Abstract
Background:
Global longitudinal strain (GLS) is a sensitive marker for identifying subclinical myocardial dysfunction in obstructive coronary artery disease (CAD). Little is known about the relationship between GLS and ischemia in patients with myocardial ischemia and no obstructive CAD (INOCA).
Objectives:
To investigate the relationship between resting GLS and ischemia on stress echocardiography (SE) in patients with INOCA.
Methods:
Left ventricular GLS was calculated offline on resting SE images at enrollment (n=144) and 1-year follow-up (n=120) in the CIAO-ISCHEMIA study, which enrolled participants with moderate or severe ischemia by local SE interpretation (>3 segments with new or worsening wall motion abnormality and no obstructive (<50% stenosis) CAD on coronary CT angiography.
Results:
GLS values were normal in 83.3% at enrollment and 94.2% at follow-up. GLS values were not associated with a positive SE at enrollment (GLS −21.5% positive SE vs. GLS −19.9% negative SE, p=0.443), or follow-up (GLS −23.2% positive SE vs. GLS −23.1% negative SE, p=0.859). Significant change in GLS was not associated with positive SE in follow-up (p=0.401). Regional strain was not associated with co-localizing ischemia at enrollment or follow-up. Changes in GLS and number of ischemic segments from enrollment to follow-up showed a modest but not clinically meaningful correlation (β=0.41, 95% CI 0.16, 0.67, p=0.002).
Conclusions:
In this cohort of INOCA patients, resting GLS values were largely normal and did not associate with the presence, severity or location of stress-induced ischemia. These findings may suggest the absence of subclinical myocardial dysfunction detectable by echocardiographic strain analysis at rest in INOCA.
Keywords: strain, echocardiography, ischemia, no obstructive disease, INOCA, stress testing
Graphical Abstract
Among a cohort of patients with ischemia on stress echocardiography and no obstructive coronary artery disease on cCTA, resting GLS values on stress echocardiogram did not predict stress test positivity or presence of ischemia at enrollment or 1-year follow up. Normal resting GLS defined as <18.0%. GLS was largely normal and did not provide additional diagnostic utility in INOCA.
Introduction
Individuals with ischemia and no obstructive coronary artery disease (INOCA) experience high symptom burden, impaired quality of life, and increased adverse cardiac events compared with asymptomatic individuals [1–3]. Multiple studies demonstrate a high prevalence of coronary microvascular dysfunction (CMD), coronary vasospasm, or both in INOCA [4–6]. However, there is poor correlation between angina and ischemia on both invasive and non-invasive cardiac testing[4, 7–10]. For example, in the CIAO-ISCHEMIA (Changes in Ischemia and Angina over One year in ISCHEMIA trial screen failures with no obstructive coronary artery disease on CT angiography) study of patients with INOCA as determined by stress echocardiography (SE) and coronary computed tomography angiography (CCTA), there was no association between the number of ischemic segments on SE and angina severity[10] and half of patients with moderate or severe ischemia at baseline had normal SE 1-year later[10].
Resting GLS has incremental diagnostic utility as compared with conventional wall motion assessment in discriminating the presence and location of obstructive coronary artery disease (CAD)[11, 12] among symptomatic patients undergoing resting echocardiography for suspected CAD. The diagnostic utility of resting GLS among individuals with INOCA has not been rigorously studied.
To address this gap, we leveraged the international, multi-center CIAO-ISCHEMIA study, to investigate 1) the frequency of abnormal GLS as a subclinical marker of myocardial dysfunction at rest, and 2) the relationship between resting GLS and the presence and severity of ischemia on SE in INOCA. We hypothesized that GLS would be associated with subclinical myocardial dysfunction and predict stress-induced ischemia among patients with INOCA.
Materials and Methods
Study Population
212 participants were enrolled from 39 participating international sites into the CIAO-ISCHEMIA study [10]. CIAO-ISCHEMIA participants were enrolled but not randomized into the International Study of Comparative Health Effectiveness with Medical and Invasive Approaches (ISCHEMIA) trial. They had ischemic symptoms (chest pain or other potential ischemic equivalent) and moderate to severe ischemia on SE (> 3 ischemic segments) as determined by the local enrolling site but no obstructive CAD on CCTA (no stenosis > 50% in major epicardial vessel). Patients with <50% stenosis in all epicardial vessels are unlikely to have flow limitation based on invasive measurement of fractional flow reserve[13]. Demographics were assessed at enrollment. Angina status, assessed by Seattle Angina Questionnaire (SAQ) [14], and ischemia by SE, were assessed at enrollment and 1-year follow-up. Only patients who had SE images technically suitable for GLS analysis were included (Figure 1). The study was approved by the NYU Grossman School of Medicine Institutional Review Board (IRB), and by each site’s local IRB or ethics committee. All participants provided written informed consent.
Figure 1 –
Study Flow Diagram
192 CIAO participants with enrollment and 1-year SE images were eligible for inclusion in this analysis. SE images were analyzed on 144 individuals at enrollment. 48 participants (25% of total eligible cohort) were excluded as resting images at enrollment were not suitable for strain analysis. At 1-year follow-up SE images were analyzed on 120 individuals. A further 24 participants were excluded between enrollment and 1-year follow-up due to poor image quality.
