Is exercise stress echocardiography useful in patients with suspected obstructive coronary artery disease who have resting left bundle branch block?

Abstract

Background

Current guidelines support exercise stress echocardiography (ESE) for evaluation of suspected obstructive coronary artery disease (OCAD) in ambulant patients with left bundle branch block (LBBB). Data regarding the diagnostic utility of ESE in patients with LBBB are limited.

Hypothesis

We hypothesized that the diagnostic performance of ESE for the assessment of suspected OCAD is reduced in the context of LBBB.

Methods

We studied 191 consecutive patients with resting LBBB undergoing ESE for the investigation of suspected OCAD between 2008 and 2015 at our center. The studies were categorized as inconclusive, normal, or abnormal. Patients with an abnormal response were subcategorized as regional ischemic response or globally abnormal.

Results

Eighty‐two patients (43%) demonstrated a normal left ventricular contractile response (LVCR) to exercise; 92 (48%) developed an abnormal LVCR to exercise, including 70 patients with globally abnormal and 22 patients with regional ischemic responses. Of the patients with abnormal responses, 62 patients had anatomic imaging, only 29 of whom had significant OCAD, conferring an overall specificity of ESE for significant OCAD of 21% and accuracy of 52%. Of patients who developed a regionally abnormal response, 89% had significant OCAD.

Conclusions

For patients with LBBB who develop a globally abnormal LVCR during ESE, the specificity of ESE for reliably excluding significant OCAD is significantly reduced. ESE appears to be a suboptimal test for the evaluation of OCAD in patients with resting LBBB, as about 50% of patients will have an abnormal response, the majority due to globally abnormal contraction where OCAD cannot be reliably diagnosed. Alternative testing should be considered for the investigation of suspected OCAD in patients with resting LBBB.

1. INTRODUCTION

Exercise stress echocardiography (ESE) is a well‐validated technique for the assessment of myocardial ischemia in patients with suspected obstructive coronary artery disease (OCAD).1, 2 In clinical practice, ESE is used to assess for OCAD in patients with resting left bundle branch block (LBBB), based largely on the class I guideline recommendation from the American College of Cardiology Foundation/American Heart Association (ACCF/AHA) that “exercise stress with nuclear myocardial perfusion imaging (MPI) or echocardiography is recommended for patients with an intermediate to high pretest probability of ischemic heart disease who have an uninterpretable electrocardiogram (ECG) and at least moderate physical functioning or no disabling comorbidity.”3 This approach is also supported by the American Society of Echocardiography (ASE) recommendation that ESE is an appropriate investigation to assess for OCAD in patients with resting LBBB.4

Surprisingly, limited data are available in the literature regarding the diagnostic utility of ESE for the assessment of OCAD in patients with LBBB. The guidelines refer to a meta‐analysis of noninvasive imaging techniques for the diagnosis of OCAD in patients with LBBB, which included 43 MPI studies and 6 stress echocardiography studies.5 Of the stress echocardiography studies, only 1 study involved exercise stress6; and, interestingly, this small study showed that a significant proportion of patients with LBBB (18/35) had a fall in left ventricular ejection fraction (LVEF) post‐stress, which represents an abnormal response.7 This is consistent with our own empiric observation.

We investigated the diagnostic utility of ESE for the detection of OCAD in patients with resting LBBB referred for clinical ESE at our center.

2. METHODS

2.1. Patient selection

Consecutive ESE studies performed at our hospital (tertiary referral center) over an 8‐year period from January 2008 to August 2015 were analyzed after institutional review board approval was obtained from our center. Patients with resting LBBB, referred for the assessment of suspected underlying OCAD, were identified retrospectively. Test indication was noted, and standard clinical and anthropometric data were collected.

2.2. Exercise stress protocol

All patients performed symptom‐limited treadmill exercise according to the standard Bruce protocol, with continuous ECG monitoring and 2‐minute blood pressure (BP) measurements. Tests were concluded for the usual endpoints. Exercise capacity, symptom status, resting and maximum heart rates (HR), and BPs were recorded in all patients. In the literature, a hypertensive response to exercise has been defined as ≥220/95 mmHg in men and ≥190/95 mmHg in women.8 In this study, a patient was categorized as having a hypertensive response to exercise if both the systolic and diastolic BP thresholds were reached or exceeded, or if the systolic BP thresholds were reached or exceeded and the diastolic BPs were within 10 mmHg of the reported thresholds.

