Cardiovascular Outcomes in Patients with Intraventricular Conduction Blocks: a Sixteen-Year Follow-Up in a State-Wide Database John S

Hellenic Society of Cardiology (2017) 58, 194e201

Available online at www.sciencedirect.com ScienceDirect

journal homepage: http://www.journals.elsevier.com/ hellenic-journal-of-cardiology/

ORIGINAL ARTICLE Cardiovascular outcomes in patients with intraventricular conduction blocks: A sixteen-year follow-up in a state-wide database John S. Pantazopoulos, MD*, Alice David, MS, PhD, William J. Kostis, PhD, MD, Nora M. Cosgrove, RN, John B. Kostis, MD for the Myocardial Infarction Data Acquisition System (MIDAS 30) study group

Cardiovascular Institute, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, USA

Received 11 July 2016; received in revised form 14 November 2016; accepted 22 November 2016 Available online 10 December 2016

KEYWORDS Abstract Background: To assess the adverse clinical effects of left anterior hemiblock alone Intraventricular or in combination with right bundle branch block and of complete left bundle branch block in blocks; comparison with isolated right bundle branch block and the relationship of these effects with Mortality; altered mechanoelectric factors resulting in left ventricular dysfunction. Cardiovascular death; Methods: In a 16-year follow-up study using a statewide database, we studied the occurrence Heart failure; of mortal and morbid cardiovascular (CV) events among patients without apparent ischemic Mitral regurgitation; heart disease who had left anterior hemiblock (LAHB, nZ4273, right bundle branch block Atrial ﬁbrillation (RBBB) with LAHB (BFBB, nZ1857) and left bundle branch block (LBBB, nZ9484 compared to isolated RBBB (nZ25288). Results: After adjustment for demographics, co-morbidities and insurance, LAHB was associated with a signiﬁcant excess risk of all-cause death (HR 1.134, 95% CI 1.061-1.213, pZ0.0002) and CV death (HR 1.329, 95% CI 1.174-1.501, p<0.0001). BFBB was associated with excess HF (HR 1.190, 95% CI 1.048-1.351, p<0.0071), all-cause death (HR 1.440, 95% CI 1.045- 1.252, pZ0.0036) and CV death (HR 1.210, 95% CI 1.020-1.436, p<0.0001). LBBB was associated with an excess risk of MR (HR 1.307, 95% CI 1.116-1.530, p<0.0009), HF 1.177, 95% CI1.097-1.263, p<0.0001) and CV death (HR 1.220, 95% CI 1.106-1.345, p<0.0001).

* Corresponding author. John S. Pantazopoulos, M.D., Cardiovascular Institute, Rutgers Robert Wood Johnson Medical School, 125 Paterson Street, CAB-4180A, New Brunswick, NJ 08901, USA. Tel.: 732 235 7685; fax: 732 235 7039. E-mail address: [email protected] (J.S. Pantazopoulos). Peer review under responsibility of Hellenic Society of Cardiology. http://dx.doi.org/10.1016/j.hjc.2016.11.034 1109-9666/ª 2016 Hellenic Society of Cardiology. Publishing services by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Intraventricular conduction block and cardiovascular outcome 195

Conclusions: In patients without apparent ischemic heart disease, the presence of LAHB alone or in combination with RBBB imparts increased risk of CV and all-cause death compared to isolated RBBB. BFBB is also associated with an increased risk of HF. ª 2016 Hellenic Society of Cardiology. Publishing services by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).