Echocardiographic Analysis
All SEs were evaluated by a blinded investigator (MHP) at the ISCHEMIA trial echocardiography core laboratory (Massachusetts General Hospital) for the presence, severity and location of ischemia based on standardized 16-segment myocardial segmentation model [15]. Moderate or severe ischemia was defined as the presence of stress-induced moderate or severe hypokinesis, akinesis or dyskinesis in > 3 segments [16]. Presence and severity of ischemia was based on core lab adjudication. Offline cardiac strain analysis using TomTec 2D-CPA software (version TTA 2.3) was performed by two investigators (EFD, DRC), blinded to patients’ SE results. Subendocardial GLS analysis was undertaken on apical two, three, and four chamber resting SE images. Strain analysis was performed if >12 of 16 cardiac segments were of adequate quality for analysis. Contrast enhanced images were included as the feasibility and accuracy of such analysis been previously established [17].
Regional longitudinal strain (RLS) was defined based on coronary vascular supply as follows: anterior region (left anterior descending artery): basal anterior, basal anterior septum, mid anterior, mid anterior septum, apical anterior, apical septal; inferior region (right coronary artery): basal inferoseptum, basal inferior, mid inferoseptum, mid inferior, apical inferior; and lateral region (left circumflex artery): basal inferolateral, basal anterolateral, mid inferolateral, mid anterolateral, apical lateral. A RLS value was calculated for each coronary territory and ischemia was defined as stress-induced wall motion abnormalities in at least 2 segments of that region. Normal strain was defined as < −18.0% based on published values [18–20]. We defined improvement in GLS as change toward a more negative value (farther from 0) and worsening in GLS as change toward a more positive value (closer to 0). A significant change in GLS from enrollment to 1-year was defined as >15.0% relative change and/or absolute change of >3 % [21]. Interobserver and intraobserver reproducibility of GLS was undertaken on a subset (15%, n = 22) of patients.
Statistical Methods
We computed descriptive statistics of baseline characteristics, stress test results and symptoms, presented as median and inter-quartile range (IQR) for continuous variables, and frequencies and percentages for categorical variables. We evaluated differences in patients’ stress test characteristics between enrollment and 1-year using the Wilcoxon rank-sum test for continuous variables and McNemar’s test for categorical variables. Comparison within each time point was conducted with Student’s t-test and Pearson’s chi-square test as appropriate.
We assessed whether changes in GLS measurements between baseline and follow-up were associated with key variables of interest, namely, stress test results, number of ischemic segments, and wall motion score index (WMSI). We fit separate linear mixed effects models of GLS with each variable, including a random intercept to account for potential correlations among measurements belonging to the same patient. To examine both the cross-sectional and the within-subject (longitudinal) associations between each key variable and GLS measurements, we included the mean value for each participant and the time-varying value centered at the participant-level mean as covariates [22, 23]. For example, for WMSI, we included the mean WMSI for a given participant, and the time-varying WMSI centered at the participant-level mean WMSI. Unadjusted modeling included only the variable of interest. In adjusted analysis, we extended each model to control for sex, age, body mass index (BMI), enrollment left ventricular ejection fraction (LVEF) at rest, SAQ Angina Frequency Score, and systolic blood pressure (SBP, in mmHg). To facilitate interpretation, WMSI was scaled for a 0.1-unit change, while LVEF and SBP were each scaled to a 10-unit change. All analyses were conducted in R software. To evaluate statistical significance, we set the 2-tailed type 1 error to 0.05.
Results
Enrollment Characteristics
144 patients had SE images suitable for GLS analysis at enrollment, and 120 of these had suitable SE images at both enrollment and 1-year follow-up (Figure 1). The ICC for interobserver reproducibility was 0.91 (95% CI 0.79–0.96) and the coefficient of variance was 2.56%. Intraobserver ICCs were 0.79 (DRC 95% CI 0.56–0.91) and 0.94 (EFD 95% CI 0.80–0.98), respectively. Among the 144 included individuals, 67% (n = 97) were female with a median age of 61 years (Table 1). Symptoms leading to SE and angina frequency are reported in Table 1. The median time from SE to enrollment was 63 days (IQR 35–132 days) Apart from a lower BMI among included patients (28.4 kg/m2 vs 30.9 kg/m2, p = 0.032), there were no significant differences in demographic, clinical or stress test characteristics between included versus excluded from analysis at either time point (tested variables as in Tables 1 and 2, data not shown).
