular activation pattern caused by RV apical pacing produces interventricular and intraventricular dyssynchrony, resulting in decreased cardiac output, increased myocardial workload and oxygen consumption, and changes in neurohormonal and electrophysiological activity, resulting in ventricular remodeling. This may lead to left ventricular systolic and diastolic dysfunction, heart failure, and atrial fibrillation [4–6].
Due to these deleterious effects of RV apical pacing and dyssynchrony, there have been attempts to identify a better target site. Possible targets include RV outflow tract, RV septum, His bundle, left ventricular, and biventricular pacing [7–9]. RV outflow tract and septal pacing are attractive options due to the relative technical ease of lead implantation compared to other techniques, with leads traditionally used for RV apical pacing [10]. However, observational studies and small-scale randomized controlled trials failed to confirm its prognostic benefit, although electrocardiographic, hemodynamic, echocardiographic, and neurohormonal benefits have been reported [2,11,12]. This study was performed to examine the clinical efficacy of RV septal pacing in a large cohort study with medium-term follow-up and propensity-matched analysis. 2. Methods
⁎ Corrresponding author at: Department of Cardiology, Kameda Medical Center, 929 Higashi-cho, Kamogawa, Chiba 296-8602, Japan. E-mail address: [email protected] (A. Mizukami). 1 All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
http://dx.doi.org/10.1016/j.ijcard.2016.06.250 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
2.1. Study population This study retrospectively included 982 consecutive patients receiving their first pacemaker implantation with RV lead placement between January 2008 and
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218 December 2013 at two centers in Japan (Kameda Medical Center and Yokohama Rosai Hospital; 51.4% male, age 76.1 ± 10.6 years, 64.3% septal pacing). The indications for pacemaker implantation were decided according to the guidelines of the Japanese Circulation Society [13]. The target site of RV lead placement was decided by the caring physician on the bases of patient background and operator preference. The location of the RV lead was assessed at the time of implantation by right anterior oblique and left anterior oblique fluoroscopic projections, as well as paced QRS morphology during implantation using the methods reported previously [14], and was followed-up by biplane chest radiography and 12-lead ECG after implantation. RV outflow tract pacing was included in the RV septal pacing group. 2.2. Outcome analysis The primary endpoint of this study was a combination of all-cause death and hospitalization due to heart failure. The secondary endpoints included the individual components of the primary endpoint, as well as adverse events related to the implantation procedure, including lead perforation and dislodgement. Data at the time of implantation procedure were collected, including age, sex, diagnosis for implantation (AV block, sick sinus syndrome [SSS], or others), past history (hypertension, hyperlipidemia, diabetes mellitus, heart failure, atrial fibrillation, and ischemic heart disease), medications (beta-blockers, angiotensin converting enzyme inhibitors/angiotensin receptor blockers, and calcium channel blockers), ECG parameters (QRS interval, presence of complete left bundle branch block [CLBBB]), laboratory parameters (hemoglobin, estimated glomerular filtration rate [eGFR], and B-type natriuretic peptide [BNP]), and left ventricular ejection fraction (LVEF) on transthoracic echocardiography. The diagnosis of AV block included any degree of AV block with indication for pacemaker implantation. Hypertension, hyperlipidemia, and diabetes mellitus were scored based on the previous diagnosis and initiation for therapy. Heart failure, atrial fibrillation, and ischemic heart disease were scored based on previous history. The Modification of Diet in Renal Disease (MDRD) study equation with Japanese coefficient was used to calculate eGFR. This new Japanese equation is currently recommended by the Japanese Society of Nephrology for accuracy in the Japanese population [15]. Latest follow-up date and value of BNP were also obtained. Data regarding outcome were obtained by a single investigator who was unaware of the patients' information, including RV pacing site. Adverse events related to the ventricular lead implantation procedure were also assessed and compared between the two groups. The number and rate of micro- and macro-lead dislodgements and lead perforation were compared between the two pacing sites. Other complications related to device implantation were not assessed. Ethical approval was obtained from the institutional review board of each participating hospital. 2.3. Statistical analysis Continuous variables are expressed as the means ± standard deviation or medians (interquartile range [IQR]). All categorical variables are presented as the number and percentage in each group. For the whole cohort, comparisons of continuous variables were tested by Student's t test or Mann–Whitney U-test according to the data distribution. Comparisons of categorical variables were compared by chi-square analysis or Fisher's exact test. In propensity-matched analysis, the variables included in the logistic regression model for calculation of the propensity score (PS) included age, diagnosis for implantation, history of hypertension, history of heart failure, history of atrial fibrillation, presence of CLBBB, QRS interval duration, LVEF b 50%, hemoglobin, and eGFR. Matching was performed using the logit-transformed PS. An optimal-matching algorithm with a caliper width of 0.2 standard deviations of the logit of the PS was used [16]. The balance of the variables between the two groups was judged by standardized differences, and standardized difference b 10% was taken to be satisfactory [16,17]. If the standardized differences for all covariates were b 10%, PS matching was generally considered successful, i.e., two groups are well-balanced as if they were randomized at least in the observed covariates. “Time 0” for survival analyses was the date of pacemaker implantation. For the total cohort, comparison of the probability of freedom from the prognostic binary endpoints between groups was performed by Kaplan–Meier survival analysis with estimation of the hazard ratio from a Cox regression model. Univariate and multivariate Cox regression analyses were also performed to determine the prognostic implications of each variable, including RV septal pacing, on the endpoints. Variables with P b 0.1 on univariate analysis were entered into multivariate analysis. Logarithmic transformations of the variables without a normal distribution were used for Cox regression analysis. The estimates of the parameters were given with their 95% confidence intervals. All P-values reported are 2-sided, and P b 0.05 was considered statistically significant. All statistical analyses were performed with R (The R Foundation for Statistical Computing, Vienna, Austria, version 3.1.1).