1. Introduction algorithm was found to have a sensitivity of 79% and posi- tive predictive value of 99%.5 Intraventricular conduction blocks (ICBs) are associated To obtain an estimate of deaths occurring outside the with adverse cardiovascular outcomes in studies of state of NJ, 7308 randomly selected MIDAS records from the ischemic and non-ischemic cardiomyopathy.1e4 Left bundle years 1986 to 2004 were matched to the National Death branch block (LBBB) is consistently associated with higher Index. Only 14 deaths (0.19% of the sample) occurred out- risk compared to isolated right bundle branch block of-state within 30 days of discharge and 119 (1.63%) (RBBB).3 A suggested explanation for these adverse out- occurred within 1 year of discharge. Deaths occurring in comes has been that ICBs are associated with inter and/or hospice or home hospice were included in the in-hospital intraventricular dyssynchrony and specific regional deaths. contraction abnormalities of the ventricles. It has been Entering the data is the responsibility of the individual proposed that the presence of left anterior hemiblock hospitals. However, state and federal authorities audit the (LAHB) alone or in combination with RBBB is associated with completeness and accuracy of the data. This database was delayed depolarization of the posterobasal/lateral seg- previously audited using a random sample of charts to ments of the left ventricle (LV), leading to adverse conse- verify the accuracy of the information. quences in overall left ventricular function and possibly NJ residents hospitalized outside the state of NJ are not mitral valve competency. Considerable controversy exists included in this analysis. While this introduces some error, regarding specific contraction abnormalities associated the number is probably small. with ICBs, their importance in placing the left ventricular lead during cardiac resynchronization therapy (CRT) and on 2.2. Study Patients the effect of ICBs on long term clinical outcomes. In a statewide database with up to 16 year follow-up We examined the discharge records of 4,407,455 patients (average 5.8 years), we investigated whether LAHB in and identified 51,408 patients with a first admission with a isolation or with RBBB (BFBB), is a discrete marker of higher diagnosis of ICB. Among these patients with ICBs, 10,384 risk compared to isolated RBBB, by examining the occur- were admitted with a diagnosis of IHD during the follow-up rence of mortal and morbid events among 41,024 patients period and were excluded. The remaining 41,024 patients without clinical evidence of ischemic heart disease (IHD). are the subjects of the current study. They were admitted to New Jersey hospitals between 1994 and 2013 with a first 2. Methods admission with a diagnosis of ICB and were over the age of 20 years. Patients with an index diagnosis of ischemic heart disease prior to the index admission for ICB were excluded. 2.1. Data Sources Each patient was followed up to December 31, 2013 or intervening death. Isolated RBBB was chosen as the control We used data from the Myocardial Infarction Data Acquisi- group as early LV depolarization is entirely normal. tion System (MIDAS) database for this study. MIDAS includes the hospital discharge records of patients with myocardial 2.2.1. Study Variables infarction and invasive cardiovascular procedures who were The ICBs were categorized as RBBB without LAD (ICD9 admitted to New Jersey (NJ) non-federal acute care hos- 426.4), LAHB (ICD 462.2), RBBBþLAHB (BFBB; ICD9 426.52) pitals since 1986. It contains abstracted discharge data, and LBBB (ICD9 426.3). LBBB was undifferentiated in terms including the primary reason for admission and up to eight of the presence or absence of LAD. Five separate outcome additional diagnoses, derived from the NJ statewide hos- variables were considered for this study: mitral regurgita- pital uniform billing system. Starting in 1994, the database tion (MR; ICD9 424.0), atrial fibrillation (AF; ICD9 427.31), was made more comprehensive by