Table 1 –
Enrollment Clinical and Demographic Characteristics
| Age (Years) | 61.0 (56 – 70) |
| Sex (Female) | 67.4 (n = 97) |
| Body Mass Index (kg/m2) | 28.4 (25.0 – 31.5) |
| Race | |
| White | 84.7% (n = 122) |
| Black | 5.6% (n = 8) |
| Asian | 6.9% (n = 10) |
| Other/Unknown | 2.8% (n =4) |
| Ethnicity | |
| Hispanic or Latino | 11.1% (n = 16) |
| Cardiovascular Risk Factors | |
| Hypertension | 62.5% (n = 90) |
| Diabetes | 15.4% (n = 22) |
| Active Smoking | 6.2% (n = 9) |
| Statin Therapy | 82.5% (n = 118) |
| LDL Cholesterol (mg/dL) | 99.0 (77.0 – 125.6) |
| Family History of premature CHD | 36.8% (n = 53) |
| Cardiovascular History | |
| Prior MI | 2.1% (n = 3) |
| Prior Revascularization | 4.9% (n = 7) |
| Heart Failure | 0.7% (n = 1) |
| Atrial Fibrillation | 4.2% (n = 6) |
| Valvular heart disease | 5.0% (n = 7) |
| Cardiovascular Medications | |
| Antiplatelet | 73.6% (n = 106) |
| Beta-blocker | 55.6% (n = 80) |
| Calcium Channel Blocker | 14.6% (n = 21) |
| Short Acting Nitrate | 10.4% (n = 15) |
| Long-acting nitrate | 9.7% (n =14) |
| Anticoagulant | 4.2% (n = 6) |
| Symptoms leading to stress echo | |
| Typical Chest pain | 49.3% (n = 71) |
| Atypical Chest pain | 34.0% (n = 49) |
| Shortness of Breath | 53.5% (n = 77) |
| Nausea | 2.8% (n = 4) |
| Sweating | 4.2% (n = 6) |
| Angina | |
| SAQ Angina Frequency Scale | 83 (66–93) |
| SAQ Angina Frequency Scale Score | 90 (70–100) |
| Angina Frequency at Enrollment | |
| None | 37.5% (n = 53) |
| Daily | 2.1% (n = 3) |
| Weekly | 13.6% (n = 19) |
| Monthly | 46.4% (n = 65) |
LDL = Low density Lipoprotein, MI = Myocardial infarction, SAQ = Seattle Angina Questionnaire
Table 2 –
Stress test Characteristics – Enrollment and Follow-up
| Enrollment | Follow up | p-value* | ||
|---|---|---|---|---|
| Whole Cohort (n = 144) | Participants with 1-year follow up. (n=120) | |||
| Pharmacological | 14.0 % (n = 20) | 15.1% (n = 18) | 20.0% (n = 24) | |
| MET Achieved (MET) | 7 (6.1 – 9.2) | 7 (6.1 – 9.2) | 7.9 (6.8 – 10.1) | 0.013 |
| 85% Maximum Heart Rate Achieved # | 82.1% (n = 115) | 84% (n = 97) | 75% (n = 90) | 0.076 |
| Baseline SBP (mmHg) | 139 (124 – 150) | 137 (123 – 150) | 134.5 (120,F 149) | 0.535 |
| Baseline DBP (mmHg) | 81 (76 – 87.5) | 81 (76 – 87) | 80 (70, 86) | 0.010 |
| Systolic Hypertension at Baseline | 48.9% (n = 67) | 45% (n = 51) | 45% (n = 54) | 1.00 |
| Diastolic Hypertension at Baseline | 20% (n = 23) | 19% (n = 19) | 18% (n = 20) | 1.00 |
| Systolic Blood pressure (Stress) | 170 (156 – 190) | 170 (156 – 190) | 166 (151.2, 189.8) | 0.109 |
| Diastolic Blood Pressure (Stress) | 84 (80 – 95) | 84 (80 – 95) | 80 (72, 90.2) | 0.014 |
| Hypertensive response to Stress | 16.2% (n = 22) | 16.1% (n = 18) | 18.0% (n = 21) | 1.00 |
| Resting LVEF | 62% (58–66) | 62% (58–66) | 63% (59–67) | 0.065 |
| LVEF Stress | 61% (58 – 67) | 61% (58–67) | 67% (60–72) | <0.001 |
| Stress Test Positive | 93.1% (n = 134) | 92% (n =111) | 42% (n = 51) | <0.001 |
| Number of Ischemic Segments | 4.0 (3.0 – 4.0) | 4.0 (2.8 – 4.0) | 0.0 (0.0 – 2.0) | <0.001 |
| Inferior and Lateral | 7.6% (n = 11) | 6.7% (n = 8) | 2.0% (n = 3) | 0.180 |
| Other | 7.0% (n = 10) | 7.6 (n=9) | 16.0% (n = 19) | 0.052 |
Reported p-values are for subjects with paired enrollment and one year follow-up data.
Refers to patients who underwent exercise testing only, Abbreviations: DBP = Diastolic Blood Pressure LVEF = Left Ventricular Ejection Fraction, MET = Metabolic Equivalent, SBP = Systolic Blood Pressure, MET
Stress test characteristics at enrollment and follow-up are presented in Table 2. At least 85% of the maximal predicted heart rate was achieved in 82.1% of individuals at enrollment and 75.0% at follow-up. A hypertensive response to stress was present in 16.2% at enrollment and 18.0% at follow-up.
While enrolling sites interpreted all participants as having moderate or severe ischemia, by core echocardiography lab adjudication, the SE was positive in 93.1% at enrollment and moderate or severe ischemia was present in 77.8%. At 1-year follow-up, SE was positive for ischemia in 42.5%, with moderate or severe ischemia at 1-year in 22.5%. At enrollment, there was no statistical difference in demographic or clinical characteristics between those with positive and negative SE (Table 3). At 1-year follow-up, those with positive SE had a higher (more favorable) SAQ angina frequency subscale score than those with negative SE (median 95, IQR 80–100 with positive SE, vs. 80, IQR 70–100 with negative SE, p = 0.030). Cardiovascular risk factors were not associated with SE positivity at either time point. At 1-year follow-up there was an increase in the use of calcium channel blockers from 17% at baseline to 29% at follow up (p=0.022). There were no other significant changes in medical therapy.
