Author + information
- Received June 2, 2016
- Revision received June 28, 2016
- Accepted July 7, 2016
- Published online November 1, 2016.
- Chance M. Witt, MDa,
- Gang Wu, MDa,
- Dachun Yang, MDa,
- David O. Hodge, MSb,
- Veronique L. Roger, MDa and
- Yong-Mei Cha, MDa,∗ ()
- aDivision of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
- bBiostatistics Unit, Mayo Clinic, Jacksonville, Florida
- ↵∗Reprint requests and correspondence
: Dr. Yong-Mei Cha, Division of Cardiovascular Diseases, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905.
Objectives This study aimed to define the prognosis for patients with left bundle branch block (LBBB) and a mildly to moderately reduced left ventricular ejection fraction (LVEF) (36% to 50%) as well as to clarify whether LBBB remained a negative prognostic marker in this group.
Background LBBB is associated with worse outcomes in patients with heart failure in the setting of severely reduced LVEF. The level of morbidity and mortality associated with LBBB in the setting of a mildly to moderately reduced LVEF (36% to 50%) has not been clearly characterized. This knowledge is important to clarify the potential benefit of cardiac resynchronization therapy in this group.
Methods All patients identified as having an LBBB from 1994 to 2014 were included in the study if they had a baseline echocardiogram within 1 year and an LVEF between 36% and 50%. A control group without intraventricular conduction abnormality matched on age, sex, baseline LVEF, and date of echocardiogram was created. Outcomes were compared between the 2 groups.
Results Of 1,436 patients meeting inclusion criteria, 54% were male. Mean age was 67 ± 13 years, and mean LVEF at baseline was 44 ± 4%. There was no difference in baseline heart failure diagnosis between groups. There were significantly higher rates of baseline coronary artery disease in the control group and higher rates of aortic stenosis in the LBBB group. LBBB was associated with significantly worse mortality (hazard ratio [HR]: 1.17; 95% confidence interval [CI]: 1.00 to 1.36), an LVEF drop to 35% or less (HR: 1.34; 95% CI: 1.09 to 1.63), and the need for an implantable cardioverter-defibrillator (HR: 1.50; 95% CI: 1.10 to 2.10). Mortality remained significantly higher in the LBBB group when controlled for heart failure, coronary artery disease, and aortic stenosis (p = 0.04).
Conclusions Patients with a mildly to moderately reduced LVEF and LBBB have poor clinical outcomes that are significantly worse than those for patients without conduction system disease. This group may obtain benefit from cardiac resynchronization therapy and deserves to be studied in prospective trials.
- cardiac resynchronization therapy
- left bundle branch block
- left ventricular systolic function
In patients with heart failure (HF) and a significantly reduced left ventricular ejection fraction (LVEF), left bundle branch block (LBBB) has been associated with an increase in mortality (1). Fortunately, cardiac resynchronization therapy (CRT) has proven to be of benefit in mitigating the dyssynchrony caused by the LBBB (2–5). The proven reduction in mortality, HF symptoms, and HF hospitalizations with CRT has led major guideline groups to recommend this therapy for many patients with HF and an LVEF of 35% or less (6,7).
Current U.S. HF guidelines place an important focus on prevention. This is highlighted by the inclusion of American College of Cardiology/American Heart Association stages A and B HF, which identify potential HF patients without current symptoms (8). As a potentially treatable cause of worsening left ventricular dysfunction, LBBB may be thought of as a target for patients with mild or asymptomatic HF. Data are scarce in the group with a reduced LVEF of 36% to 50% who are not meeting current guideline recommendations for CRT placement (≤35%). Some consider right ventricular pacing to have similar negative pathophysiologic effects to LBBB, which may allow us to extrapolate some data from the BLOCK HF (Biventricular versus Right Ventricular Pacing in Heart Failure Patients with Atrioventricular Block) trial. This study showed that patients with an LVEF <50%, with or without symptoms, requiring high amounts of pacing have improved outcomes with CRT when compared with outcomes associated with typical right ventricular pacing (9).
Interestingly, a recent study aimed at assessing the benefit of CRT in these patients with LVEF of 36% to 50% was stopped early, primarily because of difficulty with enrollment (10). Although the investigators discuss several contributing difficulties, a problem may be a limited knowledge of the expected prognosis for patients in this situation. Further clarification of the severity of outcomes in this group of patients may help guide discussions in clinical practice and in future similar trials.