including all admissions heart failure (HF; ICD9 428.0 to 428.4), cardiovascular and of patients with any cardiovascular diagnosis with or all-cause mortality. The covariates considered were age, without a myocardial infarction. gender, race, health insurance and diagnoses other than On an annual basis, MIDAS data were merged with the NJ ICBs, listed in the abstracted discharge data mentioned Death Registration Files to obtain death information (date under data sources, either as the primary diagnosis or as and cause of death), including out-of-hospital deaths. comorbid conditions. These included hypertension (ICD9 Matching was performed using the Link King, a record 401 to 405), chronic pulmonary disease (ICD9 490 to 508), linkage and consolidation software that integrates a diabetes (ICD9 250), renal disease (ICD9 580 to 589) and deterministic algorithm with a probabilistic algorithm. The cancer (ICD9 140). Patients with a first admission for ICD 196 J.S. Pantazopoulos et al. were not evaluated for the prior occurrence of MR, atrial RBBB). LBBB was present in 9,484 patients; LAHB in 4,273, fibrillation or CHF. RBBB with LAHB in 1,857 and isolated RBBB in 25,288. The study population primarily consisted of white (80.7%), 2.3. Statistical analysis males (50.5%) who were on average 65 years old (63.717.8) and were mostly either on Medicare (44.2%) or We compared patient characteristics across ICB categories. a commercial insurance (47.3%). They were followed for up Event rates were calculated using Cox proportional hazards to 16 years with an average follow-up time of 5.8 years. regression models. In each Cox model, event rates were Most of this population had co-morbid conditions, such as compared between each of the study groups (LAHB, hypertension (52.7%), chronic obstructive pulmonary dis- RBBBþLAHB, and LBBB) and the control group (RBBB) using ease (14.9%), diabetes (15.3%), cancer (10.2%) and chronic hazard ratios with 95% confidence intervals. Kaplan-Meier renal disease (4.6%) (Table 1). survival curves for each outcome (MR, AF, HF, all cause Patients with RBBB þ LAHB and those with isolated RBBB death and cardiovascular (CV) death) were plotted over were predominantly male (58.4% and 56.2%, respectively). time separately for the three study groups and the control Those with LBBB were mostly female (66.9%). However, group. The models were adjusted for the covariates listed there was no gender difference among those with LAHB above, as baseline characteristics differed among groups. (51.5% male). Diagnoses included in the abstracted discharge data, exclusive of the outcomes listed above, were not consid- 3.1.1. Outcomes ered. Statistical analyses were performed with the use of The following five outcomes were evaluated: MR, AF, HF, SAS 9.3 software (SAS Institute). cardiovascular and all-cause mortality. The Kaplan-Meier The institutional review boards of the State of New Curves show that those with RBBB þ LAHB had lower sur- Jersey Department of Health and Senior Services and the vival rates for HF, cardiovascular or all-cause mortality. UMDNJ-Robert Wood Johnson Medical School approved the Patients with LBBB had the lowest event-free rates in re- study. gard to MR, HF and CV death. LAHB patients had lower survival rates for cardiovascular and all cause death 3. Results (Figures 1 and 2). Table 2 shows the adjusted hazard ratios for MR, AF, HF, cardiovascular and all-cause mortality. Compared to iso- 3.1. Patient Characteristics lated RBBB patients, MR occurred more frequently among patients with LBBB (HR 1.307, 95% CI: 1.116-1.530, The number of patients with RBBB and left posterior p<0.0009). Regarding AF, none of the study groups were hemiblock (122) was very small and had few events (no MR, associated with a significant excess risk. The risk of heart 9 Afib, 8 HF, 4 CV death and 24 all-cause deaths, all non- failure was significantly higher in two groups (RBBBþLAHB: significantly different compared to patients with isolated HR 1.19 95% CI: 1.048- 1.351, pZ0.0071 and LBBB: HR 1.177