3. Results 3.1. Patient characteristics Between January 2008 and December 2013, a total of 982 patients (51.4% male, age 76.1 ± 10.6 years) undergoing their first pacemaker implantation were included in this analysis. The RV lead was placed at the septum in 631 (64.3%) patients, and the apex in 351 (35.7%)
215
patients. LVEF was preserved (≥50%) in 94% of the patients. The patient variables of the study population before and after PS matching for each site of RV lead placement are shown and compared in Table 1. Before PS matching, patients in the apical pacing group were older, lower proportion of males, less SSS, and more AV block as diagnosis for implantation compared to patients in the septal pacing group. There was less history of atrial fibrillation and ischemic heart disease, wider QRS complex, more CLBBB, less hemoglobin, less eGFR, and less LVEF at baseline in the apical pacing group than the septal pacing group. PS matching successfully matched 446 patients into two groups with 223 patients in each group. After matching, the baseline characteristics of the two groups were well balanced with standardized difference of all baseline variables between matched groups b 10% (Table 1). 3.2. Paced QRS intervals Paced QRS intervals were compared between the two groups after pacemaker implantation in 828 patients for whom records of ECG with RV pacing immediately after implantation were available. In the whole cohort, the paced QRS interval of the septal pacing group was shorter than that of the apical pacing group, but the difference was small and not statistically significant (153.7 ± 17.6 ms vs. 155.3 ± 23.3 ms, P = 0.306). Records of RV paced ECG immediately after implantation were available for 384 patients in the propensity-matched cohort. The differences were even smaller and also not significant in the propensity-matched cohort (153.6 ± 17.4 ms, vs. 153.9 ± 24.5 ms, P = 0.891). Although obvious fusion and pseudo-fusion ECGs were excluded, fusion with intrinsic QRS could not be totally eliminated due to the retrospective nature of this study. 3.3. Primary outcome In the propensity-matched cohort of 446 patients, the primary endpoint (combined endpoint of all-cause death and hospitalization due to heart failure) was observed in 61 patients (13.7%) over a median follow-up of 2.1 years (IQR = 1.1–3.5 years). No improvement of prognosis with septal pacing was observed in the matched cohort (HR = 1.10, 95%CI = 0.60–2.04, P = 0.752, Fig. 1). 3.4. Secondary outcome All-cause death and hospitalization due to heart failure were also assessed separately to further clarify the effects of septal pacing. Allcause death was observed in 32 patients (7.2%) in the propensitymatched cohort, and RV septal pacing showed no significant effect (HR = 1.86, 95%CI = 0.74–4.66, P = 0.187, Fig. 2). In the propensity-matched cohort with heart failure hospitalization in 40 patients (9.0%), no improvement of heart failure hospitalization was observed with RV septal pacing (HR = 0.93, 95%CI = 0.44–1.98, P = 0.847, Fig. 3). Data regarding the changes in BNP values on admission and follow-up ([admission value] - [follow-up value]) were also collected, and was compared between the two pacing sites. The median interval of the admission and the follow-up tests were 387 days (IQR = 15–919 days), without significant differences between the septal and apical pacing groups (median 372 days [IQR = 30–776 days] vs. median 400 days [IQR = 10–978 days], P = 0.639). There were no statistically significant differences in the changes of the BNP values between the two groups (− 26 pg/ml [IQR = − 225–10 pg/ml] vs. -30 pg/ml [IQR = − 197–17 pg/ml, P = 0.726). 3.5. Data from the total cohort Among 982 patients, the primary endpoint was observed in 159 patients (16.2%) in the total cohort over the median follow-
216
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218
Table 1 Demographic and clinical characteristics of the study population before and after propensity matching. Before matching
Age, years Male sex, n (%) Diagnosis AV block, n (%) Sick sinus syndrome, n (%) Others, n (%) Past history Hypertension, n (%) Hyperlipidemia, n (%) Diabetes mellitus, n (%) Atrial fibrillation, n (%) Heart failure, n (%) Ischemic heart disease, n (%) Medication β-Blocker, n (%) ACE-I/ARB, n (%) Calcium blocker, n (%) Non-paced ECG parameters QRS complex duration, ms CLBBB, n (%) Laboratory parameters Hemoglobin, g/dl Estimated GFR, ml/min/1.73m2 B-type natriuretic peptide, pg/ml Echocardiographic parameters Left ventricular EF,% LVEF ≥ 50%, n (%)
After matching
Apical pacing
Septal pacing
(n = 351)
(n = 631)
78 ± 10 162 (46)
75 ± 11 343 (54)
196 (56) 138 (39) 17 (5)
P-value
Apical pacing
Septal pacing
Std. mean difference
(n = 223)
(n = 223)
Before
After
b0.001 0.014
78 ± 9 100 (45)
78 ± 9 105 (47)
29.96 16.05
0.22 4.01
277 (44) 308 (49) 46 (7)
b0.001 0.005 0.134
130 (58) 92 (41) 8 (4)
120 (54) 103 (46) 11 (5)
24.15 20.21 8.38
8.05 9.93 4.95
208 (59) 96 (27) 76 (22) 100 (28) 71 (20) 48 (14)
410 (65) 214 (34) 159 (25) 240 (38) 112 (18) 136 (22)
0.075 0.034 0.212 0.003 0.339 0.003
132 (59) 60 (27) 47 (21) 61 (27) 39 (18) 31 (14)
139 (62) 68 (30) 49 (22) 65 (29) 42 (19) 31 (14)
12.38 6.39 7.09 21.31 5.10 21.08
6.13 6.63 2.43 4.43 5.17 0.00
71 (20) 160 (46) 151 (43)
149 (24) 327 (52) 286 (45)
0.223 0.061 0.486
41 (18) 106 (48) 100 (45)
45 (20) 116 (52) 108 (48)
9.67 12.01 4.02
5.06 7.98 6.01
111 (97–143) 42 (12)
100 (90–128) 36 (6)
b0.001 b0.001
110 (97–145) 24 (11)
114 (92–144) 23 (10)
42.94 21.13
7.53 3.26
12 ± 2 55 (40–70) 190 (81–448)
13 ± 2 60 (44–71) 161 (78–349)
b0.001 0.004 0.086
12 ± 2 56 (42–70) 154 (69–381)
12 ± 2 58 (41–69) 168 (84–385)
30.13 20.63 9.48
2.02 0.95 7.08
69 (63–74) 294 (94)
70 (63–77) 483 (94)
0.030 0.877
70 (64–75) 214 (96)
71 (63–78) 213 (96)
13.25 1.10
8.28 0.49
Values are reported as the means ± standard deviation, median (interquartile range), or n (%). ACE-I, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; CLBBB, complete left bundle branch block; GFR, glomerular filtration rate; EF, ejection fraction.
up period of 2.2 years (IQR, 1.1–3.8 years). Superiority of septal pacing was not observed in Kaplan–Meier survival analysis (Log-rank, P = 0.094, Fig. 1 in Ref [18]), as well as multivariate Cox regression analysis (HR = 1.35, 95%CI = 0.90–2.04, P = 0.149, Table 1 in Ref [18]), consistent with the results of propensity-matched analysis. All-cause death occurred in 104 patients (10.6%) in the whole cohort. Septal pacing failed to prove the prognostic benefit in Kaplan–
Meier survival analysis (Log-rank, P = 0.079, Fig. 2 in Ref [18]) and multivariate Cox regression analysis (HR = 0.86; 95%CI = 0.0.52– 1.47, P = 0.616). Hospitalization due to heart failure occurred in 78 patients (7.9%) in the whole cohort. Septal pacing did not show any advantage in Kaplan–Meier survival analysis (Log-rank, P = 0.682, Fig. 3 in Ref [18]) or univariate Cox regression analysis (HR = 0.91, 95%CI = 0.57–1.44, P = 0.682).
