Author + information
- Received August 31, 2017
- Revision received November 15, 2017
- Accepted November 16, 2017
- Published online January 17, 2018.
- John Aalen, MDa,b,c,
- Petter Storsten, MDa,b,c,
- Espen W. Remme, Dr.ing., PhDa,b,
- Per A. Sirnes, MD, PhDd,
- Ola Gjesdal, MD, PhDc,
- Camilla K. Larsen, MDa,b,c,
- Erik Kongsgaard, MD, PhDb,c,
- Espen Boe, MDa,b,
- Helge Skulstad, MD, PhDa,b,c,e,
- Jonny Hisdal, PhDe and
- Otto A. Smiseth, MD, PhDa,b,c,e,∗ ()
- aInstitute for Surgical Research, Oslo University Hospital, Rikshospitalet, Oslo, Norway
- bCenter for Cardiological Innovation, Oslo University Hospital, Rikshospitalet, Oslo, Norway
- cDepartment of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Norway
- dOstlandske Hjertesenter, Moss, Norway
- eInstitute of Clinical Medicine, University of Oslo, Oslo, Norway
- ↵∗Address for correspondence:
Dr. Otto A. Smiseth, Department of Cardiology, Oslo University Hospital, Rikshospitalet, N-0027 Oslo, Norway.
Objectives This study sought to investigate the hypothesis that patients with left bundle branch block (LBBB) are hypersensitive to elevated afterload.
Background Epidemiological data suggest that LBBB can provoke heart failure in patients with hypertension.
Methods In 11 asymptomatic patients with isolated LBBB and 11 age-matched control subjects, left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS) were measured by echocardiography. Systolic arterial pressure was increased by combining pneumatic extremity constrictors and handgrip exercise. To obtain more insight into mechanisms of afterload response, 8 anesthetized dogs with left ventricular (LV) micromanometer and dimension crystals were studied during acutely induced LBBB and aortic constriction. Regional myocardial work was assessed by LV pressure-dimension analysis.
Results Consistent with normal afterload dependency, elevation of systolic arterial pressure by 38 ± 12 mm Hg moderately reduced LVEF from 60 ± 4% to 54 ± 6% (p < 0.01) in control subjects. In LBBB patients, however, a similar blood pressure increase caused substantially larger reduction in LVEF (p < 0.01), from 56 ± 6% to 42 ± 7% (p < 0.01). There were similar findings for GLS. In the dog model, aortic constriction abolished septal shortening (p < 0.02), and septal work decreased to negative values (p < 0.01). Therefore, during elevated systolic pressure, the septum made no contribution to global LV work, as indicated by net negative work, and instead absorbed energy from work done by the LV lateral wall.
Conclusions Moderate elevation of arterial pressure caused marked reductions in LVEF and GLS in patients with LBBB. This reflects a cardiodepressive effect of elevated afterload in the dyssynchronous ventricle and was attributed to loss of septal function.
Left bundle branch block (LBBB) is relatively common in patients with congestive heart failure and is associated with reduced left ventricular (LV) function and increased mortality (1). Asymptomatic individuals with structurally normal hearts can also have LBBB, and the prevalence increases with age (2). These subjects are at increased risk of developing cardiovascular disease, including congestive heart failure, which might reflect progression of underlying cardiac disease (3). An alternative mechanism of heart failure may be a negative effect of LBBB on cardiac function, potentially acting together with other cardiovascular disturbances (4–6). As observed in the LIFE (Losartan Intervention for Endpoint Reduction in Hypertension) study, which followed more than 500 patients with LBBB for nearly 5 years, the combination of LBBB and hypertension was associated with a 2-fold higher risk of hospitalization for heart failure compared with hypertensive control patients (7). This may suggest that ventricles with LBBB have reduced tolerance to arterial hypertension, which is consistent with the observation that LBBB leads to impaired LV function (8,9) and therefore is expected to render the ventricle more sensitive to afterload elevation (10,11).