Table 1 Baseline Characteristics Stratiﬁed by BBB. LAHB RBþLA LBBB RBBB RBþLP Total Total 4,273 1,857 9,484 25,288 122 41,024 n%n%n%n%n%n % GENDER Male 2,202 51.5 1,085 58.4 3,143 33.1 14,219 56.2 70 57.4 20,719 50.5 RACE White 3,560 83.3 1,525 82.1 7,885 83.1 20,048 79.3 97 79.5 33,115 80.7 Black 338 9.1 190 10.2 740 7.8 2,508 9.9 8 6.6 3,834 9.3 Other 325 7.6 142 7.6 859 9.1 2,732 10.8 17 13.9 4,075 9.9 HEALTH INSURANCE Commercial 1,728 40.4 649 34.9 3,974 41.9 12,988 51.4 61 50.0 19,400 47.3 Medicare 2,238 52.4 1,103 59.4 4,913 51.8 9,842 38.9 50 41.0 18,146 44.2 Self/Medicaid 307 7.2 105 5.7 597 6.3 2,458 9.7 11 9.0 3.478 8.5 COMORBID CONDITIONS Hypertension 2,327 54.5 1,102 59.3 5,786 61.0 12,357 48.9 57 46.7 21,629 52.7 Diabetes 713 16.7 385 20.7 1,410 14.9 3,765 14.9 22 18.0 6,295 15.3 Cancer 539 12.6 248 13.4 981 10.3 2,397 9.5 15 12.3 4,180 10.2 Renal 197 4.6 150 8.1 435 4.6 1,115 4.4 4 3.3 1,901 4.6 COPD 681 15.9 303 16.3 1,245 13.1 3,843 15.2 22 18.0 6,094 14.9 Age Mean SD 67.415.9 71.515.2 68.315.7 60.818.4 61.516.8 63.717.8 Survival Time 5.95.6 5.15.0 5.85.2 5.24.8 5.85.3 5.45.0 Mean SD Intraventricular conduction block and cardiovascular outcome 197