Fig. 1. Kaplan–Meier curves for combined primary endpoint of all-cause death and heart failure hospitalization of propensity-matched cohort No significant difference was observed between the two pacing sites.
Fig. 2. Kaplan–Meier curves for all-cause death of propensity-matched cohort No significant difference was observed between the two pacing sites.
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218
Fig. 3. Kaplan–Meier curves for heart failure hospitalization of propensity-matched cohort No significant difference was observed between the two pacing sites.
3.6. Adverse events There were no statistically significant differences in incidence rates of ventricular lead macro-dislodgement (4 [1.1%] vs. 6 [1.0%] patients, P = 0.778), lead micro-dislodgement (4 [1.1%] vs. 6 [1.0%] patients, P = 0.778), or lead perforation (1 [0.3%] vs. 2 [0.3%] patients, P N 0.999) between the apical and septal pacing groups. 4. Discussion
217
duration (especially b 1 year) have been argued as reasons for negative results in some studies, the Protect-Pace trial investigators recently reported the results of an international multicenter randomized controlled trial in 240 patients with high-grade AV block requiring N 90% ventricular pacing with baseline LVEF N 50% over a follow-up period of 2 years [19]. There were no significant differences in the primary endpoint of intra-patient changes in LVEF, heart failure hospitalization, mortality, burden of atrial fibrillation, or plasma brain natriuretic peptide levels between the RV apical and RV high septal lead position. Although this study was not powered to detect the prognostic endpoints, this result suggests that septal pacing is unlikely to produce a positive prognostic effect even in highly pacemaker-dependent patients if LVEF is preserved. The results of our study, which is one of the largest cohort studies and first to use propensity matching, were consistent with the Protect-Pace trial regarding the lack of an effect of septal pacing on mortality and heart failure hospitalization, even with a relatively long follow-up period. It should be emphasized that our study population consisted mainly of patients with preserved EF (LVEF ≥ 50% in 96% of PS-matched cohort and 94% of the whole cohort), similar to the Protect-Pace study cohort. More favorable results may be obtained in patients with reduced cardiac performance considering the results of the above systematic review and metaanalysis by Shimony et al. [12], but definitive evidence is still lacking. Moreover, the BLOCK HF trial clearly showed that in pacemakerdependent patients with heart failure and LVEF ≤ 50%, cardiac resynchronization therapy was superior to RV pacing alone with regard to prognosis and negative remodeling [20]. The differences in paced QRS intervals between RV apical and septal pacing groups were inconsistent in previous studies, with some reports of shorter QRS intervals with RV septal pacing, and others without differences, which may theoretically affect the efficacy of septal pacing [11,21]. In this study, no differences were observed between the two pacing sites.
4.1. Main findings There have been no large-scale randomized controlled trials to determine the efficacy of RV septal pacing on the medium to long term prognostic clinical endpoint in patients with pacemaker indications. This is one of the largest cohort studies with a population of 982 patients (446 matched patients) and a relatively long follow-up period, and the first to use propensity matching to assess the differences in prognostic endpoints between RV apical and septal pacing. This study indicated no benefit of RV septal pacing with regard to the primary combined endpoint of all-cause death and heart failure hospitalization, as well as all-cause death and heart failure hospitalization when assessed separately. These results were consistent between the two statistical methods used in the analysis; propensity-matched analysis and Cox regression analysis of the whole cohort, strengthening the reliability of the results obtained. 4.2. Comparison with previous studies Previous studies comparing RV apical vs. septal pacing yielded variable results. Most studies assessed the effects on left ventricular function, parameters of neurohormonal activation, and heart failure severity in limited numbers of patients. There have been few prognostic endpoint studies. A systematic review and meta-analysis by Shimony et al. [12], including 14 randomized controlled trials with 754 patients, indicated better LVEF at end of follow-up especially in studies with more than 1 year of follow-up, and in studies including patients with LVEF b 40%–45%. However, the authors concluded that the results were limited, variable, and inconclusive with regard to clinical and prognostic endpoints, such as exercise capacity, functional class, quality of life, and survival. As low % ventricular pacing and short follow-up
4.3. Study limitations Our study has some limitations: This was a retrospective observational study performed at two community hospitals in Japan, including all pacemaker-implanted patients with variable etiology, pacing dependency, cardiac function, and operator experience. The precise site of the RV septal pacing lead was not confirmed by imaging studies, such as echocardiography or computed tomography. However, left anterior oblique and right anterior oblique projection, which is currently the standard for confirming the septal location of the RV lead during the procedure [2], was used in all patients. The paced QRS intervals were calculated from ECG after implantation retrospectively without a set protocol in the mode of pacing for calculation, and may have included fused QRS. The % ventricular pacing, which is an important variable that may significantly affect the results of our study was not assessed, as these data were lacking in a cohort of patients with pacemakers from certain manufacturers and patient backgrounds, and the fusion beats could not be eliminated. We used PS matching to adjust for known confounders; however, unknown and/or unmeasured confounders were not adjusted, and this may have affected the results. This study cannot lead to any final conclusion in regard to the prognostic benefit of septal pacing. Large-scale multi-center prospective randomized controlled trial with sufficient power for clinical prognostic endpoints are needed to confirm the results of our study. Many factors other than prognostic parameters (the procedure and fluoroscopic time during implantation, tricuspid valve function, ease of extraction, etc.) may influence the pacing site preference, but were not assessed in this study. Clinically important follow-up data such as NYHA class and LVEF were unavailable.
218
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218
5. Conclusions In conclusion, in this large-scale retrospective cohort study with propensity matching, RV septal pacing did not show any mediumterm advantage compared to RV apical pacing with regard to all-cause death and hospitalization due to heart failure in a generalized population of patients with predominantly preserved EF.
[6]
[7]
[8]
Disclosures Dr. Matsue received lecture honoraria from Otsuka pharmaceutical. Dr. Suzuki received lecture honoraria from Otsuka Pharmaceutical, Bayer Pharma AG, and Fukuda Denshi. Dr. Nogami received lecture honoraria from St. Jude Medical and Boston Scientific; and an endowment from Medtronic and Johnson & Johnson. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
[9]
[10] [11]
[12]
Grant support [13]
None.
[14]
Acknowledgements
[15]
Source of funding: None. Data collection/analysis: Drs. Maki Ono, Masao Soeda, Shunsuke Kuroda, Masahiro Hoshino, Ryo Ninomiya, Kenji Yoshioka, Yuji Matsuda, Yoshiyasu Takeda, Yuichi Hanaki, Yasutoshi Shinoda, and Tagayasu Anzai.