The general objective of the present study was to investigate LV mechanical response to elevated afterload in patients with LBBB. The specific objectives were to: 1) determine to what extent acute elevation of arterial pressure in patients with LBBB causes more reduction in global LV function than in control subjects with normal electrical conduction; and 2) investigate how the early-activated septum and the late-activated LV free wall contribute to the response. As measures of global LV function, we used left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS). Furthermore, segmental function was studied by use of a noninvasive left ventricular pressure (LVP) estimate and segmental work by pressure-strain analysis. Patients with LBBB and preserved LV systolic function were compared with control subjects of similar age. Blood pressure elevation was induced by a combination of extremity cuff inflation and handgrip exercise. In an experimental LBBB model, we studied afterload responses by LVP-segment length analysis, which is considered the gold standard for assessment of LV function.
Eleven patients (62 ± 8 years of age) with LBBB were recruited through an outpatient cardiology practice, and a control group of 11 individuals of similar age (60 ± 10 years) were recruited through voluntary enrollment in the community (Table 1). LBBB was defined according to Strauss et al. (12). All patients were in sinus rhythm. One subject in the control group had complete right bundle branch block. Medical history, clinical examination, electrocardiogram, and echocardiography were obtained in all participants. Patients with coronary artery disease were excluded.
The study was approved by the regional ethics committee, and written informed consent was obtained from all study participants.
Echocardiography and strain analysis
Two-dimensional gray-scale echocardiographic recordings were conducted during baseline and increased afterload using apical 2-, 3-, and 4-chamber views (Vivid E9, GE Vingmed Ultrasound, Horten, Norway). In addition, during baseline, 2-dimensional gray-scale images were obtained in the parasternal long-axis view to measure LV dimensions. Doppler echocardiography was used to assess heart valve function. Ventricular volumes and LVEF were calculated by the biplane Simpson method.
Global and segmental strain analyses were performed off-line with speckle tracking echocardiography (Echopac, GE Vingmed Ultrasound). Average frame rate was 59 ± 6/s. Segmental strains from the septum and LV lateral wall were calculated as the average end-systolic value from single wall analysis in the apical 4-chamber view.
Estimation of regional work
An index of segmental myocardial work was calculated by LV pressure-strain analysis using a semiautomated analysis tool (Echopac, version 202, GE Vingmed Ultrasound). The method, which includes a noninvasive estimate of LVP, was validated previously and has been described in detail (13). The work index (mm Hg·%) was calculated by multiplying the rate of segmental shortening (strain rate) with instantaneous LVP. This resulted in a measure of instantaneous power, which was integrated over time to give work as a function of time in systole, defined as the time interval from mitral valve closure to mitral valve opening (14).
In the present study, the net work for a myocardial segment was calculated as the sum of positive and negative work. Normally, LV segments shorten in systole when LVP is rising, and by definition do positive work. A segment that becomes elongated in systole when LVP is rising, by definition does negative work because it is stretched as a result of contraction in other segments.
Elevation of afterload
Afterload was elevated by combining 30 s of isometric handgrip exercise (80% of maximum voluntary contraction) performed by the right hand with inflation of cuffs placed around the 2 lower extremities and the right upper extremity. Simultaneous with handgrip exercise, the blood pressure cuffs were inflated to suprasystolic pressure to cause compression of extremity arteries. In a pilot study, the combined intervention was shown to give higher blood pressure than either of the methods alone. Arterial blood pressure was continuously recorded during the intervention by a photoplethysmographic pressure recording device (Finometer, FMS Finapres Medical System, Amsterdam, the Netherlands) placed at the left hand. The pressure recording device was calibrated against brachial blood pressure. The intervention was repeated to obtain echocardiographic images from one apical view at a time, which allowed us to use heart beats with similar timing and afterload when measuring LVEF and GLS. Figure 1 shows recordings from a representative individual.