This study aimed to define the prognosis for patients with LBBB and a mildly to moderately reduced LVEF (36% to 50%) as well as to clarify whether LBBB remained a negative prognostic marker in this group.
All adult patients with a diagnosis of LBBB coded on their electrocardiogram (ECG) at Mayo Clinic in Rochester, Minnesota, from 1994 to 2014 were screened for inclusion in the study. All ECG with an LBBB diagnosis were individually reviewed and confirmed by an electrophysiologist (G.W. or D.Y.). Patients with a diagnosis of LBBB were cross-matched with the Mayo Clinic echocardiography database to search for patients who had an LVEF of 36% to 50% within 1 year of the index ECG LBBB diagnosis. Patients without a baseline transthoracic echocardiogram within 1 year of the index ECG were excluded. To create the desired study group, we excluded patients with an LVEF ≤35% or >50%.
A control group was then created with patients matched in a 1:1 ratio on the basis of age, sex, baseline LVEF, and time of baseline echocardiogram. Potential control patients were excluded if their ECG had a diagnosis of right bundle branch block or a fascicular block as these diagnoses, by definition, could not have occurred in the patients with LBBB (Figure 1).
Outcomes including death, HF hospitalization, sustained ventricular tachycardia/ventricular fibrillation (VT/VF), and implantable cardioverter-defibrillator (ICD) placement were then retrieved from the electronic medical record and device database. Death was verified with the national database. HF hospitalization was defined as a hospitalization with a primary diagnosis of HF. The progression of LVEF was recorded for patients with follow-up echocardiograms. A follow-up echocardiogram was not required for a patient to be included in the study, and therefore not all patients had these data. The follow-up LVEF measurement was primarily used to determine which patients had a drop below 35%, a clinically important variable for consideration of ICD or CRT. These data were also analyzed to examine smaller changes through time. To improve the homogeneity of comparison intervals, the follow-up echocardiograms were divided on the basis of their occurrence at <2 years or between 2 and 5 years after the baseline echocardiogram.
Overall comparisons between the groups were completed with chi-square tests for categorical variables and Wilcoxon rank sum tests for continuous variables. Follow-up events such as overall mortality, LVEF reduced to <35%, HF admission, VT/VF event, and ICD implantation using the Kaplan-Meier method. Overall comparisons between the cases and control patients for these follow-up events were completed using proportional hazards models. All analyses were completed using SAS (version 9.4, SAS Institute Inc., Cary, North Carolina).
Baseline patient characteristics
Between 1994 and 2014, LBBB was diagnosed on ECG in 12,879 individual patients. Of those patients, 5,322 had a transthoracic echocardiogram within 1 year of the index ECG. Patients were included in the study only if their baseline LVEF was >35% and ≤50%. The 1,993 patients with an LVEF ≤35% and the 1,893 with an LVEF >50% were excluded. This left 1,436 patients in the study group, who had a mean LVEF of 44 ± 4%. Patients in the LBBB group were 54% male and had a mean age of 67 ± 13 years.
The control group was matched on age, sex, and baseline LVEF, so there was no difference in these variables between groups. Groups were also assessed for baseline HF or coronary artery disease diagnosis, as well as for previous myocardial infarction or revascularization procedure. Using baseline echocardiogram data, patients were also assessed for moderate or greater aortic stenosis (AS) (Table 1). There were significantly higher rates of coronary artery disease, myocardial infarction, and percutaneous or surgical revascularization in the control group. There were significantly higher rates of AS in the LBBB group, although numbers were low in both groups. There was no difference between groups for baseline HF, and the rates were low overall.
Changes in LVEF
The most clinically valuable data regarding LVEF was a drop to 35% or less, which is included as a clinical outcome. We also analyzed changes in LVEF at follow-up echocardiograms occurring at <2 years and in the period 2 to 5 years after the baseline echocardiogram (Table 2). This analysis showed significant variation in LVEF changes between groups at <2 years, with the largest decreases occurring more commonly in the LBBB group. Similar trends were seen at the longer follow-up interval but were not significant in the setting of smaller numbers of patients.