Figure 1 Survival of Persons by Type of Intraventricular Block. Kaplan-Meier survival curves for all cause mortality by type of intraventricular conduction block. Red denotes LAHB; blue denotes RBBBþLAHB; black denotes LBBB; green denotes RBBB.

95% CI: 1.097- 1.263, p<0.0001). Cardiovascular death was further increased as the QRS duration is prolonged >140 higher in the LAHB (HR 1.329 95% CI: 1.174 -1.505, ms”.6 The prevalence of LAHB in the normal population p<0.0001); RBBBþLAHB (HR 1.210 95% CI: 1.020-1.436) and varies from 0.9% to 6.2%. In our study of a hospital-based LBBB (HR 1.220 95% CI: 1.106-1.345, p<0.0001 for both) population, the overall prevalence was 10.6%, while the groups. For all-cause death, excess risk was identified for prevalence among the non-ischemic group was 10.4%. ICBs the LAHB (HR 1.134 95% CI: 1.061-1.213, pZ0.0002) and are considered to be due to abnormalities of the His Pur- (RBBBþLAHB: HR 1.44 95% CI: 1.045-1.252, pZ0.0036) kinje System (HPS), its distribution and transmission ve- groups. locities. The ventricular conduction system comprises the In summary, in this study of patients without admission main bundle, RBBB, LBBB and the subendocardial periph- for ischemic heart disease, patients with LAHB, BFBB (RBBB eral conduction system, thought to develop from the þ LAD) and LBBB (undifferentiated as to frontal axis) had trabecular ventricular component. Studies in embryology worse outcomes compared to patients with isolated RBBB. indicate that, before the onset of overt chamber formation, an ECG can be recorded. In the early embryonic heart 4. Discussion (stage HH14), no specialized/distinct cells of a conduction system can be recognized.7 Considerable uncertainty exists regarding the clinical significance of isolated LAHB, with one study in 2007 stating 4.1. Myocardial Structure and Function that isolated LAHB does not itself imply a risk factor of cardiac morbidity or mortality, and in a healthy population, The ventricular myocardial band (VMB), a concept devel- it must be regarded as an incidental electrocardiographic oped over the last 50 years, has led to an explanation of the finding, which is not the case in hospital populations. way the heart performs its double function, ejection and Additionally, ICBs have been associated with increased suction of blood. Diffusion-tension MRI studies on intact morbidity and mortality in long-term follow-up studies of human beating hearts have confirmed that myocardial fi- individuals with initially apparently normal cardiac func- bers within the ventricular mass are arranged in layers of tion.7,11,20,21 While LBBB has been associated with the most counter-wound helices, encircling the ventricular cavities. adverse outcomes, in a study that identified the frontal QRS Patterns of contractile activities within the ventricular axis,4 LBBB with LAD was associated with even more myocardium may or may not precisely correspond with adverse outcomes. In the same study, RBBB was found to be patterns of electrical excitation. The apical loop of the VMB associated with similar adverse outcomes as LBBB without is the principal motor of the heart. The ventricles have the LAD. In a recent analysis of the risk of incident HF in the ability to translate 15% linear sarcomere shortening into an ARIC study, it was concluded that “VCDs except for isolated ejection fraction of greater than 50% and wall thickening 8,9 RBBB are strong predictors of incident HF, and HF risk is greater than 30%. 198 J.S. Pantazopoulos et al.