Eight mongrel dogs of either sex and with a body weight of 34 ± 2 kg were anesthetized by use of barbiturates and opioids (n = 7; thiopentone 25 mg/kg and morphine 100 mg IV, followed by infusion of morphine 50 to 100 mg/h and pentobarbital 50 mg IV every hour) or, because of change of anesthetic protocol in our laboratory, propofol and opioids (n = 1; single dose of methadone 0.2 mg/kg, followed by propofol 3 to 4 mg/kg and a bolus of fentanyl 2 to 3 μg/kg, then a continuous infusion of propofol 0.2 to 1 mg/kg/min and fentanyl 5 to 40 μg/kg/h). The animals were ventilated and surgically prepared as described previously (15). To increase afterload, an inflatable silicone occluder was placed around the ascending aorta. The National Animal Experimentation Board approved the study. The animals were supplied by the Center for Comparative Medicine (Oslo University Hospital, Rikshospitalet, Oslo, Norway).
Pressures, dimensions, and electromyograms
Aortic, LV, and left atrial (LA) pressures were measured by 5-F micromanometer-tipped catheters (model MPC 500, Millar Instruments Inc., Houston, Texas). The LV and LA micromanometers were zero referenced to LA pressure measured via a fluid-filled line during diastasis, using postextrasystolic beats with long diastoles.
LV dimensions were measured as circumferential segment lengths by sonomicrometry using 2-mm-wide crystals (Sonometrics Corporation, London, Ontario, Canada) implanted in the inner third of the myocardium in the interventricular septum and in the posterior, anterior, and lateral LV walls. The ultrasonic crystals were combined with a bipolar electrode for recording of intramyocardial electromyograms. Data were sampled at 200 Hz.
Induction of LBBB (Online Figure 1) was performed by radiofrequency ablation as described previously (16). Response to elevated afterload by aortic constriction was studied shortly before and soon after induction of LBBB. By recording beat-to-beat changes at onset of aortic constriction, we could study direct afterload responses with minimal influence from reflexes induced by the procedure. Data were recorded with the ventilator temporarily switched off.
Segmental work was obtained by a similar principle as in the clinical study, but with the micromanometer instead of an estimate of LVP and absolute dimensions by sonomicrometry instead of percentage strain. Net systolic shortening was measured as end-diastolic minus end-systolic dimension. End diastole was defined as onset of the septal electromyogram.
Values are presented as mean ± SD. Comparisons between 2 groups were performed with independent-samples t tests, whereas comparisons within the same group were performed by paired-samples t tests. Bonferroni correction was applied to adjust for testing in 2 groups. In Tables 2 and 3⇓⇓, significance of the interaction effect from an analysis of variance was reported to highlight differences in afterload effect between LBBB and control conditions. A value of p < 0.05 was considered significant. SPSS version 24.0 (SPSS Inc., Chicago, Illinois) was used for the analyses.
Characteristics of the study participants are provided in Table 1. None of the subjects had more than mild valvular regurgitation.
There were no significant differences between the groups in LV dimensions, volumes, or LVEF; however, dimensions and volumes tended to be larger and LVEF somewhat lower in the LBBB group (Tables 1 and 2). LV longitudinal shortening by GLS was 17.1 ± 2.2% in the LBBB group compared with 20.8 ± 2.5% in the control group (p < 0.01). In the LBBB group, septal systolic shortening was 11.7 ± 3.8% compared with 18.2 ± 2.6% in the control group (p < 0.01). There were also markedly lower values for septal work in LBBB patients than in control subjects (Table 2). In the LV lateral wall, however, the work index showed similar values in LBBB patients and control subjects.