Kaplan-Meier outcome analysis for LBBB versus control patients showed that 5-year mortality (28% vs. 24%; hazard ratio [HR]: 1.17; 95% confidence interval [CI]: 1.00 to 1.36) was higher in the LBBB group. Five-year LVEF decrease to 35% or less (37% vs. 30%; HR: 1.34; 95% CI: 1.09 to 1.63) and need for ICD (13% vs. 9%; HR: 1.50; 95% CI: 1.10 to 2.10) were also significantly more common in the LBBB group. There was no significant difference in VT/VF event or HF admission rate (Table 3). Figure 2 shows the Kaplan-Meier estimate for probability of 5-year mortality, LVEF drop to 35% or less, ICD implantation, and VT/VF event in LBBB and control groups.
The Kaplan-Meier 20-year probability of survival in both groups is shown in Figure 3. Patients with LVEF of 36% to 50% and LBBB had lower survival rates at 5 years (72% vs. 76%; p = 0.04), 10 years (53% vs. 61%; p < 0.001), and 20 years (27% vs. 46%; p < 0.001) compared with control patients who had no intraventricular conduction delay.
To assess for the contribution of baseline HF, coronary artery disease, or AS to the mortality endpoint, we performed a multivariate analysis. This showed that LBBB was still associated with a significantly higher rate of mortality at 5 years (HR: 1.18; 95% CI: 1.01 to 1.37; p = 0.04).
The novel finding of this study is that patients with LBBB and LVEF of 36% to 50% have exceedingly poor outcomes. The overall data, considered together in the context of their clinical implications, are striking. At 5 years, 59% of patients had died, had further reduction in their LVEF to 35% or less, had a sustained ventricular arrhythmia, or required device implantation. Stated another way, within 5 years a majority of the patients with LBBB had died, had a device placed, or had a clinical event that would typically prompt device placement. This level of mortality and other adverse events identifies a group in which preventive or therapeutic measures could have a high impact on improving prognosis.
Clinical outcome and survival
LBBB has been associated with negative outcomes in several clinical scenarios. Some (11–13), but not all (14), studies show that patients in the general population with LBBB, with and without known cardiovascular disease, have worse outcomes than their counterparts with no conduction disease. Other studies have shown that LBBB is clearly an adverse prognostic sign in patients with HF in general (1,15–17), including those with severely reduced LVEF (18,19). Our study provides new evidence along these lines, confirming an association between LBBB and negative outcomes in patients with LVEF of 36% to 50%, most of whom did not have a HF diagnosis at baseline.
Mortality rate, the most definitive endpoint, was significantly higher in the LBBB group than the control group in these patients with mildly to moderately reduced LVEF. The Kaplan-Meier survival curves in the LBBB and control groups started separating after 2 years and had persistent further reduction, with a significantly lower survival rate in the LBBB group. The more difficult-to-obtain endpoints, HF admission and ventricular arrhythmia, were not significantly different between groups, although there was a trend toward a higher rate of ventricular arrhythmia in the LBBB group. These outcomes are often difficult to obtain because of patients presenting to their local hospitals rather than returning to our tertiary care center, which is a potential limitation for this study.
Further LVEF reduction in LBBB
In patients who had follow-up echocardiograms, the results clearly show a higher clinically significant drop in LVEF in the patients with LBBB than in the control patients. The difference between the 2 groups was most apparent in those with LVEF decrease of >10% at follow-up. In addition, the LBBB group underwent implantation of ICD at a much higher rate than control patients did.
These outcomes suggest that patients with LBBB are even more vulnerable than other patients with this level of reduced LVEF. This difference was seen despite a significantly higher rate of baseline coronary artery disease, myocardial infarction, and revascularization in the control group. There was a higher rate of moderate or greater AS in the LBBB group; however, the absolute number of patients was low (around 5% in the LBBB group), whereas the number of patients with CAD was quite large (around 50% in the control group). To clarify this issue, we performed a multivariate analysis, which included baseline HF, coronary artery disease, and AS. This confirmed that mortality remains significantly higher in the LBBB group even when these confounding variables are taken into account.