2.3. Imaging protocol

Baseline and post‐exercise transthoracic imaging was obtained by experienced sonographers using Vivid 7/Vivid E9 cardiovascular ultrasound systems (GE Healthcare, Little Chalfont, United Kingdom).

Detailed resting imaging was performed from the parasternal and apical windows including evaluation of left heart chamber dimensions, LVEF, regional wall motion, left ventricular (LV) long‐axis contraction [assessed by measuring the early diastolic mitral annular velocities [e'] at the LV septum and lateral wall), cardiac valvular function, and estimated right ventricular systolic pressure.

Pre‐ and post‐exercise images were compared using the parasternal long‐ and short‐axis windows, and the apical 4‐chamber, apical 2‐chamber, apical long‐axis windows as well as an apical short‐axis window. Each study was interpreted by 1 of 5 expert echocardiologists. Equivocal studies were independently reread by 1 experienced echocardiologist (SM). An equivocal study was defined as a study for which there was uncertainty in the interpretation of the LV response following exercise (ie, globally abnormal vs regionally abnormal).

2.4. Stress echocardiogram categorization

Each stress echocardiogram was categorized as either inconclusive, normal, or abnormal. In patients achieving <85% of the age‐predicted maximum HR during exercise, the test was considered inconclusive.

A normal left ventricular contractile response (LVCR) to exercise was defined as a visible reduction in LV cavity dimension with an increase in LVEF following exercise, without inducible systolic wall‐thickening abnormality in a nonseptal segment.

An abnormal LVCR to exercise was categorized as either (1) regional ischemic response, reflecting new inducible regional hypokinesis (not involving the septum) within a coronary territory irrespective of changes in LV cavity size or ejection fraction; or (2) globally abnormal response: reflecting transient global LV hypokinesis causing LV dilatation and reduction in ejection fraction, or failure of LV augmentation post exercise reflecting no change in LV cavity size or ejection fraction post stress. Both regionally abnormal and globally abnormal responses were determined on the basis of visual assessment of multiple side‐by‐side comparisons of rest and peak stress images obtained from standard parasternal and apical imaging windows.

2.5. Reproducibility of stress echocardiogram categorization

To define the reproducibility of stress echocardiogram categorization by the echocardiologists, 20 patients were randomly selected from the study cohort. Two blinded echocardiologists independently interpreted the stress echocardiogram studies and categorized each study as either normal, regional ischemic response, or globally abnormal response.

2.6. Anatomic imaging

Study patients who underwent subsequent anatomical imaging with invasive coronary angiography or CTCA at our center within 100 days were identified. Significant OCAD was defined as >70% stenosis in a major epicardial artery. At the time of invasive angiography, the severity of each coronary artery lesion was determined visually by an independent interventional cardiologist. At the time of CTCA, the severity of each coronary artery lesion was determined visually by both an experienced imaging cardiologist and an experienced cardiovascular radiologist. Both physicians independently assessed and interpreted the severity of each coronary artery lesion. For coronary artery lesions where the lesion severity differed between the 2 reporting physicians on initial reporting, reassessment of the lesions would occur before reaching a consensus for the final report.

2.7. Statistical analysis

Continuous data were expressed in terms of mean ± SD. Categorical data were expressed in terms of absolute count and percentage of cohort. Continuous variables were compared using 1‐way ANOVA, followed by post hoc Bonferroni tests. The association of selected variables with a globally abnormal LVCR was assessed using a binary logistic regression model. The following covariates were analyzed: age, sex, height, body surface area, body mass index, weight, exercise duration, exercise capacity, resting HR, peak HR, resting systolic and diastolic BP, peak systolic and diastolic BP, hypertensive response to exercise, and resting LVEF. Odds ratios (ORs) with corresponding 95% confidence intervals (CIs) were estimated. The reproducibility of stress echocardiogram categorization was evaluated by calculating the interobserver variability using the Cohen κ coefficient. A P value <0.05 was considered statistically significant. Data were analyzed using SPSS version 16.0 (SPSS Inc., Chicago, IL).