Cuff inflation in combination with handgrip effort resulted in increases in systolic blood pressure of 34 ± 13 and 38 ± 12 mm Hg in LBBB patients and control subjects, respectively (p = NS). There was no change in QRS duration, but a moderate increase in heart rate in the 2 groups (22 ± 10 compared with 20 ± 11 beats/min in LBBB patients and control subjects, respectively; p = NS). As illustrated in Figure 2, the effect of elevated afterload on LVEF and GLS was different in the 2 groups. In the control group, elevation of systolic pressure was associated with moderate reductions in LVEF (from 60 ± 4% to 54 ± 6%; p < 0.01) and in GLS (from 20.8 ± 2.5% to 18.4 ± 2.4%; p < 0.01). In LBBB patients, however, there were marked reductions in LVEF (from 56 ± 6% to 42 ± 7%; p < 0.01) and in GLS (from 17.1 ± 2.2% to 12.4 ± 1.8%; p < 0.01). In absolute terms, the reduction in LVEF was 14% in patients with LBBB compared with 7% in control subjects (95% confidence interval: 11% to 17% and 3% to 10%, respectively; p < 0.01), and the reductions in GLS were 4.7% and 2.3% (95% confidence interval: 3.7% to 5.7% and 1.0% to 3.6%, respectively; p < 0.01). There was a similar trend in stroke volume reduction (15 ± 10 ml compared with 8 ± 6 ml in LBBB patients and control subjects, respectively).
As illustrated in Figure 2, the reduction in GLS with elevated afterload in patients with LBBB was attributed to reduced septal function. There was reduction in systolic shortening from 11.7 ± 3.8% to 6.3 ± 4.6% (p < 0.01) (Table 2), which was attributed to reduction in shortening during the LV ejection phase (p < 0.01) and a trend toward increased early systolic lengthening. In LBBB patients, septal contraction during elevated afterload was highly inefficient, because the pressure-strain loops showed clockwise rotation during a large portion of systole, which implies systolic lengthening and therefore a large component of negative work (Figure 3C). The result was a marked decrease in net septal work during elevation of afterload (Table 2). Myocardial work in the lateral wall was maintained, although systolic shortening tended to decrease (Table 2).
Induction of LBBB caused depression of global LV function, as indicated by a moderate fall in maximum rate of rise of LVP (LV dP/dtmax) (p < 0.01), and there was a marked reduction in septal systolic shortening (Table 3). In contrast to patients, there was no accompanying tachycardia suggesting reflex activation during increased afterload.
Aortic constriction increased LV systolic pressure by 24 ± 12 and 24 ± 11 mm Hg before and after induction of LBBB, respectively. Before LBBB, aortic constriction caused only moderate reductions in septal and LV lateral wall systolic shortening (Table 3). During LBBB, however, aortic constriction caused a marked reduction in septal systolic shortening (p < 0.02) because of reduced shortening during LV ejection (p < 0.01) (Figure 4). There was also a trend toward an increase in systolic lengthening, in particular early-systolic lengthening, which contributed to the reduction in systolic shortening during aortic constriction. An increase in systolic lengthening implies increased negative septal work (p < 0.05), and a reduction in septal shortening implies less positive work (p < 0.02); the result was a shift from net positive to net negative septal work (p < 0.01) (Figure 3D, Figure 5, and Table 3). LV lateral wall shortening and work, however, were maintained during aortic constriction. Septal pre-ejection shortening was unaffected by elevated afterload.
As shown in Table 3, responses in the time constant (tau) of LV relaxation and LV end-diastolic pressure were similar during control conditions and LBBB. Minimum LVP, however, increased modestly but significantly more during LBBB.
The present study demonstrates increased afterload sensitivity in patients with LBBB and preserved LV systolic function, as indicated by marked reductions in LVEF and GLS in response to moderate elevation of systolic arterial pressure. When subjects without LBBB were exposed to a similar rise in arterial pressure, they showed only moderate reductions in LVEF and GLS, consistent with normal afterload dependency of myofiber shortening (17,18). This observation may have impact on strategies for management of patients with LBBB.
Mechanism of reduction in systolic function during elevated afterload
Consistent with previous studies, we found impaired LV systolic function (8,9) and abnormal septal motion in patients with LBBB (19). The cardiodepressive effect of elevated afterload was attributed to aggravation of septal dysfunction, which included reduction in septal shortening during LV ejection and septal systolic lengthening. The responses to elevated arterial pressure in patients were reproduced in the experimental model, which provided a measure of myocardial segment length and not just strain.