Table 3 –
Characteristics associated with abnormal stress test at enrollment and 1-year follow-up.
| Enrollment | 1-year follow up | |||||
|---|---|---|---|---|---|---|
| Stress test negative | Stress test positive | p-value | Stress test negative | Stress test positive | p-value | |
| Enrollment | N = 8 | N =136 | N = 69 | N =51 | ||
| Age (Years) | 59.0 (58.0 – 60.2) | 62 (55.8 – 70.0) | 0.450 | 61.0 (56.0–67.0) | 62.0 (55.0–71.5) | 0.505 |
| Sex (% female) | 50.0% (n = 4) | 68.4% (n = 93) | 0.438 | 66.7% (n = 46) | 70.6% (n = 36) | 0.796 |
| Other | 0% (n = 0) | 2.9% (n = 4) | 2.9% (n = 2) | 3.9% (n = 2) | ||
| Hypertension | 75.0% (n = 6) | 61.8% (n = 84) | 0.710 | 55.1% (n = 38) | 68.6% (n = 35) | 0.189 |
| Diabetes | 0% (n = 0) | 16.3% (n = 22) | 0.609 | 13.0% (n = 9) | 14.0 (n= 7) | 1.00 |
| BMI (kg/m2) | 30.3 (26.6 – 33.2) | 28.4 (25.0 – 31.2) | 0.297 | 28.1 (24.8 – 31.0) | 29.4 (26.5 – 32.2) | 0.154 |
| LDL (mg/dL) | 101.0 (83.1 – 107.0) | 99 (76.2 – 125.7) | 0.974 | 100.0 (75.9 – 124.5) | 99.0 (80.0 – 127.0) | 0.668 |
| Smoking | 0% (n = 0) | 6.6% (n = 9) | 1.00 | 5.8% (n = 4) | 5.9% (n = 3) | 1.00 |
| Prior Revascularization | 12.5% (n=1) | 4.4% (n = 6) | 0.336 | 5.8% (n = 4) | 3.9 (n = 2) | 1.00 |
| Monthly | 62.5% (n = 5) | 45.5 (n = 60) | 55.9% (n = 38) | 35.4% (n 17) | ||
| Long acting nitrate | 0% (n = 0) | 6.6% (n = 9) | 1.00 | 8.7 (n = 6) | 2% (n = 1) | 0.236 |
| GLS on Enrollment Echo | −19.95 (−22.10 to −18.68) | −21.50 (−24.07 to −19.09) | 0.443 | −21.00 (−23.82 to −19.11) | −22.02 (−24.46 to −18.50) | 0.627 |
| GLS on Follow-up Echo | −21.94 (−23.64 to −20.50) | −23.16 (−25.68 to −21.08) | 0.502 | −23.09 (−25.68 to −21.19) | 23.19 (−25.52 to −20.16) | 0.859 |
| Abnormal GLS | 25.0% (n = 2) | 16.2% (n = 22) | 0.621 | 4.3% (n = 3) | 7.8% (n = 4) | 0.456 |
Resting GLS
Median GLS was −21.4% (−24.1% to −19.0%) at enrollment and −23.1% (−25.6% to −20.9%) at 1-year follow-up. GLS was normal in the majority at both enrollment (83.3%, n = 120) and follow-up (94.2%, n = 113). Clinical factors associated with normal and abnormal enrollment GLS are presented in Table 4. Abnormal GLS was associated with higher BMI (31.0 vs. 28.1 kg/m2, p = 0.023) and Hispanic ethnicity (p = 0.006).
Table 4 –
Clinical, Demographic and Stress Test Factors associated with Normal and Abnormal GLS at Enrollment.
| Normal Strain (n = 120) |
Abnormal strain (n = 24) |
p-value | |
|---|---|---|---|
| Age (Years) | 61 (56, 70) | 61.5 (56.2, 70) | 0.944 |
| Sex (Female)) | 70.8% (n = 90) | 50.0% (n = 12) | 0.080 |
| Other | 2.5% (n = 3) | 4.2% (n = 1) | |
| Hispanic or Latino | 7.5% (n = 9) | 29.2% (n = 7) | 0.006 |
| Hypertension | 60.8% (n = 73) | 70.8% (n = 17) | 0.488 |
| Diabetes | 15.1% (n = 18) | 16.7% (n = 4) | 0.765 |
| BMI (kg/m2) | 28 (24.7 – 31.1) | 30.9 (28 – 32.7) | 0.023 |
| LDL (mg/dL) | 94.0 (74.9 – 121.5) | 115.3 (93.1 – 138.1) | 0.036 |
| Smoking | 5.8% (n = 7) | 8.3% (n = 2) | 0.645 |
| Family History of Premature Coronary Heart Disease | 38.3% (n = 46) | 29.2% (n = 7) | 0.237 |
| Prior MI | 2.5% (n = 3) | 0% (n = 0) | 1.000 |
| Prior Revascularization (%) | 5.0% (n = 6) | 4.2% (n = 1) | 1.000 |
| Sweating | 2.5% (n = 3) | 12.5% (n = 3) | 0.058 |
| LVEF at Rest (%) | 63 (59 – 66) | 57 (55 – 60.2) | <0.001 |
| LVEF with Stress (%) | 61 (57.5 – 67) | 60 (57.5 – 67.2) | 0.468 |
| Exercise Stress Test | 85.7% (n = 102) | 87.5% (n = 21) | 1.000 |
| Exercise Time (secs) | 395 (300 – 510.8) | 420 (345 – 563) | 0.472 |
| MET | 7.0 (6.1 – 9.4) | 7.6 (6.1 – 8.4) | 0.879 |
| Baseline DBP (mmHg) | 80 (76 – 89) | 82 (79.2 – 86.2) | 0.855 |
| Peak Diastolic BP (mmHg) | 85 (80 – 95) | 80.0 (77.5 – 88.5) | 0.173 |
| Hypertensive Response to Stress | 16.7% (n = 19) | 13.6% (n =3) | 1.000 |
| Stress Test Positive | 93.3% (n = 112) | 91.7% (n = 22) | 0.673 |
| Ischemic Segments (n) | 3 (3 – 4) | 4 (2.8 – 5) | 0.232 |
| WMSI | 1.3 (1.3 – 1.4) | 1.4 (1.2 – 1.5) | 0.410 |
| Other | 5.9% (n = 7) | 12.5% (n = 3) | 0.372 |
| SAQ Angina frequency scale score | 90 (70 – 100) | 90 (80 – 100) | 0.271 |
| Statin | 84.9% (n = 101) | 70.8% (n = 17) | 0.137 |
| GLS | −22.4 (−24.6 to −20.1) | −16.7 (−17.5 to −16.1) | Not applicable |
The median change in GLS between enrollment and follow-up was −2.1 percentage points (−4.3 to 1.1). GLS improved in 38.3% of individuals (n = 46, median change in GLS −5.2%) and worsened in 10.8% of individuals (n = 13, median change in GLS +4.3%).