3. RESULTS

3.1. Patient cohort

Between January 2008 and August 2015, a total of 13614 ESEs were performed at our center. We identified 191 consecutive patients (53% female; mean age, 65 ± 11 years) with resting LBBB who underwent ESE according to the Bruce protocol during this period (Table 1). The most common indications for ESE were chest pain for investigation (41%), exclusion of coronary artery disease (CAD) in patients with LBBB with/without cardiac risk factors (35%), and dyspnea for investigation (18%; Table 1).

Table 1.

Baseline clinical and echocardiographic data of the entire study cohort

Characteristic	Value
No. of patients	N = 191
Percentage with LBBB as total number of ESE	1.4
Patients achieving ≥85% of maximum age‐predicted HR	174 (91)
Indication for ESE
Chest pain for investigation	79 (41)
Dyspnea	35 (18)
Exclusion of CAD in presence of LBBB and cardiac risk factors	67 (35)
Other: valvular heart disease, arrhythmias, cardiomyopathy	10 (6)
Age, y	65 ± 11
Sex
M	47
F	53
Height, cm	164 ± 10
Weight, kg	77.5 ± 15.6
BSA, m2	1.84 ± 0.23
BMI, kg/m2	28.7 ± 5.0
Exercise duration, min	7.2 ± 2.8
Exercise capacity, METs	8.8 ± 2.9
Resting HR, bpm	78 ± 15
Peak HR, bpm	149 ± 23
Resting BP, mm Hg	135/78 ± 17/11
Peak BP, mm Hg	174/80 ± 26/12
Hypertensive response to exercise	31 (16)
Resting LVEF, %
Normal (>55)	118 (62)
Mildly reduced (45–55)	45 (24)
Moderately reduced (35–45)	17 (9)
Severely reduced (<35)	11 (5)
Resting early diastolic mitral annular velocities, cm/s
Septal	5 ± 2
Lateral	7 ± 2
Further anatomical imaging	80
CTCA	22
Invasive coronary angiography	58
Severe OCAD on anatomical imaging	32

Abbreviations: BMI, body mass index; BP, blood pressure; BSA, body surface area; CAD, coronary artery disease; CTCA, computed tomography coronary angiography; ESE, exercise stress echocardiography; F, female; HR, heart rate; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; M, male; MET, metabolic equivalent of task; OCAD, obstructive coronary artery disease; SD, standard deviation.

Data are presented as n (%), mean ± SD, or percentage alone (%).

3.2. Stress echocardiogram results

Seventeen (9%) patients were deemed to have an inconclusive study, as they failed to achieve 85% of the age‐predicted maximum HR. 174 consecutive patients exercised to achieve ≥85% of the age‐predicted maximum HR (Table 1). Of these, 82 patients (47%) demonstrated a normal LVCR to exercise; 92 (53%) had an abnormal LVCR, including 70 patients with a globally abnormal response and 22 patients with a regional ischemic response. Characteristics of patients with different LV responses are detailed in Table 2. In essence, there were more males in the group with a regionally ischemic response, and a higher percentage of patients with a normal LVCR had normal resting LVEF when compared with patients in the regionally abnormal or globally abnormal LVCR groups.

Table 2.

Clinical and echocardiographic data of the different patient groups, categorized according to type of LVCR to exercise