When evaluating myocardial function in terms of regional work, we confirmed the results of previous studies that have shown markedly reduced work in the septum during LBBB and during pacing-induced LV dyssynchrony (13,20,21). In the clinical study, elevation of afterload caused further reduction in septal work, in part because of increased negative septal work, which reflects septal lengthening caused by contractions in the LV lateral wall. In the dog model, septal work was converted to net negative values during elevation of afterload, which implies that the septum made no contribution to global LV work but instead absorbed energy from work done in the LV free wall. Myocardial work in the LV lateral wall was maintained during aortic constriction. Therefore, in patients with LBBB, the LV lateral wall carries a relatively larger mechanical burden than in normal hearts, and this imbalance was even stronger during elevated afterload. The additional workload on the LV lateral wall when LBBB is combined with hypertension may represent a stimulus to remodeling (22). A similar mechanism could explain upregulation of proteins involved in myocardial hypertrophy in late activated myocardium in dyssynchronous ventricles (23).
It was suggested in previous studies that the association between LBBB and congestive heart failure in otherwise apparently healthy individuals reflects progression of coexisting subclinical myocardial disease (3,24). As shown in the present study, response to elevated afterload during LBBB in the acute dog model was in principle similar to the response in patients, which excludes that the increased afterload sensitivity was entirely caused by preexisting or subclinical myocardial dysfunction. Instead this supports the notion that LV afterload hypersensitivity is a direct consequence of LBBB itself.
The observation that a moderate acute elevation of aortic pressure caused a marked reduction in LVEF in patients with LBBB raises the question whether such patients are more vulnerable to arterial hypertension. Potentially, the increased risk of congestive heart failure for hypertensive patients with LBBB (7) is attributable to afterload-induced LV dysfunction in a ventricle with increased sensitivity to afterload. If there is a similar response to chronic elevation of arterial pressure, the grossly abnormal septal function and altered LV free wall function could lead to remodeling and precipitation of heart failure. There is need for future studies to examine whether antihypertensive treatment has beneficial effects on LV contractile function in patients with LBBB and whether treatment goals should be stricter than for hypertensive patients in general.
The reductions in LVEF and GLS in response to elevated afterload in individuals without LBBB is consistent with normal myocardial physiology (17). Afterload dependency is a known limitation of all ejection phase indices of LV systolic function, including GLS (15). The augmented afterload sensitivity during LBBB could make interpretation of LV systolic function more challenging in this group of patients. If similar afterload dependency exists for heart failure patients with LBBB, it could be important to take blood pressure into account when using LVEF as a criterion for selection of patients for cardiac resynchronization therapy (CRT). In 4 of 11 patients with LBBB, there was a fall in LVEF to values less than 40% with elevated arterial pressure, which approaches the levels of LVEF when CRT may be indicated.
To improve responder rates, different septal deformation patterns have been suggested as predictors of CRT response (25–27). The finding that septal motion is highly affected by differences in afterload suggests that arterial blood pressure should be taken into account when evaluating patients for CRT based on such patterns.
In the clinical study, afterload was increased by a procedure that activated reflexes, thereby increasing sympaticoadrenal tone (28). This was reflected in the accompanying tachycardia and unchanged end-diastolic volume, which means the Frank-Starling mechanism was not activated. Therefore, this was not a pure afterload intervention. In the dog model, however, we confirmed that responses to elevated afterload were immediate, occurring as beat-to-beat changes, and therefore could not be attributed to reflex responses but reflected direct mechanical interactions between afterload and contracting myofibers.
Measurement of arterial blood pressure at the finger could have overlooked effects from the present intervention on arterial wave reflection and pressure augmentation and thereby included an error in the assessment of central blood pressure. However, because the main objective was to apply a similar afterload increase in patients with LBBB and control subjects, the absolute numbers for systolic and diastolic blood pressure were not essential.
Myocardial work was calculated with LVP as a substitute for force. Measurement of force would have been optimal but is complicated, because radius of curvature and wall thickness change continuously during the heart cycle. In a previous experimental study from our laboratory, we showed that pressure-length loops during myocardial ischemia reflected myocardial work measured from force-segment length loops (29). In a validation study in LBBB, we found excellent correlations between the noninvasive work estimate and work by pressure-length analysis in a dog model and against LVP by micromanometer and strain in a clinical study (13). However, because strain is a relative measure, caution should be exercised when comparing work by pressure-strain analysis from hearts with different sizes. In the present study, we confirmed a detrimental effect of elevated afterload on septal work when using sonomicrometry to measure absolute dimension.