Figure 2 Survival Curves for mitral regurgitation, atrial ﬁbrillation, heart failure and cardiovascular death by type of intraventricular block. Kaplan-Meier survival curves (from top to bottom) for mitral regurgitation, atrial ﬁbrillation, heart failure and cardiovascular death by type of intraventricular conduction block. Red denotes LAHB; blue denotes RBBBþLAHB; black denotes LBBB; green denotes RBBB. Intraventricular conduction block and cardiovascular outcome 199

The most important part of ventricular diastole (i.e., the rapid ﬁlling phase), during which it receives more than 70% of the stroke volume, belongs to the active muscular 0.0002 0.0036 0.2291 contraction of the ascending segment of the apical band. It is widely accepted that electrical coupling between myocardial cells is accomplished by cell-to-cell transmission via the gap junctions.

4.2. Prior Observations CI: 1.061-1.213 CI: 1.045-1.252 CI: 0.979-1.091 Carl Wiggers, in his monograph published in 1928 (chapter 8), discussed “the fractionate nature of ventricular contraction”. He stated that the initial concave rise in LV pressure 0.0001 1.134 0.0001 1.440 0.0001 1.034 < < < may be accounted for by the continued entry of new fractionate contractions. This early phase of isometric contraction, averaging 0.038 sec, he designated “the entrant phase”. Changes in the gradient and amplitude of the total muscular contraction of a cardiac chamber are caused either by variation of the number of fractions excited or by alter- ations of the state of reactivity.10 He concluded that frac- CI: 1.174-1.505 CI: 1.020-1.436 CI: 1.106-1.345 tionate contractions continue after the point arbitrarily adopted (the incisura) as a criterion of the end of mechanical systole.10 In 1930, Katz demonstrated a drop of intraventricular pressure below 0 during the rapid filling phase, 0.0001 1.220 0.2680 1.329 0.0071 1.210 < preceded by a large and rapid increase of the ventricular volume capable of producing effective blood suctioning.9 The direction, magnitude and spatial distribution of early and late vectors on VCG most likely represent the fractional entry into contraction (regional strain) and the amount of the additive muscle mass involved.11 These prior observations are consistent with studies on CI: 0.959-1.161 CI: 1.048-1.351 CI: 1.097-1.263 regional strain and the continuing contraction of segments of the LV during the early diastolic period of rapid filling. Post-systolic shortening was reported predominately in the apical and basal segments of the septal, anteroseptal and 0.9140 1.055 0.4705 1.190 0.3865 1.177 anterior walls, and its duration of approximately 1/6 of the systolic duration results in a relative increase in strain of 19þ6% after aortic valve closure. In high resolution MRI tagging, the first image that can be used for strain analysis has a delay of 39 ms with respect to the R wave (i.e., beyond the isovolemic contraction), thus missing early 12 CI: 0.869-1.135 CI: 0.892-1.281 CI: 0.946-1.155 contraction events (strain). The early diastolic phase is very important for the LV preload, as shown in recent studies of global longitudinal strain and its predictive value of future outcomes.13 Early 0.1868 0.993 0.9871 1.069 0.0009 1.045 diastolic relaxation is an active energy-dependent process that is rapidly initiated in the basal segments of the LV and propagates towards the apex. Diastolic dysfunction is increased by the abnormal timing of contraction of the ascending limb of the apical VMB. A recent study of the significance of abnormal early diastolic strain on outcomes suggests such a role.14 In normal situations, it has been CI: 0.656-1.086 CI: 0.701-1.417 CI: 1.116-1.530 demonstrated that mechanical activation and contraction of the lateral wall occur earlier than in the septum, with a late and protracted peak of contraction of the lateral