Cohort	Normal LVCR	Regional Ischemic LVCR	Globally Abnormal Response	P Value
No. of patients	82 (47)	22 (13)	70 (40)	N/A
Age, y	64 ± 12	66 ± 10	67 ± 11	0.201
Sex
M	30 (37)	17 (77)	32 (46)	0.003
F	52 (63)	5 (23)	38 (54)
Height, cm	163 ± 11	170 ± 8	164 ± 10	0.021
Weight, kg	76.4 ± 15.8	82.7 ± 14.8	79.4 ± 18.6	0.256
BSA, m2	1.8 ± 0.2	1.9 ± 0.2	1.8 ± 0.3	0.744
BMI, kg/m2	28.8 ± 5.0	28.6 ± 4.2	29.4 ± 6.8	0.166
Exercise duration, min	8.2 ± 2.4	6.9 ± 3.1	6.3 ± 2.8	<0.05
Exercise capacity, METs	9.9 ± 2.6	8.8 ± 3.0	7.9 ± 2.9	<0.05
Resting HR, bpm	78 ± 13	78 ± 17	81 ± 16	0.485
Peak HR, bpm	154 ± 17	144 ± 24	153 ± 24	0.103
Resting BP, mm Hg	134/79 ± 18/9	133/77 ± 12/13	139/78 ± 18/13	0.164
Peak BP, mm Hg	176/81 ± 20/10	172/78 ± 22/10	176/81 ± 29/15	0.688
Hypertensive response to exercise	10 (12)	3 (14)	18 (26)	0.074
Resting LVEF, %
Normal, >55	72 (88)	8 (36)	30 (43)	<0.05
Mildly reduced, 45–55	6 (7)	8 (36)	25 (36)	<0.05
Moderately reduced, 35–45	4 (5)	5 (23)	7 (10)	0.035
Severely reduced, <35	0 (0)	1 (5)	8 (11)	0.006
Resting early diastolic mitral annular velocities
Septal E', cm/s	5 ± 2	5 ± 2	5 ± 2	0.621
Lateral E', cm/s	7 ± 2	8 ± 2	7 ± 2	0.888
Further anatomical imaging				N/A
CTCA	1		18
Invasive coronary angiography	10	18	26
Severe OCAD on anatomical imaging	2	16	13	N/A

Abbreviations: BMI, body mass index; BP, blood pressure; BSA, body surface area; CTCA, computed tomography coronary angiography; F, female; HR, heart rate; LV, left ventricular; LVCR, left ventricular contractile response; LVEF, left ventricular ejection fraction; M, male; MET, metabolic equivalent of task; N/A, not available/applicable; OCAD, obstructive coronary artery disease; SD, standard deviation.

Data are presented as n (%) or mean ± SD.

3.3. Anatomical imaging

Eighty patients underwent anatomical imaging within 100 days of the ESE, including 58 invasive coronary angiograms and 22 computed tomography coronary angiography (CTCA) studies. Of patients who underwent subsequent anatomical imaging, the subsequent anatomical imaging occurred within 30 days of the ESE for 76% of the patients.

Seven of 17 (41%) patients with an inconclusive ESE underwent subsequent anatomical imaging (1 of whom had OCAD), and 11/82 (13%) patients with a normal LVCR underwent subsequent anatomical imaging (2 of whom had OCAD).

Of 92 patients with an abnormal LVCR, 62 (67%) had anatomical imaging; 32 (40%) patients who underwent anatomical imaging were found to have OCAD. This included 16 of 18 patients with a regional ischemic LVCR, and 13 of 44 patients with a globally abnormal LVCR.

3.4. Diagnostic accuracy

Of the 73 patients who achieved ≥85% maximum predicted HR and underwent anatomic imaging, ESE was highly sensitive for detection of OCAD for patients with resting LBBB (94%; 29/31 patients), but the specificity was poor (21%; 9/42 patients). The positive predictive value was 47% (29/62 patients), and the test accuracy was 52% (38/73 patients).

3.5. Associations with globally abnormal LVCR

On binary logistic regression analysis, the following factors were found to be associated with a globally abnormal LVCR in the setting of resting LBBB: a higher peak HR during exercise (OR: 1.06, 95% CI: 1.03‐1.09, P < 0.001) and a hypertensive response to exercise (OR: 7.26, 95% CI: 1.81‐29.07, P = 0.005). Of note, a globally abnormal LVCR was not associated with age, sex, height, body surface area, body mass index, weight, exercise duration, exercise capacity, resting HR, resting systolic and diastolic BP, or reduced resting LVEF (P > 0.05).

3.6. Interobserver variability

Cohen's κ for the categorization of stress echocardiogram response by 2 blinded echocardiologists was 0.73 (95% CI: 0.45‐1.00), indicating a good strength of agreement.

4. DISCUSSION

The results of this study show that, contrary to current guidelines, ESE appears to be a suboptimal technique for the evaluation of OCAD in patients with resting LBBB. About half of our cohort demonstrated an abnormal LVCR, predominantly related to global rather than regional abnormality, and less than half of the patients with an abnormal LVCR who underwent subsequent anatomical imaging were found to have OCAD, giving rise to poor overall test specificity (21%) and accuracy (52%).