In the experimental study, we assessed animals that were heavily instrumented and during general anesthesia, both of which can cause deterioration in cardiac function. Furthermore, we cannot exclude that myocardial function might have deteriorated during the time that elapsed between control and LBBB conditions. However, when control conditions and LBBB were compared, there were no significant reductions in global LV function given as stroke work, stroke volume, and LV end-diastolic pressure, which indicates limited deterioration.
The present study demonstrated marked reductions in LVEF and GLS in response to a moderate elevation of blood pressure in individuals with LBBB and preserved LV systolic function, and this response far exceeded reductions in LVEF and GLS when afterload was elevated in a control population. The mechanism of this exaggerated afterload response during LBBB was septal dysfunction. Future studies should explore whether a similar interaction between afterload and LV contractile function occurs in heart failure patients with LBBB.
COMPETENCY IN MEDICAL KNOWLEDGE: Patients with LBBB are less tolerant to acute elevations of blood pressure.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: When evaluating patients for CRT, the clinician should be aware that moderate fluctuations in blood pressure, which occur frequently in daily clinical practice, may have a significant impact on LVEF and thus on the decision whether CRT is indicated.
TRANSLATIONAL OUTLOOK 1: There is a need for prospective trials to evaluate whether treatment goals for hypertension should be different in patients with LBBB.
TRANSLATIONAL OUTLOOK 2: Future studies should emphasize whether LVEF and GLS are less reliable markers for changes in systolic function in patients with LBBB.
The authors thank Dr. Anders Opdahl, Dr. Kristoffer Russell, and Surgical Nurse Aurora Pamplona for their contribution to the animal experiments.
Dr. Aalen was supported by a grant from the Norwegian Health Association. Drs. Storsten and Kjellstad Larsen were recipients of clinical research fellowships from the South-Eastern Norway Regional Health Authority. Professor Smiseth is co-inventor but no longer has ownership of the patent “Method for myocardial segment work analysis,” which was used to calculate myocardial work in the clinical study. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- cardiac resynchronization therapy
- global longitudinal strain
- left atrial
- left bundle branch block
- left ventricular
- left ventricular ejection fraction
- left ventricular pressure
- Received August 31, 2017.
- Revision received November 15, 2017.
- Accepted November 16, 2017.
- 2018 American College of Cardiology Foundation
- Baldasseroni S.,
- Opasich C.,
- Gorini M.,
- et al.,
- on behalf of the Italian Network on Congestive Heart Failure Investigators
- Eriksson P.,
- Hansson P.O.,
- Eriksson H.,
- Dellborg M.
- Vaillant C.,
- Martins R.P.,
- Donal E.,
- et al.
- Myhre E.S.P.,
- Johansen A.,
- Bjornstad J.,
- Piene H.
- Urheim S.,
- Edvardsen T.,
- Torp H.,
- Angelsen B.,
- Smiseth O.A.
- Gjesdal O.,
- Remme E.W.,
- Opdahl A.,
- et al.
- Siegel J.H.,
- Sonnenblick E.H.
- Leite-Moreira A.F.,
- Gillebert T.C.
- McDonald I.G.
- Prinzen F.W.,
- Hunter W.C.,
- Wyman B.T.,
- McVeigh E.R.
- Chakir K.,
- Daya S.K.,
- Aiba T.,
- et al.
- Spragg D.D.,
- Leclercq C.,
- Loghmani M.,
- et al.
- Risum N.,
- Tayal B.,
- Hansen T.F.,
- et al.
- Kivowitz C.,
- Parmley W.W.,
- Donoso R.,
- Marcus H.,
- Ganz W.,
- Swan H.J.C.
- Skulstad H.,
- Edvardsen T.,
- Urheim S.,
- et al.