GLS and Stress Test Results
At enrollment, GLS was not significantly different between those with a positive or negative SE (GLS −21.5% positive SE vs. GLS −19.9% negative SE, p = 0.443, Figure 2). Abnormal GLS was not more common in those with a positive SE (p = 0.673, Table 3) or in those with moderate or severe ischemia on SE (p = 0.093) compared to those with mild or no ischemia, although few patients had abnormal GLS (16.7%, n = 24). There was no correlation between GLS and number of ischemic segments (rho = 0.079, p = 0.344) or stress WMSI (rho = 0.078, p = 0.352) at enrollment.
Figure 2 –
Resting GLS at Enrollment and Enrollment Stress Echocardiogram Outcome
Resting GLS values at enrollment in individuals with a positive (n=136 in red) and negative (n=8 in blue) SE enrollment. Values were not significantly different between those with a positive or negative stress echocardiogram (p=0.443).
At 1-year follow-up, there was no difference in GLS between those with a positive or negative SE at that time point (GLS −23.2% positive SE vs. GLS −23.1% negative SE, p = 0.859, Figure 3). Abnormal GLS was not more common in participants with a positive SE at 1-year. (p = 0.456, Table 3). Moderate or severe ischemia was not more common (p = 0.654) among the patients with abnormal GLS at 1 year (5.8%, n = 7). There was no correlation between 1-year GLS and 1-year number of ischemic segments (rho = −0.002, p = 0.984) or WMSI (rho = −0.005, p = 0.954). Enrollment GLS did not differ between those with positive and negative SE at 1-year follow-up (GLS −22.0% positive SE vs. GLS −21.1% negative SE, p = 0.678).
Figure 3:
Resting 1-Year Global Longitudinal Strain and 1-Year Stress Test Outcome
GLS values at 1-year follow-up in individuals with a positive (n = 51 in red) and negative (n = 69 in blue) SE at 1-year follow-up. Values were not significantly different between those with a positive or negative SE (p=0.859).
Regional Longitudinal Strain and Regional Ischemia
RLS was not significantly different in corresponding areas of regional ischemia at enrollment or follow-up (Table 5). RLS was not worse in the presence of corresponding segmental ischemia at either enrollment or follow-up.
Table 5 –
Regional Longitudinal Strain in the Presence and Absence of Corresponding Regional Ischemia at Enrollment and 1-Year Follow-up
| Enrollment | 1-Year Follow-up | |||||
|---|---|---|---|---|---|---|
| Anterior Ischemia Absent (n = 79) |
Anterior Ischemia Present (n = 63) |
p-value | Anterior Ischemia Absent (n = 101) |
Anterior Ischemia Present (n = 18) |
p-value | |
| Anterior Regional Strain | −21.7 (−25.9, −19.7) | −22.2 (−25.7, −19.5) | 0.833 | −23.9 (−27.0, −20.7) | −24.3 (−27.9, −19.9) | 0.982 |
| Lateral Ischemia Absent (n = 101) |
Lateral Ischemia Present (n = 39) |
p-value | Lateral Ischemia Absent (n = 112) |
Lateral Ischemia Present (n = 6) |
p-value | |
| Lateral Regional Strain | −21.0 (−24.0, −16.9) | −21.3 (−24.5, −18.9) | 0.671 | −21.8 (−24.3, −19.6) | −23.2 (−25.8, −19.9) | 0.557 |
| Inferior Ischemia Absent (n = 87) |
Inferior Ischemia Present (n = 54) |
p-value | Inferior Ischemia Absent (n = 104) |
Inferior Ischemia Present (n = 12) |
p-value | |
| Inferior Regional Strain | −18.5 (−21.0, −16.2) | −17.7 (−20.9, −16.0) | 0.678 | −19.9 (−21.8, −17.3) | −19.6 (−20.7, −18.1) | 0.906 |
Change in GLS and Stress Test Results
Patients who experienced a significant change in GLS between enrollment and follow-up did not show differences in the rates of positive SE (p = 0.401), number of ischemic segments (p = 0.545), or stress WMSI (p = 0.544) at 1-year compared to those who did not experience a significant change in GLS.