[16]
References
[18]
[1] S. Furman, J.B. Schwedel, An intracardiac pacemaker for Stokes-Adams seizures, N. Engl. J. Med. 261 (1959) 943–948. [2] R.J. Hillock, H.G. Mond, Pacing the right ventricular outflow tract septum: time to embrace the future, Europace 14 (2012) 28–35. [3] A.M. Gillis, Optimal pacing for right ventricular and biventricular devices: minimizing, maximizing, and right ventricular/left ventricular site considerations, Circ. Arrhythm. Electrophysiol. 7 (2014) 968–977. [4] M.A. Lee, M.W. Dae, J.J. Langberg, J.C. Griffin, M.C. Chin, W.E. Finkbeiner, et al., Effects of long-term right ventricular apical pacing on left ventricular perfusion, innervation, function and histology, J. Am. Coll. Cardiol. 24 (1994) 225–232. [5] M.O. Sweeney, A.S. Hellkamp, K.A. Ellenbogen, A.J. Greenspon, R.A. Freedman, K.L. Lee, et al., Adverse effect of ventricular pacing on heart failure and atrial fibrillation
[17]
[19]
[20]
[21]
among patients with normal baseline QRS duration in a clinical trial of pacemaker therapy for sinus node dysfunction, Circulation 107 (2003) 2932–2937. L.F. Tops, M.J. Schalij, J.J. Bax, The effects of right ventricular apical pacing on ventricular function and dyssynchrony implications for therapy, J. Am. Coll. Cardiol. 54 (2009) 764–776. F. Victor, C. Leclercq, P. Mabo, D. Pavin, A. Deviller, C. de Place, et al., Optimal right ventricular pacing site in chronically implanted patients: a prospective randomized crossover comparison of apical and outflow tract pacing, J. Am. Coll. Cardiol. 33 (1999) 311–316. E. Occhetta, M. Bortnik, A. Magnani, G. Francalacci, C. Piccinino, L. Plebani, et al., Prevention of ventricular desynchronization by permanent para-Hisian pacing after atrioventricular node ablation in chronic atrial fibrillation: a crossover, blinded, randomized study versus apical right ventricular pacing, J. Am. Coll. Cardiol. 47 (2006) 1938–1945. C.M. Yu, J.Y. Chan, Q. Zhang, R. Omar, G.W. Yip, A. Hussin, et al., Biventricular pacing in patients with bradycardia and normal ejection fraction, N. Engl. J. Med. 361 (2009) 2123–2134. C. Medi, H.G. Mond, Right ventricular outflow tract septal pacing: long-term followup of ventricular lead performance, Pacing Clin. Electrophysiol. 32 (2009) 172–176. L. Molina, R. Sutton, W. Gandoy, N. Reyes, S. Lara, F. Limon, et al., Medium-term effects of septal and apical pacing in pacemaker-dependent patients: a double-blind prospective randomized study, Pacing Clin. Electrophysiol. 37 (2014) 207–214. A. Shimony, M.J. Eisenberg, K.B. Filion, G. Amit, Beneficial effects of right ventricular non-apical vs. apical pacing: a systematic review and meta-analysis of randomizedcontrolled trials, Europace 14 (2012) 81–91. J.C.S.J.W. Group, Guidelines for Non-Pharmacotherapy of Cardiac Arrhythmias (JCS 2011), Circ. J. 77 (2013) 249–274. H.G. Mond, R.J. Hillock, I.H. Stevenson, A.D. McGavigan, The right ventricular outflow tract: the road to septal pacing, Pacing Clin. Electrophysiol. 30 (2007) 482–491. S. Matsuo, E. Imai, M. Horio, Y. Yasuda, K. Tomita, K. Nitta, et al., Revised equations for estimated GFR from serum creatinine in Japan, Am. J. Kidney Dis. 53 (2009) 982–992. P.C. Austin, Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies, Pharm. Stat. 10 (2011) 150–161. P.C. Austin, Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples, Stat. Med. 28 (2009) 3083–3107. A. Mizukami, Y. Matsue, Y. Naruse, S. Kowase, K. Kurosaki, M. Suzuki, et al., Implications of Right Ventricular Septal Pacing for Medium-Term Prognosis: Kaplan–Meier Survival Analysis and Cox Regression Analyses of the Data from the Total Cohort, 2016 (Data in Brief submitted for publication). G.C. Kaye, N.J. Linker, T.H. Marwick, L. Pollock, L. Graham, E. Pouliot, et al., Effect of right ventricular pacing lead site on left ventricular function in patients with highgrade atrioventricular block: results of the Protect-Pace study, Eur. Heart J. (2014). A.B. Curtis, S.J. Worley, P.B. Adamson, E.S. Chung, I. Niazi, L. Sherfesee, et al., Biventricular pacing for atrioventricular block and systolic dysfunction, N. Engl. J. Med. 368 (2013) 1585–1593. X. Gong, Y. Su, W. Pan, J. Cui, S. Liu, X. Shu, Is right ventricular outflow tract pacing superior to right ventricular apex pacing in patients with normal cardiac function? Clin. Cardiol. 32 (2009) 695–699.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Implications of right ventricular septal pacing for medium-term prognosis: Propensity-matched analysis Akira Mizukami a,⁎,1, Yuya Matsue a,1, Yoshihisa Naruse b,1, Shinya Kowase c,1, Kenji Kurosaki c,1, Makoto Suzuki a,1, Akihiko Matsumura a,1, Akihiko Nogami b,1, Kazutaka Aonuma b,1, Yuji Hashimoto a,1 a b c
Department of Cardiology, Kameda Medical Center, 929 Higashi-cho, Kamogawa, Chiba 296-8602, Japan Cardiovascular Division, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan Department of Cardiovascular Medicine, Yokohama Rosai Hospital, 3211 Kozukue-cho, Kohoku-Ku, Yokohama, Kanagawa 222-0036, Japan
a r t i c l e
i n f o
Article history: Received 24 January 2016 Received in revised form 15 May 2016 Accepted 26 June 2016 Available online 28 June 2016 Keywords: RV septal pacing RV non-apical pacing Prognosis Heart failure Mortality Propensity matching
a b s t r a c t Background: The effect of right ventricular (RV) septal pacing as opposed to RV apical pacing on prognosis in patients undergoing pacemaker implantation remains controversial. This study was performed to examine the clinical efficacy of RV septal pacing in a large cohort with medium-term follow-up and propensitymatched analysis. Methods: A total of 982 consecutive patients with first pacemaker implantation between 2008 and 2013 at two centers in Japan (51.4% male, age 76.1 ± 10.6 years, 64.3% septal pacing, 94% preserved ejection fraction [EF]) were enrolled. Propensity matching successfully matched 446 patients into RV septal and apical pacing groups. The primary endpoint, a combination of all-cause death and hospitalization due to heart failure, was compared between the two groups. Results: In the propensity-matched cohort, the primary endpoint was observed in 61 patients (13.7%) over a median follow-up period of 2.1 years (interquartile range, 1.1–3.5 years). The effects of septal pacing on prognosis were not statistically significant (hazard ratio [HR] = 1.10, 95% confidence interval [CI] = 0.60–2.04, P = 0.752). No significant benefit of septal pacing was observed on all-cause death (HR = 1.86, 95%CI = 0.74–4.66, P = 0.187) and heart failure hospitalization (HR = 0.93, 95%CI = 0.44–1.98, P = 0.847) when assessed separately. Conclusion: Septal pacing did not show medium-term advantages in prognosis in this large-scale retrospective cohort study with propensity matching of patients with predominantly preserved EF. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The right ventricular (RV) apex has long been the preferred ventricular pacing site for transvenous cardiac pacemakers due to technical ease of implantation, good stability with both passive and active fixation leads, and reliability based on its history with substantial clinical data [1–3]. However, the non-physiological ventricular activation pattern caused by RV apical pacing produces interventricular and intraventricular dyssynchrony, resulting in decreased cardiac output, increased myocardial workload and oxygen consumption, and changes in neurohormonal and electrophysiological activity, resulting in ventricular remodeling. This may lead to left ventricular systolic and diastolic dysfunction, heart failure, and atrial fibrillation [4–6].