4.1. Diagnostic accuracy of ESE in LBBB

The current ACCF/AHA Guidelines for Diagnosis and Management of Patients With Stable Ischemic Heart Disease recommend ESE as a class 1 indication for patients with an intermediate to high pretest probability of ischemic heart disease who have an uninterpretable ECG and at least moderate physical functioning or no disabling comorbidity.3 These recommendations largely stem from a meta‐analysis by Biagini et al. that included 43 MPI studies and 6 stress echocardiography studies.5 The 6 stress echocardiography studies included 5 dobutamine stress echocardiogram studies, and the 1 exercise stress echocardiogram study by Peteiro et al.6 This meta‐analysis suggested that the sensitivity for the detection of OCAD in LBBB was higher in qualitatively analyzed MPI (88.5%) vs stress echocardiography (74.6%) but conversely, specificity in LBBB was significantly higher for stress echocardiography (88.7%) vs MPI (41.2%), with similar overall accuracy.5

The only published study on the diagnostic utility of ESE in the setting of LBBB involved 35 patients from a single center who all underwent ESE and angiography, with 17 patients being diagnosed with OCAD.6 The results of this small study suggested a sensitivity of 76%, a specificity of 83%, and an accuracy of 80% of ESE for the detection of significant OCAD in the setting of resting LBBB. Of note, a significant proportion of the study cohort (8 patients; 23%) did not achieve ≥85% of the maximum predicted heart rate, and interestingly just over half (18/35) of patients in this study were also reported to have a decrease in LVEF post‐exercise (12/17 patients with OCAD; 6/18 patients without OCAD).6 It is unclear whether these patients were categorized as having an abnormal response, as they would be according to contemporary reporting standards.7 Our study cohort of 191 patients is the largest in the literature investigating the diagnostic utility of ESE in LBBB; and with stress echocardiography categorized by contemporary reporting standards,7 our results demonstrate a poor overall specificity of ESE for the detection of OCAD in patients with resting LBBB. The group with a globally abnormal response was most confounding, as this group represented 76% of all abnormal tests, and only 30% of these patients had OCAD. Additionally there were no reliable pretest predictors of a globally abnormal LVCR. In comparison, the incidence of significant OCAD in patients who developed a regional ischemic LVCR was high (89%).

Although ESE has low specificity for detection of OCAD for patients with resting LBBB, it may still be a useful investigation in this population of patients. In particular, exercise stress provides valuable physiologic and prognostic information, and the presence of a normal LVCR in the setting of LBBB indicates a good prognosis, as demonstrated by Supariwala et al.9 They analyzed 7214 patients undergoing stress echocardiography over a 14‐year period (51% of these patients underwent dobutamine stress echocardiography; 49% of these patients underwent ESE).9 Fifty patients with LBBB were identified.9 Among patients with a normal stress echocardiogram, those with LBBB had similar mortality to those without LBBB, and patients with LBBB and an abnormal stress echocardiogram had >2× greater risk of all‐cause mortality.9 Additionally, Bouzas‐Mosquera et al. examined the prognostic value of ESE in 609 patients with LBBB over a 12‐year period.10 Of these patients, those with a regional ischemic LVCR were reported to have a higher 5‐year mortality rate (24.6% vs 12.6%; P < 0.001 for no ischemia) and 5‐year major adverse cardiac events rate (18.1% vs 9.7%; P = 0.003).10 The rate of major adverse cardiac events among patients with LBBB and a normal LVCR to exercise was reported as 0.92% per year, compared with 3.6% per year for patients with LBBB and a regional ischemic LVCR to exercise.10 However, this study did not specifically investigate the subgroup of patients with LBBB and a globally abnormal LVCR to exercise.