Longitudinal Modeling of GLS
There was no significant cross-sectional association across the entire population between GLS and ischemia on SE whether ischemia was defined as a positive stress test (unadjusted β=−0.43, p=0.669), the number of ischemic segments (unadjusted β=0.11, p=0.516), or the stress WMSI (unadjusted β=0.15, p=0.372). For example, a 0.1-unit increase in stress WMSI was not associated with mean GLS. In contrast, for an individual patient, longitudinal change in these ischemia measures was associated with change in GLS between baseline and follow-up, but the magnitude of the associations was small: change in SE positivity (unadjusted β=1.91, p=0.001), change in number of ischemic segments (unadjusted β=0.43, p<0.001), and the stress WMSI (unadjusted β=0.44, p=0.001). For both the population as a whole and for individual patients, the associations of GLS and exposures of interest were largely unchanged after adjustment for hypothesized confounders (Supplemental Table 1). LVEF and SBP were independently associated with GLS in all models. Unadjusted and adjusted models for the relationship between GLS and ischemic and other variables are presented in Supplemental Table 1.
Discussion
Although GLS has been associated with improved diagnostic accuracy in patients with obstructive CAD, our study may suggest that in INOCA, GLS is 1) largely normal, 2) not correlated with inducible ischemia, and 3) not predictive of SE findings.
In this cohort of patients with INOCA, GLS values were normal in 83% at enrollment and 94% at follow-up. As thresholds for normal GLS ranges have not been previously defined in this population, we defined abnormal GLS as worse than −18.0% based on accepted values among healthy individuals [19, 20, 24]. Women, who represented two- thirds of our cohort, have been reported to have higher GLS values than men [19, 25]; however, in our cohort, female sex was not associated with GLS. Whereas GLS is load dependent and higher systemic blood pressures associate with worse GLS values [19], we found that hypertension was not associated with abnormal GLS, potentially reflecting the reasonable level of BP control in this cohort at the time of SE. It is possible that that on an individual patient level changes in blood pressure between baseline and follow-up were also responsible for the improvement in GLS between baseline and follow-up seen in 38% of the population and the worsening seen in 11% of participants, however our relatively small numbers meant that such changes did not reach statistical significance. Given the influence of systolic blood pressure, assessment of myocardial work in the INOCA population may be of interest in the future. Although further study is warranted, our results provide insight regarding the range of GLS values among patients with INOCA.
In INOCA, myocardial ischemia occurs secondary to coronary microvascular dysfunction (CMD) from either fixed structural remodeling of the microvasculature, functional dynamic obstruction of the microcirculation, or epicardial vasospasm [2, 26]. Multiple studies have demonstrated worse GLS values among patients with obstructive CAD suggestive of myocardial dysfunction at rest secondary to irreversible myocardial injury and/or fibrosis resulting from repetitive ischemia [11, 12]. A similar mechanism has been hypothesized to underlie the observation that patients with INOCA have increased risk of heart failure with preserved ejection fraction [27]. The predominantly normal GLS values seen in our population, despite inducible ischemia on SE, argue against the presence of subclinical myocardial dysfunction, detectable by echocardiographic strain analysis at rest, in the majority of individuals. Normal myocardial function at rest among patients with INOCA is further supported by the absence of myocardial fibrosis on cardiac magnetic resonance imaging (cMRI) among symptomatic patients with no obstructive CAD and CMD [7, 28].
CMD among patients with INOCA is typically thought to cause diffuse subendocardial ischemia. Accordingly, non-invasive stress imaging including SE [9, 29] has previously demonstrated significant variability in the detection of ischemia among patients with no obstructive CAD and symptoms of typical angina (previously termed cardiac syndrome X). It remains to be determined whether regional ischemia on SE or single-photon emission CT in the absence of obstructive CAD, as observed in up to 20% of patients with moderate or severe ischemia enrolled in ISCHEMIA, represents the same process as global ischemia as detected by stress CMR imaging or positron emission tomography[10, 28, 30]. We observed resolution of ischemia in 50% of our cohort at follow-up despite similar stress test characteristics and no significant changes in treatment. This variation likely reflects the heterogeneity in pathophysiological mechanisms among this patient population, the dynamic nature of CMD, and the lower sensitivity of SE for detecting global subendocardial ischemia as compared with regional ischemic changes more commonly associated with obstructive CAD [31, 32]. We hypothesized that GLS, as a global measure, might be associated with persistence of ischemic abnormalities at 1-year follow-up in the CIAO cohort, but this was not the case. The observed correlation between individual patients’ changes in GLS and changes in stress test positivity, number of ischemic segments, and change in WMSI were in the clinically expected direction (worsening GLS was associated with more ischemia). These changes were statistically significant but the magnitude of these changes were too small to be considered clinically relevant (<1.2% absolute difference in GLS) in an individual patient. These within-patient longitudinal trends serve to strengthen the overall plausibility of our results, however, the lack of association between GLS and ischemia on a population-level continues to suggest that GLS may not be a useful predictor of ischemia on SE in this population.
Another possible explanation for the findings might be that GLS assesses subendocardial fibers which are affected in obstructive CAD since ischemia due to this process affects the subendocardium first. In contrast, INOCA may occur due to CMD which can spare the subendocardium explaining the normal GLS values we observed.