Due to these deleterious effects of RV apical pacing and dyssynchrony, there have been attempts to identify a better target site. Possible targets include RV outflow tract, RV septum, His bundle, left ventricular, and biventricular pacing [7–9]. RV outflow tract and septal pacing are attractive options due to the relative technical ease of lead implantation compared to other techniques, with leads traditionally used for RV apical pacing [10]. However, observational studies and small-scale randomized controlled trials failed to confirm its prognostic benefit, although electrocardiographic, hemodynamic, echocardiographic, and neurohormonal benefits have been reported [2,11,12]. This study was performed to examine the clinical efficacy of RV septal pacing in a large cohort study with medium-term follow-up and propensity-matched analysis. 2. Methods
⁎ Corrresponding author at: Department of Cardiology, Kameda Medical Center, 929 Higashi-cho, Kamogawa, Chiba 296-8602, Japan. E-mail address: [email protected] (A. Mizukami). 1 All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
http://dx.doi.org/10.1016/j.ijcard.2016.06.250 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
2.1. Study population This study retrospectively included 982 consecutive patients receiving their first pacemaker implantation with RV lead placement between January 2008 and
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218 December 2013 at two centers in Japan (Kameda Medical Center and Yokohama Rosai Hospital; 51.4% male, age 76.1 ± 10.6 years, 64.3% septal pacing). The indications for pacemaker implantation were decided according to the guidelines of the Japanese Circulation Society [13]. The target site of RV lead placement was decided by the caring physician on the bases of patient background and operator preference. The location of the RV lead was assessed at the time of implantation by right anterior oblique and left anterior oblique fluoroscopic projections, as well as paced QRS morphology during implantation using the methods reported previously [14], and was followed-up by biplane chest radiography and 12-lead ECG after implantation. RV outflow tract pacing was included in the RV septal pacing group. 2.2. Outcome analysis The primary endpoint of this study was a combination of all-cause death and hospitalization due to heart failure. The secondary endpoints included the individual components of the primary endpoint, as well as adverse events related to the implantation procedure, including lead perforation and dislodgement. Data at the time of implantation procedure were collected, including age, sex, diagnosis for implantation (AV block, sick sinus syndrome [SSS], or others), past history (hypertension, hyperlipidemia, diabetes mellitus, heart failure, atrial fibrillation, and ischemic heart disease), medications (beta-blockers, angiotensin converting enzyme inhibitors/angiotensin receptor blockers, and calcium channel blockers), ECG parameters (QRS interval, presence of complete left bundle branch block [CLBBB]), laboratory parameters (hemoglobin, estimated glomerular filtration rate [eGFR], and B-type natriuretic peptide [BNP]), and left ventricular ejection fraction (LVEF) on transthoracic echocardiography. The diagnosis of AV block included any degree of AV block with indication for pacemaker implantation. Hypertension, hyperlipidemia, and diabetes mellitus were scored based on the previous diagnosis and initiation for therapy. Heart failure, atrial fibrillation, and ischemic heart disease were scored based on previous history. The Modification of Diet in Renal Disease (MDRD) study equation with Japanese coefficient was used to calculate eGFR. This new Japanese equation is currently recommended by the Japanese Society of Nephrology for accuracy in the Japanese population [15]. Latest follow-up date and value of BNP were also obtained. Data regarding outcome were obtained by a single investigator who was unaware of the patients' information, including RV pacing site. Adverse events related to the ventricular lead implantation procedure were also assessed and compared between the two groups. The number and rate of micro- and macro-lead dislodgements and lead perforation were compared between the two pacing sites. Other complications related to device implantation were not assessed. Ethical approval was obtained from the institutional review board of each participating hospital. 2.3. Statistical analysis Continuous variables are expressed as the means ± standard deviation or medians (interquartile range [IQR]). All categorical variables are presented as the number and percentage in each group. For the whole cohort, comparisons of continuous variables were tested by Student's t test or Mann–Whitney U-test according to the data distribution. Comparisons of categorical variables were compared by chi-square analysis or Fisher's exact test. In propensity-matched analysis, the variables included in the logistic regression model for calculation of the propensity score (PS) included age, diagnosis for implantation, history of hypertension, history of heart failure, history of atrial fibrillation, presence of CLBBB, QRS interval duration, LVEF b 50%, hemoglobin, and eGFR. Matching was performed using the logit-transformed PS. An optimal-matching algorithm with a caliper width of 0.2 standard deviations of the logit of the PS was used [16]. The balance of the variables between the two groups was judged by standardized differences, and standardized difference b 10% was taken to be satisfactory [16,17]. If the standardized differences for all covariates were b 10%, PS matching was generally considered successful, i.e., two groups are well-balanced as if they were randomized at least in the observed covariates. “Time 0” for survival analyses was the date of pacemaker implantation. For the total cohort, comparison of the probability of freedom from the prognostic binary endpoints between groups was performed by Kaplan–Meier survival analysis with estimation of the hazard ratio from a Cox regression model. Univariate and multivariate Cox regression analyses were also performed to determine the prognostic implications of each variable, including RV septal pacing, on the endpoints. Variables with P b 0.1 on univariate analysis were entered into multivariate analysis. Logarithmic transformations of the variables without a normal distribution were used for Cox regression analysis. The estimates of the parameters were given with their 95% confidence intervals. All P-values reported are 2-sided, and P b 0.05 was considered statistically significant. All statistical analyses were performed with R (The R Foundation for Statistical Computing, Vienna, Austria, version 3.1.1).