Although the overall specificity of ESE for detecting significant OCAD was suboptimal for the entire study cohort, it should be emphasized that this resulted predominantly from the reduced diagnostic performance of ESE in patients with LBBB, who developed a globally abnormal LVCR. It should be highlighted that for patients with LBBB and a regionally abnormal LVCR, the positive predictive value of a regionally abnormal LVCR was excellent, and comparable with patients without LBBB (89%; 16/18 patients). Therefore, a regionally abnormal LVCR in the setting of LBBB predicts the presence of significant OCAD. The specific type of LVCR in patients with LBBB undergoing ESE (regional vs global) likely offers different diagnostic, and possibly prognostic, information, and warrants further investigation.

An additional caveat in the interpretation and extrapolation of the findings of the current study into clinical practice is the fact that resting LV systolic function, as assessed by LVEF, was heterogeneous in the study cohort. Resting LVEF was abnormal in 64 patients, whereas LVEF was >55% for 110 patients (Table 2). It could be argued that ESE is not a good investigation for the 37% of patients with resting LV systolic dysfunction, because wall thickening is not evaluated during graded exercise, but only immediately post–peak exercise. The possibility of a “biphasic response” cannot be excluded in the current study. Although this may not account for false‐positive results, it may contribute to false‐negative results. Therefore, our data would support the hypothesis that ESE would be most useful clinically in patients with LBBB, when resting global and regional are normal.

Unfortunately, prognostic information was not available in our cohort, and prognostic outcomes of patients with globally abnormal responses have not been studied to date. It is possible that a globally abnormal LVCR for patients with resting LBBB and no significant OCAD as identified by our study, may reflect early cardiomyopathy. It is interesting to observe that the patients in our cohort almost invariably had reduced early diastolic mitral annular tissue velocities on tissue Doppler imaging (Table 2), consistent with reduced LV long‐axis function.11, 12 This reinforces the concept that LBBB can be an early manifestation of cardiomyopathy and significant myocardial dysfunction.13, 14

4.2. Alternative testing

Established data support the use of dobutamine stress echocardiography15, 16 and MPI5, 15 for assessment of OCAD in patients with resting LBBB. Additionally, CTCA has a high sensitivity and specificity for detecting significant OCAD.17, 18 320‐detector‐row CTCA has been reported to have a sensitivity of 94% and specificity of 87% for detecting significant CAD, and it is not affected by resting LBBB on ECG.19

4.3. Study limitations

This study was a single‐center retrospective study. Not all patients in our cohort underwent anatomical testing for OCAD. Referral bias in the performance of subsequent anatomical testing likely affected the calculated diagnostic performance of ESE in patients with LBBB: only 13% (11/82 patients) with a normal LVCR underwent further anatomical testing. Additionally, the majority of patients in the globally abnormal and regionally abnormal stress response groups had reduced LV systolic function at rest. Some of these patients likely had cardiomyopathy at rest. A proportion of these patients would be expected to develop abnormal responses to stress in the absence of significant CAD, given the known effect of increased afterload and the expected decrease of coronary perfusion reserve in the setting of cardiomyopathy. Further, the determination of CAD severity on subsequent anatomical testing was based on visual assessment of experienced, independent cardiologists. The study would have been strengthened by quantitative coronary angiography data. Moreover, the full clinical risk‐factor profiles of individual patients and their medication regimens at the time of ESE were not available. However, we presented the largest cohort of patients in the literature specifically investigating the real‐life diagnostic utility of ESE for the assessment of OCAD in LBBB, with a significant proportion of our patients (67%, 62/92 patients) undergoing anatomical testing.

5. CONCLUSION

For patients with LBBB who develop a globally abnormal LVCR during ESE, the specificity of ESE for reliably excluding significant OCAD is significantly reduced. For patients with LBBB who develop a regionally abnormal LVCR during ESE, the diagnostic performance of ESE for reliably detecting significant OCAD remains excellent. Because the majority of abnormal responses to exercise were as a result of globally abnormal LVCR, to reliably exclude significant OCAD for patients with LBBB, especially for those with resting global or regional LV systolic dysfunction, an alternative imaging strategy may be appropriate.

Conflicts of interest

The authors declare no potential conflicts of interest.

Xu B, Dobson L, Mottram PM, Nasis A, Cameron J, Moir S. Is exercise stress echocardiography useful in patients with suspected obstructive coronary artery disease who have resting left bundle branch block? Clin Cardiol. 2018;41:360–365. 10.1002/clc.22875