n HR P value HR P value HRsegment; P value in the HR septum, it occurs P value HR later and peaks P value earlier. This sequence assures that the lateral wall will not bulge Adjusted Hazard Ratio Deltas of Outcomes Stratiﬁed by BBB. outwards with the early contraction peak of the septum.12 Considering the velocity of conduction in the HP system and

LA 1,857 0.997 the excitation maps vs. timing and the intensity of þ contraction (peak time) of the septum vs. the lateral wall, Table 2 Type of BBB Total MR AFIB HF CV Death Death LAHB 4,273 0.844 RB LBBB 9,484 1.307 200 J.S. Pantazopoulos et al. the signal starting the earlier contraction of the lateral wall The limitations of the study include its retrospective in normal subjects must arrive there via faster mechano- nature, the requirement for hospitalization, coding varia- electric means.12 Our data are similar to other significant tions on the discharge document, a lack of information reports from different ethnic populations.15e17 regarding the presence of left axis deviation in the LBBB group as well as confounders not included in the analysis, 4.3. Mechanical Implications of Left Axis Deviation e.g., mechano-electric feedback, risk factors such as smoking, lipid levels, and medications. An extensive re- 25 Ventricular electrical remodeling (VER) is triggered by view regarding the pathophysiology of LBBB and a case 26 mechanical rather than electrical stimuli via mechano- study by Becker Muscular Dystrophy stressed the impor- electric feedback.18,19 tance of the definition/presence of true LBBB in affecting The time course of ventricular electrical remodeling the response and outcomes of resynchronization therapy. In (VER) antecedes the development of hypertrophy.18 Ven- addition, we did not exclude prior admissions with di- tricular electrical remodeling in late activated segments is agnoses of non-fatal outcomes (MR, AF and HF), although triggered by mechanoelectric feedback induced by the the long follow-up period partially ameliorates this limita- change in cardiac activation sequence. VER involves a tion. The average follow-up of 5.8 years might be too short complex cascade of events that may be initiated by to demonstrate adverse outcomes, which require longer stretch.19 follow-up to 9.0 years. In a previous smaller study, it was Dyssynchronous hearts displayed significantly shorter noted that the survival curves did not diverge until 4 APDs in the lateral wall relative to the anterior wall approximately 5 years of follow-up. (approximately a 16% reduction). Lateral LV segments from LBBB hearts exhibited substantial Cx43 signal lateraliza- 5. Conclusions tion, consistent with VER. Abnormal spatial sequences and/ or the displacement of electrical depolarization lead to In patients without apparent ischemic heart disease, the abnormalities of the sequential strain and strain rate. presence of LAHB alone or in combination with RBBB im- Mechanoelectric factors lead to abnormal sequences of the parts an increased risk of CV and all-cause death compared regional strain that go beyond the local change of initial to isolated RBBB. BFBB is also associated with an increased depolarization, affecting remote areas of the myocardium. risk of HF. This study of patients without evidence of The regional pairing of sections of the ventricular myocar- ischemic heart disease during follow-up raises the issue of dium (in series or in parallel) effects changes and/or im- the significance of mechanoelectric factors in determining 20 pairs both ejection and early suction. regional and sequential ventricular strain and its origins on embryology. The noted adverse outcomes with ICBs in non- 4.3.1. Septal Depolarization ischemic patients reported here suggest that functional The abnormal septal depolarization noted in LBBB and in factors play a role in abnormalities of ventricular pump LAHB becomes evident by studying the direction, rotation function (ejection and suction) and raises the possibility and size of the 10 msec vector, and is the initiating factor of a that the QRS pattern, in addition to QRS duration, is a very specific type of dyssynchrony with LAD the finding on the primary driver of worse outcomes. 11 12 lead ECG. In pure RBBB, the 10 msec vector in RBBB is To the best of our knowledge, this is the largest reported 21 entirely normal. What is now recognized as septal flash, study comparing ICDs to pure RBBB (without LAD), which 22 23 septal beak or apical rocking represents the mechanical has no effect on the initial direction of septal expression of an abnormal direction of early septal depo- depolarization. larization. This is present in LAHB alone and in conjunction with RBBB and LBBB with LAD.24 An abnormal left axis suggests that a particular variant of abnormal septal repolari- Funding zation sets in motion a sequence of regional myocardial strain that affects the early isovolumic contraction phase This study was funded, in part, by the Robert Wood Johnson and also the late strain during early relaxation (during which Foundation. basal segments of LV contract) with implications for the early “suction” phase of LV diastole. Patients with LBBB and a large septal flash (SF) have areas of conduction blocks with a U- Conflict of interest shaped activation pattern on non-contact mapping (NCM), and patients with this pattern have favorable acute and The authors have no conflicts of interest. chronic responses to CRT.21 In patients with a large SF, there may be a block in the proximal left bundle; and accordingly, the endocardium of the LV would be expected to be acti- References vated from the RV with a delay of approximately 40e50 ms, caused by slow muscle-to-muscle trans-septal conduction.21 1. Bussink BE, Holst AG, Jespersen L, Deckers JW, Jensen GB, Prescott E. Right bundle branch block: prevalence, risk factors, The absence of septal flash implies preserved septal activa- and outcome in the general population: results from the tion by the proximal left bundle. It is interesting that 80% of e 21 Copenhagen City Heart Study. Eur Heart J. 2013;34:138 146. the patients with a large septal flash were non-ischemic. 2. Schneider JF, Thomas HE, Kerger BE, McNamara PM, Sorlie P, Apical rocking is more prominent in non-ischemic cardio- Kannel WB. Newly acquired right bundle-branch block. Ann 23 myopathy patients. Intern Med. 1980;92:37e44. Intraventricular conduction block and cardiovascular outcome 201