3. Results 3.1. Patient characteristics Between January 2008 and December 2013, a total of 982 patients (51.4% male, age 76.1 ± 10.6 years) undergoing their first pacemaker implantation were included in this analysis. The RV lead was placed at the septum in 631 (64.3%) patients, and the apex in 351 (35.7%)
215
patients. LVEF was preserved (≥50%) in 94% of the patients. The patient variables of the study population before and after PS matching for each site of RV lead placement are shown and compared in Table 1. Before PS matching, patients in the apical pacing group were older, lower proportion of males, less SSS, and more AV block as diagnosis for implantation compared to patients in the septal pacing group. There was less history of atrial fibrillation and ischemic heart disease, wider QRS complex, more CLBBB, less hemoglobin, less eGFR, and less LVEF at baseline in the apical pacing group than the septal pacing group. PS matching successfully matched 446 patients into two groups with 223 patients in each group. After matching, the baseline characteristics of the two groups were well balanced with standardized difference of all baseline variables between matched groups b 10% (Table 1). 3.2. Paced QRS intervals Paced QRS intervals were compared between the two groups after pacemaker implantation in 828 patients for whom records of ECG with RV pacing immediately after implantation were available. In the whole cohort, the paced QRS interval of the septal pacing group was shorter than that of the apical pacing group, but the difference was small and not statistically significant (153.7 ± 17.6 ms vs. 155.3 ± 23.3 ms, P = 0.306). Records of RV paced ECG immediately after implantation were available for 384 patients in the propensity-matched cohort. The differences were even smaller and also not significant in the propensity-matched cohort (153.6 ± 17.4 ms, vs. 153.9 ± 24.5 ms, P = 0.891). Although obvious fusion and pseudo-fusion ECGs were excluded, fusion with intrinsic QRS could not be totally eliminated due to the retrospective nature of this study. 3.3. Primary outcome In the propensity-matched cohort of 446 patients, the primary endpoint (combined endpoint of all-cause death and hospitalization due to heart failure) was observed in 61 patients (13.7%) over a median follow-up of 2.1 years (IQR = 1.1–3.5 years). No improvement of prognosis with septal pacing was observed in the matched cohort (HR = 1.10, 95%CI = 0.60–2.04, P = 0.752, Fig. 1). 3.4. Secondary outcome All-cause death and hospitalization due to heart failure were also assessed separately to further clarify the effects of septal pacing. Allcause death was observed in 32 patients (7.2%) in the propensitymatched cohort, and RV septal pacing showed no significant effect (HR = 1.86, 95%CI = 0.74–4.66, P = 0.187, Fig. 2). In the propensity-matched cohort with heart failure hospitalization in 40 patients (9.0%), no improvement of heart failure hospitalization was observed with RV septal pacing (HR = 0.93, 95%CI = 0.44–1.98, P = 0.847, Fig. 3). Data regarding the changes in BNP values on admission and follow-up ([admission value] - [follow-up value]) were also collected, and was compared between the two pacing sites. The median interval of the admission and the follow-up tests were 387 days (IQR = 15–919 days), without significant differences between the septal and apical pacing groups (median 372 days [IQR = 30–776 days] vs. median 400 days [IQR = 10–978 days], P = 0.639). There were no statistically significant differences in the changes of the BNP values between the two groups (− 26 pg/ml [IQR = − 225–10 pg/ml] vs. -30 pg/ml [IQR = − 197–17 pg/ml, P = 0.726). 3.5. Data from the total cohort Among 982 patients, the primary endpoint was observed in 159 patients (16.2%) in the total cohort over the median follow-
216
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218
Table 1 Demographic and clinical characteristics of the study population before and after propensity matching. Before matching
Age, years Male sex, n (%) Diagnosis AV block, n (%) Sick sinus syndrome, n (%) Others, n (%) Past history Hypertension, n (%) Hyperlipidemia, n (%) Diabetes mellitus, n (%) Atrial fibrillation, n (%) Heart failure, n (%) Ischemic heart disease, n (%) Medication β-Blocker, n (%) ACE-I/ARB, n (%) Calcium blocker, n (%) Non-paced ECG parameters QRS complex duration, ms CLBBB, n (%) Laboratory parameters Hemoglobin, g/dl Estimated GFR, ml/min/1.73m2 B-type natriuretic peptide, pg/ml Echocardiographic parameters Left ventricular EF,% LVEF ≥ 50%, n (%)
After matching
Apical pacing
Septal pacing
(n = 351)
(n = 631)
78 ± 10 162 (46)
75 ± 11 343 (54)
196 (56) 138 (39) 17 (5)
P-value
Apical pacing
Septal pacing
Std. mean difference
(n = 223)
(n = 223)
Before
After
b0.001 0.014
78 ± 9 100 (45)
78 ± 9 105 (47)
29.96 16.05
0.22 4.01
277 (44) 308 (49) 46 (7)
b0.001 0.005 0.134
130 (58) 92 (41) 8 (4)
120 (54) 103 (46) 11 (5)
24.15 20.21 8.38
8.05 9.93 4.95
208 (59) 96 (27) 76 (22) 100 (28) 71 (20) 48 (14)
410 (65) 214 (34) 159 (25) 240 (38) 112 (18) 136 (22)
0.075 0.034 0.212 0.003 0.339 0.003
132 (59) 60 (27) 47 (21) 61 (27) 39 (18) 31 (14)
139 (62) 68 (30) 49 (22) 65 (29) 42 (19) 31 (14)
12.38 6.39 7.09 21.31 5.10 21.08
6.13 6.63 2.43 4.43 5.17 0.00
71 (20) 160 (46) 151 (43)
149 (24) 327 (52) 286 (45)
0.223 0.061 0.486
41 (18) 106 (48) 100 (45)
45 (20) 116 (52) 108 (48)
9.67 12.01 4.02
5.06 7.98 6.01
111 (97–143) 42 (12)
100 (90–128) 36 (6)
b0.001 b0.001
110 (97–145) 24 (11)
114 (92–144) 23 (10)
42.94 21.13
7.53 3.26
12 ± 2 55 (40–70) 190 (81–448)
13 ± 2 60 (44–71) 161 (78–349)
b0.001 0.004 0.086
12 ± 2 56 (42–70) 154 (69–381)
12 ± 2 58 (41–69) 168 (84–385)
30.13 20.63 9.48
2.02 0.95 7.08
69 (63–74) 294 (94)
70 (63–77) 483 (94)
0.030 0.877
70 (64–75) 214 (96)
71 (63–78) 213 (96)
13.25 1.10
8.28 0.49
Values are reported as the means ± standard deviation, median (interquartile range), or n (%). ACE-I, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; CLBBB, complete left bundle branch block; GFR, glomerular filtration rate; EF, ejection fraction.