Our results should be interpreted in the context of several limitations. We performed post hoc off-line strain analysis on images acquired for the primary purpose of evaluating myocardial ischemia. These images were not optimized for strain analysis and therefore as noted in Figure 1, 25% of CIAO participants were excluded due to inadequate image quality which could have led to attrition bias. However, the percentage of patients excluded was similar to prior studies undertaking strain analyses [25], and with the exception of BMI, there were no differences in characteristics between the included and excluded patient groups. We acknowledge that CMD has been previously associated with impairment of GLS at stress[33], however, we set out to evaluate if there was incremental value to the use of GLS at rest to predict ischemia among patients with INOCA. We performed strain analysis only on resting images given our post-hoc analysis was limited by variability of quality of stress images, in addition to the inherent technical challenges posed with measuring strain at peak stress, including suboptimal tracking in the setting of increased heart rate, hypercontractility, and excessive annular motion. A definitive cut off for normal GLS values has not been defined, GLS ranging from −18% to −22% has been suggested in the literature among healthy individuals [34, 35]. We used −18.0% as the cut off for normal GLS which was conservative based on the comparatively higher GLS values generated by the TomTec software in comparison to other strain software packages [36]. Given known intervendor variability in GLS measurements, future studies using other GLS software packages will be needed to confirm our findings. As we were limited by our small cohort, we did not stratify GLS based on severity of ischemia and may be underpowered to detect differences in GLS between those with positive and negative SE particularly at enrollment. The findings related to SAQ should be interpreted with caution as the range of angina frequencies was not wide enough to allow for robust analysis, specifically 83.6% of the population reported monthly or no angina at follow-up. Finally, it is important to acknowledge that given the imperfect sensitivity and specificity of stress echocardiography, some participants may not have true INOCA. However, the ischemia severity required for study entry and high reproducibility of blinded assessment at the ISCHEMIA core laboratory limit this limitation.
Although it is possible that the sensitivity of GLS at rest could vary depending on the severity of stress-induced ischemia present, the majority of our cohort had at least moderate ischemia making this less likely. Although CMD is hypothesized to be the primary driver of ischemia in the CIAO cohort, we did not perform formal testing of invasive or noninvasive coronary flow for the diagnosis of CMD thus limiting our ability to associate our findings directly with differences in coronary flow characteristics.
Conclusion
Among these individuals with INOCA, GLS was mostly normal and not associated with the presence or severity of inducible ischemia on SE at enrollment or follow-up. GLS at enrollment was not associated with ischemia on SE at follow-up.
Supplementary Material
Highlights.
Among patients with INOCA, resting GLS is largely normal.
GLS did not predict the presence, severity or location of stress-induced ischemia.
Resting GLS was not discriminatory for presence or severity of inducible ischemia.
Sources of Funding
This project is supported by NIH grants R01HL119153, U01HL105907, U01HL105462, U01HL105561, U01HL10565.
Disclosures
Esther F. Davis declares no conflicts of interest.
Daniela R. Crousillat declares no conflicts of interest.
Dr. Jesus Peteiro reports grants from National Heart, Lung and Blood Institute during the conduct of the study.
Dr. Lopez-Sendon reports grants from the National Heart, Lung, and Blood Institute during the conduct of the study; grants from Bayer; grants and personal fees from Pfizer; personal fees from Menarini; grants and personal fees from Sanofi; grants from Merck; grants and personal fees from Boeringher Infleheim; and grants from Amgen outside the submitted work.
Dr. Roxy Senior reports National Heart, Lung and Blood Institute during the conduct of the study; he also reports speaker fees from Lantheus Medical Imaging, Boston, Mass, Bracco, Milan, Italy, Philips Healthcare, Eindhoven, Holland.
Dr. Michael Shapiro reports grants from National Heart, Lung and Blood Institute during the conduct of the study; He serves as Scientific Advisory Board for Regeneron and Amgen.
Dr. Patricia Pellikka reports grants from National Heart, Lung and Blood Institute during the conduct of the study.
Dr. Radmila Lyubarova reports no conflicts of interest.
Dr. Khaled Alfakih reports grants from National Heart, Lung and Blood Institute during the conduct of the study.
Dr. Khaled Abdul-Nour reports grants from National Heart, Lung and Blood Institute during the conduct of the study.
Dr. Rebecca Anthopolos reports grants from National Heart, Lung and Blood Institute during the conduct of the study.
Yifan Xu reports grants from National Heart, Lung, and Blood Institute, duringthe conduct of the study.
Dr. Jerome Fleg reports no conflict of interest.
Dr. John Spertus reports grants from National Heart, Lung and Blood Institute, during the conduct of the study; personal fees from Bayer, Novartis, AstraZeneca, Amgen, Janssen, and United Healthcare; grants from American College of Cardiology, outside the submitted work; In addition, Dr. Spertus has a patent Copyright to Seattle Angina Questionnaire with royalties paid and Board of Directors for Blue Cross Blue Shield of Kansas City and Equity in Health Outcomes Sciences.
Dr. Judith S. Hochman is PI for the ISCHEMIA trial for which, in addition to support by National Heart, Lung, and Blood Institute grant, devices and medications were provided by Abbott Vascular; Medtronic, Inc.; Abbott Laboratories (formerly St. Jude Medical, Inc); Royal Philips NV (formerly Volcano Corporation); Arbor Pharmaceuticals, LLC; AstraZeneca Pharmaceuticals, LP; Merck Sharp & Dohme Corp.; Omron Healthcare, Inc, Sunovion Pharmaceuticals, Inc. Espero BioPharma; and Amgen, Inc; and financial donations from Arbor Pharmaceuticals LLC and AstraZeneca Pharmaceuticals LP.