3. Imanishi R, Seto S, Ichimaru S, Nakashima E, Yano K, Women’s Health Initiative )[WHI]. Am J Cardiol. 2012;110: Adahoshi M. Prognostic significance of incident complete left 1489e1495. bundle branch block observed over a 40-year period. Am J 16. Nakamura Y, Okamura T, Inohara T, Kohsaka S, Watanabe M, Cardiol. 2006;98:644e648. Higashiyama A, et al, for the NIPPON DATA 80 Research Group. 4. Miller WL, Hodge DO, Hammill SC. Association of uncompli- Prognostic values of bundle branch blocks for cardiovascular cated electrocardiographic conduction blocks with subsequent mortality in Japanese (24year follow-up of NIPPON DATA80). J cardiac morbidity in a community-based population (Olmsted Electrocardiol. 2013;46:360e365. County, Minnesota). Am J Cardiol. 2008;101:102e106. 17. Haataja P, Anttila I, Nikus K, Eskola M, Huhtala H, Nieminen T, 5. Link King v7.1.21 is http://www.the-link-king.com/, accessed et al. Prognostic implications of intraventricular conduction December 26, 2015. delays in a general population: The Health 2000 Survey. Ann 6. Zhang Z-M, Rauthaharju PM, Prineas RJ, Loehr L, Rosamond W, Med. 2015;47:74e80. Soliman EZ. Ventricular conduction defects and the risk of 18. Jeyaraj D, Wilson LD, Zhong J, Flask C, Saffitz JE, Deschenes I, incident heart failure in the Atherosclerosis Risk in Commu- et al. Mechanoelectrical feedback as novel mechanism of car- nities (ARIC) Study. J Cardiac Fail. 2015;21:307e312. diac electrical remodeling. Circulation. 2007;115:3145e3155. 7. Moorman AFM, Christoffels VM. Cardiac chamber formation: 19. Ibbus I, Rosenbaum DS. Transmural action potential changes development, genes, and evolution. Physiol Rev. 2003;83: underlying ventricular electrical remodeling. J Cardiovasc 1223e1267. Electrophysiol. 2003;14:394e402. 8. Torrent-Guasp F, Kocica MJ, Corno AF, Komeda M, Carreras- 20. Markhasin VS, Balakin A, Protsenko Y, et al. Activation Costa F, Flotats A, et al. Towards new understanding of the sequence of cardiac muscle in simplified experimental models: heart structure and function. Eur J Cardiothoracic Surg. 2005; relevance for cardiac mechano-electric coupling. In: Kohl P, 27:191e201. Sachs F, Franz MR, eds. Cardiac Mechano-electric Coupling and 9. Torrent-Guasp F, Kocica MJ, Corno A, Komeda M, Cox J, Arrhythmias. 2nd ed. Oxford University Press; 2011. Flotats A, et al. Systolic ventricular filling. Eur J Cardiothoracic 21. Duckett SG, Camara O, Ginks MR, Bostock J, Chinchapatnam P, Surg. 2004;25:376e386. Sermesant M, et al. Relationship between endocardial activa- 10. Wiggers CJ. The Pressure Pulses in the Cardiovascular System. tion sequences define by high-density mapping to early septal London: Longmans, Green and Co.; 1928. contraction (septal flash) in patients with left bundle branch 11. Benchimol A. Vectorcardiography. Baltimore, MD: The Williams block undergoing cardiac resynchronization therapy. Europace. & Wilkins Company; 1973. 2012;14:99e106. 12. Zwanenburg JJM, Gotte MJW, Kuijer JPA, Heethaar RM, van 22. Feigenbaum H. Role of m-mode technique in today’s echo- Rossum AC, Marcus JT. Timing of cardiac contraction in humans cardiography. J Am Soc Echocardiogr. 2010;23:240e257. mapped by high-temporal-resolution MRI tagging: early onset 23. Steelant B, Stankovic I, Roijakkers I, Aarones M, Bogaert J, and late peak of shortening in lateral wall. Am J Physiol Heart Desmet W, et al. The impact of infarct location and extent on Circ Physiol. 2004;286:H1872eH1880. LV motion patterns. Implications for dyssynchrony assessment. 13. Sengelov M, Jorgensen PG, Jensen JS, Bruun NE, Olsen FJ, J Am Coll Cardiol Img. 2016;9:655e664. Fritz-Hansen T, et al. Global longitudinal strainis a superior 24. Strasberg B, Rich S, Lam W, Swiryn S, Bauernfeind R, Rosen KM. predictor of all-cause mortality in heart failure with reduced M-mode echocardiography in left bundle branch block: signif- ejection fraction. JACC: Cardiovasc Imaging. 2015;8: icance of frontal plane QRS axis. Am Heart J. 1982;104: 1351e1359. 775e779. 14. Ersboll M, Andersen MJ, Valeur N, Mogensen UM, Fahkri Y, 25. Sideris S, Trachanas K, Kelesidis I, Gatzoulis K, Arsenos P, Thune JJ, et al. Early diastolic strain rate in relation to systolic Tousoulis D, et al. Cardiac resynchronization therapy: a review and diastolic function and prognosis in acute myocardial of pathophysiology and clinical applications. Hellenic J Car- infarction: a two-dimensional speckle-tracking study. Eur diol. 2015;56:451e460. Heart. 2014;35:648e656. 26. Andrikopoulos G, Kourouklis S, Trika C, Tzeis S, Rassias I, 15. Zhang Z-m, Rautaharju PM, Soliman EZ, Manson JE, Cain ME, Papademetriou C, et al. Cardiac resynchronization therapy in Martin LW, et al. Mortality risk associated with bundle branch becker muscular dystrophy. Hellenic J Cardiol. 2013;54: blocks and related repolarization abnormalities (from the 227e229.