up period of 2.2 years (IQR, 1.1–3.8 years). Superiority of septal pacing was not observed in Kaplan–Meier survival analysis (Log-rank, P = 0.094, Fig. 1 in Ref [18]), as well as multivariate Cox regression analysis (HR = 1.35, 95%CI = 0.90–2.04, P = 0.149, Table 1 in Ref [18]), consistent with the results of propensity-matched analysis. All-cause death occurred in 104 patients (10.6%) in the whole cohort. Septal pacing failed to prove the prognostic benefit in Kaplan–
Meier survival analysis (Log-rank, P = 0.079, Fig. 2 in Ref [18]) and multivariate Cox regression analysis (HR = 0.86; 95%CI = 0.0.52– 1.47, P = 0.616). Hospitalization due to heart failure occurred in 78 patients (7.9%) in the whole cohort. Septal pacing did not show any advantage in Kaplan–Meier survival analysis (Log-rank, P = 0.682, Fig. 3 in Ref [18]) or univariate Cox regression analysis (HR = 0.91, 95%CI = 0.57–1.44, P = 0.682).
Fig. 1. Kaplan–Meier curves for combined primary endpoint of all-cause death and heart failure hospitalization of propensity-matched cohort No significant difference was observed between the two pacing sites.
Fig. 2. Kaplan–Meier curves for all-cause death of propensity-matched cohort No significant difference was observed between the two pacing sites.
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218
Fig. 3. Kaplan–Meier curves for heart failure hospitalization of propensity-matched cohort No significant difference was observed between the two pacing sites.
3.6. Adverse events There were no statistically significant differences in incidence rates of ventricular lead macro-dislodgement (4 [1.1%] vs. 6 [1.0%] patients, P = 0.778), lead micro-dislodgement (4 [1.1%] vs. 6 [1.0%] patients, P = 0.778), or lead perforation (1 [0.3%] vs. 2 [0.3%] patients, P N 0.999) between the apical and septal pacing groups. 4. Discussion
217
duration (especially b 1 year) have been argued as reasons for negative results in some studies, the Protect-Pace trial investigators recently reported the results of an international multicenter randomized controlled trial in 240 patients with high-grade AV block requiring N 90% ventricular pacing with baseline LVEF N 50% over a follow-up period of 2 years [19]. There were no significant differences in the primary endpoint of intra-patient changes in LVEF, heart failure hospitalization, mortality, burden of atrial fibrillation, or plasma brain natriuretic peptide levels between the RV apical and RV high septal lead position. Although this study was not powered to detect the prognostic endpoints, this result suggests that septal pacing is unlikely to produce a positive prognostic effect even in highly pacemaker-dependent patients if LVEF is preserved. The results of our study, which is one of the largest cohort studies and first to use propensity matching, were consistent with the Protect-Pace trial regarding the lack of an effect of septal pacing on mortality and heart failure hospitalization, even with a relatively long follow-up period. It should be emphasized that our study population consisted mainly of patients with preserved EF (LVEF ≥ 50% in 96% of PS-matched cohort and 94% of the whole cohort), similar to the Protect-Pace study cohort. More favorable results may be obtained in patients with reduced cardiac performance considering the results of the above systematic review and metaanalysis by Shimony et al. [12], but definitive evidence is still lacking. Moreover, the BLOCK HF trial clearly showed that in pacemakerdependent patients with heart failure and LVEF ≤ 50%, cardiac resynchronization therapy was superior to RV pacing alone with regard to prognosis and negative remodeling [20]. The differences in paced QRS intervals between RV apical and septal pacing groups were inconsistent in previous studies, with some reports of shorter QRS intervals with RV septal pacing, and others without differences, which may theoretically affect the efficacy of septal pacing [11,21]. In this study, no differences were observed between the two pacing sites.
4.1. Main findings There have been no large-scale randomized controlled trials to determine the efficacy of RV septal pacing on the medium to long term prognostic clinical endpoint in patients with pacemaker indications. This is one of the largest cohort studies with a population of 982 patients (446 matched patients) and a relatively long follow-up period, and the first to use propensity matching to assess the differences in prognostic endpoints between RV apical and septal pacing. This study indicated no benefit of RV septal pacing with regard to the primary combined endpoint of all-cause death and heart failure hospitalization, as well as all-cause death and heart failure hospitalization when assessed separately. These results were consistent between the two statistical methods used in the analysis; propensity-matched analysis and Cox regression analysis of the whole cohort, strengthening the reliability of the results obtained. 4.2. Comparison with previous studies Previous studies comparing RV apical vs. septal pacing yielded variable results. Most studies assessed the effects on left ventricular function, parameters of neurohormonal activation, and heart failure severity in limited numbers of patients. There have been few prognostic endpoint studies. A systematic review and meta-analysis by Shimony et al. [12], including 14 randomized controlled trials with 754 patients, indicated better LVEF at end of follow-up especially in studies with more than 1 year of follow-up, and in studies including patients with LVEF b 40%–45%. However, the authors concluded that the results were limited, variable, and inconclusive with regard to clinical and prognostic endpoints, such as exercise capacity, functional class, quality of life, and survival. As low % ventricular pacing and short follow-up
4.3. Study limitations Our study has some limitations: This was a retrospective observational study performed at two community hospitals in Japan, including all pacemaker-implanted patients with variable etiology, pacing dependency, cardiac function, and operator experience. The precise site of the RV septal pacing lead was not confirmed by imaging studies, such as echocardiography or computed tomography. However, left anterior oblique and right anterior oblique projection, which is currently the standard for confirming the septal location of the RV lead during the procedure [2], was used in all patients. The paced QRS intervals were calculated from ECG after implantation retrospectively without a set protocol in the mode of pacing for calculation, and may have included fused QRS. The % ventricular pacing, which is an important variable that may significantly affect the results of our study was not assessed, as these data were lacking in a cohort of patients with pacemakers from certain manufacturers and patient backgrounds, and the fusion beats could not be eliminated. We used PS matching to adjust for known confounders; however, unknown and/or unmeasured confounders were not adjusted, and this may have affected the results. This study cannot lead to any final conclusion in regard to the prognostic benefit of septal pacing. Large-scale multi-center prospective randomized controlled trial with sufficient power for clinical prognostic endpoints are needed to confirm the results of our study. Many factors other than prognostic parameters (the procedure and fluoroscopic time during implantation, tricuspid valve function, ease of extraction, etc.) may influence the pacing site preference, but were not assessed in this study. Clinically important follow-up data such as NYHA class and LVEF were unavailable.