As was alluded to in the introduction, however, it is not imperative to show that patients with LBBB have a worse outcome than their counterparts do. It seems to be true that cardiac resynchronization is most effective in patients with LBBB (20,21). Therefore, the fact that these patients have this ECG finding suggests that they may potentially benefit from CRT, a proven effective treatment for severe HF. This may remain true in the setting of a mildly depressed LVEF and needs to be prospectively studied. CRT could be considered in these patients for 2 reasons. First, 8% of these patients had a diagnosis of HF at baseline, and many more received this diagnosis in the follow-up period. In these symptomatic patients with HF, CRT would provide an effective therapy for LBBB and early-stage HF that might not be completely corrected by medical management. For example, in previous prospective studies of CRT, some patients were entered into the study on the basis of local assessment of LVEF of <35% but were later found to have a higher LVEF on the basis of core laboratory assessment. Retrospective analyses of these studies have shown that CRT appears to also benefit the patients with an LVEF >35% who were errantly enrolled (22,23). Second, >90% of patients in our study did not have a diagnosis of HF prior to the baseline echocardiogram. For these asymptomatic patients with mildly to moderately reduced LVEF, resynchronization might be considered for prevention of developing HF symptoms, similar to what was demonstrated in the MADIT-CRT (Multicenter Automatic Defibrillator Implantation With Cardiac Resynchronization Therapy) study for patients with severely reduced LVEF and New York Heart Association functional class I HF (4). The benefit of CRT in this large group of asymptomatic patients (American College of Cardiology/American Heart Association stage B HF) needs to be prospectively studied.
The high level of negative outcomes in the non-LBBB group may be falsely elevated by the selection bias created by identifying patients by their echocardiogram. The patients without LBBB would have typically undergone echocardiography for a clinical reason, suggesting they were more likely to have symptoms or other significant cardiac disease at baseline. For the patients with LBBB, the echocardiogram may have been indicated on the basis of the ECG finding alone.
Only a baseline echocardiogram was required for inclusion in the study, and therefore the availability and timing of follow-up echocardiograms could not be strictly matched between groups. Also, due to the large number of patients and long, variable time periods, we were unable to gather medication use data for the patient groups. With multiple endpoints, it is possible that some differences between the 2 groups could occur randomly; however, only a small number of pre-selected clinically important endpoints were included, which should minimize this effect. As in all retrospective studies, these findings are limited by their retrospective nature.
Patients with LBBB and mildly to moderately reduced LVEF have remarkably poor long-term clinical outcomes that are significantly worse than those of patients with a similar LVEF but no conduction system disease. The high level of mortality and adverse events seen in this group may suggest a patient population that needs and would potentially benefit from resynchronization therapy.
COMPETENCY IN MEDICAL KNOWLEDGE: It is known that LBBB is associated with a worse prognosis in patients with HF. This study identifies the exceedingly poor long-term outcomes for patients with LBBB and only a mild to moderately reduced LVEF. This information provides a basis for patient discussions regarding prognosis. More importantly, clinicians may opt for closer monitoring and more aggressive therapy for their patients who meet these criteria.
TRANSLATIONAL OUTLOOK: The findings of this study can provide a foundation to develop future prospective studies testing therapies to prevent poor outcomes in patients with LBBB and a mild to moderately reduced LVEF. These trials could examine the benefit of CRT in these patients, including those without HF symptoms. In situations where there is resistance to attempting invasive procedures in mildly symptomatic or asymptomatic patients, these data may provide the necessary impetus.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aortic stenosis
- confidence interval
- cardiac resynchronization therapy
- heart failure
- hazard ratio
- implantable cardioverter-defibrillator
- left bundle branch block
- left ventricular ejection fraction
- ventricular fibrillation
- ventricular tachycardia
- Received June 2, 2016.
- Revision received June 28, 2016.
- Accepted July 7, 2016.
- American College of Cardiology Foundation
- Baldasseroni S.,
- Opasich C.,
- Gorini M.,
- et al.,
- for the Italian Network on Congestive Heart Failure Investigators
- Tracy C.M.,
- Epstein A.E.,
- Darbar D.,
- et al.
- Brignole M.,
- Auricchio A.,
- Baron-Esquivias G.,
- et al.
- Yancy C.W.,
- Jessup M.,
- Bozkurt B.,
- et al.
- Linde C.,
- Curtis A.B.,
- Fonarow G.C.,
- et al.
- Haataja P.,
- Anttila I.,
- Nikus K.,
- et al.
- Kutyifa V.,
- Kloppe A.,
- Zareba W.,
- et al.