218
A. Mizukami et al. / International Journal of Cardiology 220 (2016) 214–218
5. Conclusions In conclusion, in this large-scale retrospective cohort study with propensity matching, RV septal pacing did not show any mediumterm advantage compared to RV apical pacing with regard to all-cause death and hospitalization due to heart failure in a generalized population of patients with predominantly preserved EF.
[6]
[7]
[8]
Disclosures Dr. Matsue received lecture honoraria from Otsuka pharmaceutical. Dr. Suzuki received lecture honoraria from Otsuka Pharmaceutical, Bayer Pharma AG, and Fukuda Denshi. Dr. Nogami received lecture honoraria from St. Jude Medical and Boston Scientific; and an endowment from Medtronic and Johnson & Johnson. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
[9]
[10] [11]
[12]
Grant support [13]
None.
[14]
Acknowledgements
[15]
Source of funding: None. Data collection/analysis: Drs. Maki Ono, Masao Soeda, Shunsuke Kuroda, Masahiro Hoshino, Ryo Ninomiya, Kenji Yoshioka, Yuji Matsuda, Yoshiyasu Takeda, Yuichi Hanaki, Yasutoshi Shinoda, and Tagayasu Anzai.
[16]
References
[18]
[1] S. Furman, J.B. Schwedel, An intracardiac pacemaker for Stokes-Adams seizures, N. Engl. J. Med. 261 (1959) 943–948. [2] R.J. Hillock, H.G. Mond, Pacing the right ventricular outflow tract septum: time to embrace the future, Europace 14 (2012) 28–35. [3] A.M. Gillis, Optimal pacing for right ventricular and biventricular devices: minimizing, maximizing, and right ventricular/left ventricular site considerations, Circ. Arrhythm. Electrophysiol. 7 (2014) 968–977. [4] M.A. Lee, M.W. Dae, J.J. Langberg, J.C. Griffin, M.C. Chin, W.E. Finkbeiner, et al., Effects of long-term right ventricular apical pacing on left ventricular perfusion, innervation, function and histology, J. Am. Coll. Cardiol. 24 (1994) 225–232. [5] M.O. Sweeney, A.S. Hellkamp, K.A. Ellenbogen, A.J. Greenspon, R.A. Freedman, K.L. Lee, et al., Adverse effect of ventricular pacing on heart failure and atrial fibrillation
[17]
[19]
[20]
[21]
among patients with normal baseline QRS duration in a clinical trial of pacemaker therapy for sinus node dysfunction, Circulation 107 (2003) 2932–2937. L.F. Tops, M.J. Schalij, J.J. Bax, The effects of right ventricular apical pacing on ventricular function and dyssynchrony implications for therapy, J. Am. Coll. Cardiol. 54 (2009) 764–776. F. Victor, C. Leclercq, P. Mabo, D. Pavin, A. Deviller, C. de Place, et al., Optimal right ventricular pacing site in chronically implanted patients: a prospective randomized crossover comparison of apical and outflow tract pacing, J. Am. Coll. Cardiol. 33 (1999) 311–316. E. Occhetta, M. Bortnik, A. Magnani, G. Francalacci, C. Piccinino, L. Plebani, et al., Prevention of ventricular desynchronization by permanent para-Hisian pacing after atrioventricular node ablation in chronic atrial fibrillation: a crossover, blinded, randomized study versus apical right ventricular pacing, J. Am. Coll. Cardiol. 47 (2006) 1938–1945. C.M. Yu, J.Y. Chan, Q. Zhang, R. Omar, G.W. Yip, A. Hussin, et al., Biventricular pacing in patients with bradycardia and normal ejection fraction, N. Engl. J. Med. 361 (2009) 2123–2134. C. Medi, H.G. Mond, Right ventricular outflow tract septal pacing: long-term followup of ventricular lead performance, Pacing Clin. Electrophysiol. 32 (2009) 172–176. L. Molina, R. Sutton, W. Gandoy, N. Reyes, S. Lara, F. Limon, et al., Medium-term effects of septal and apical pacing in pacemaker-dependent patients: a double-blind prospective randomized study, Pacing Clin. Electrophysiol. 37 (2014) 207–214. A. Shimony, M.J. Eisenberg, K.B. Filion, G. Amit, Beneficial effects of right ventricular non-apical vs. apical pacing: a systematic review and meta-analysis of randomizedcontrolled trials, Europace 14 (2012) 81–91. J.C.S.J.W. Group, Guidelines for Non-Pharmacotherapy of Cardiac Arrhythmias (JCS 2011), Circ. J. 77 (2013) 249–274. H.G. Mond, R.J. Hillock, I.H. Stevenson, A.D. McGavigan, The right ventricular outflow tract: the road to septal pacing, Pacing Clin. Electrophysiol. 30 (2007) 482–491. S. Matsuo, E. Imai, M. Horio, Y. Yasuda, K. Tomita, K. Nitta, et al., Revised equations for estimated GFR from serum creatinine in Japan, Am. J. Kidney Dis. 53 (2009) 982–992. P.C. Austin, Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies, Pharm. Stat. 10 (2011) 150–161. P.C. Austin, Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples, Stat. Med. 28 (2009) 3083–3107. A. Mizukami, Y. Matsue, Y. Naruse, S. Kowase, K. Kurosaki, M. Suzuki, et al., Implications of Right Ventricular Septal Pacing for Medium-Term Prognosis: Kaplan–Meier Survival Analysis and Cox Regression Analyses of the Data from the Total Cohort, 2016 (Data in Brief submitted for publication). G.C. Kaye, N.J. Linker, T.H. Marwick, L. Pollock, L. Graham, E. Pouliot, et al., Effect of right ventricular pacing lead site on left ventricular function in patients with highgrade atrioventricular block: results of the Protect-Pace study, Eur. Heart J. (2014). A.B. Curtis, S.J. Worley, P.B. Adamson, E.S. Chung, I. Niazi, L. Sherfesee, et al., Biventricular pacing for atrioventricular block and systolic dysfunction, N. Engl. J. Med. 368 (2013) 1585–1593. X. Gong, Y. Su, W. Pan, J. Cui, S. Liu, X. Shu, Is right ventricular outflow tract pacing superior to right ventricular apex pacing in patients with normal cardiac function? Clin. Cardiol. 32 (2009) 695–699.