Dr. David Maron grants from National Heart, Lung and Blood Institute during the conduct of the study.
Dr. Michael Picard reports grants from National Heart, Lung and Blood Institute during the conduct of the study.
Dr. Harmony Reynolds reports grants from National Heart, Lung and Blood Institute during the conduct of the study; non-financial support from Abbott Vascular, non-financial support from BioTelemetry and Siemens, outside the submitted work.
Abbreviations:
- CIAO-ISCHEMIA
Changes in Ischemia and Angina over One year among ISCHEMIA trial screen failures with no obstructive coronary artery disease on CT angiography
- GLS
global longitudinal strain
- INOCA
ischemia and no obstructive coronary artery disease
- SE
stress echocardiography
- WMSI
Wall motion score index
References
- [1].Jespersen L, Hvelplund A, Abildstrøm SZ, Pedersen F, Galatius S, Madsen JK, et al. Stable angina pectoris with no obstructive coronary artery disease is associated with increased risks of major adverse cardiovascular events. Eur Heart J. 2012;33:734–44. [DOI] [PubMed] [Google Scholar]
- [2].Kunadian V, Chieffo A, Camici PG, Berry C, Escaned J, Maas A, et al. An EAPCI Expert Consensus Document on Ischaemia with Non-Obstructive Coronary Arteries in Collaboration with European Society of Cardiology Working Group on Coronary Pathophysiology & Microcirculation Endorsed by Coronary Vasomotor Disorders International Study Group. Eur Heart J. 2020;41:3504–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Gulati M, Cooper-DeHoff RM, McClure C, Johnson BD, Shaw LJ, Handberg EM, et al. Adverse cardiovascular outcomes in women with nonobstructive coronary artery disease: a report from the Women’s Ischemia Syndrome Evaluation Study and the St James Women Take Heart Project. Arch Intern Med. 2009;169:843–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Sara JD, Widmer RJ, Matsuzawa Y, Lennon RJ, Lerman LO, Lerman A. Prevalence of Coronary Microvascular Dysfunction Among Patients With Chest Pain and Nonobstructive Coronary Artery Disease. JACC Cardiovascular interventions. 2015;8:1445–53. [DOI] [PubMed] [Google Scholar]
- [5].Reis SE, Holubkov R, Conrad Smith AJ, Kelsey SF, Sharaf BL, Reichek N, et al. Coronary microvascular dysfunction is highly prevalent in women with chest pain in the absence of coronary artery disease: results from the NHLBI WISE study. American heart journal. 2001;141:735–41. [DOI] [PubMed] [Google Scholar]
- [6].Ford TJ, Stanley B, Good R, Rocchiccioli P, McEntegart M, Watkins S, et al. Stratified Medical Therapy Using Invasive Coronary Function Testing in Angina: The CorMicA Trial. J Am Coll Cardiol. 2018;72:2841–55. [DOI] [PubMed] [Google Scholar]
- [7].Mygind ND, Michelsen MM, Pena A, Frestad D, Dose N, Aziz A, et al. Coronary Microvascular Function and Cardiovascular Risk Factors in Women With Angina Pectoris and No Obstructive Coronary Artery Disease: The iPOWER Study. J Am Heart Assoc. 2016;5:e003064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Cassar A, Chareonthaitawee P, Rihal CS, Prasad A, Lennon RJ, Lerman LO, et al. Lack of correlation between noninvasive stress tests and invasive coronary vasomotor dysfunction in patients with nonobstructive coronary artery disease. Circ Cardiovasc Interv. 2009;2:237–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Nihoyannopoulos P, Kaski JC, Crake T, Maseri A. Absence of myocardial dysfunction during stress in patients with syndrome X. J Am Coll Cardiol. 1991;18:1463–70. [DOI] [PubMed] [Google Scholar]
- [10].Reynolds HR, Picard MH, Spertus JA, Peteiro J, Lopez Sendon JL, Senior R, et al. Natural History of Patients With Ischemia and No Obstructive Coronary Artery Disease: The CIAO-ISCHEMIA Study. Circulation. 2021;144:1008–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Montgomery DE, Puthumana JJ, Fox JM, Ogunyankin KO. Global longitudinal strain aids the detection of non-obstructive coronary artery disease in the resting echocardiogram. Eur Heart J Cardiovasc Imaging. 2012;13:579–87. [DOI] [PubMed] [Google Scholar]
- [12].Biering-Sørensen T, Hoffmann S, Mogelvang R, Iversen AZ, Galatius S, Fritz-Hansen T, et al. Myocardial Strain Analysis by 2-Dimensional Speckle Tracking Echocardiography Improves Diagnostics of Coronary Artery Stenosis in Stable Angina Pectoris. Circulation: Cardiovascular Imaging. 2014;7:58–65. [DOI] [PubMed] [Google Scholar]
- [13].Ahmadi A, Stone GW, Leipsic J, Serruys PW, Shaw L, Hecht H, et al. Association of Coronary Stenosis and Plaque Morphology With Fractional Flow Reserve and Outcomes. JAMA Cardiology. 2016;1:350–7. [DOI] [PubMed] [Google Scholar]
- [14].Spertus JA, Winder JA, Dewhurst TA, Deyo RA, Prodzinski J, McDonell M, et al. Development and evaluation of the Seattle Angina Questionnaire: a new functional status measure for coronary artery disease. J Am Coll Cardiol. 1995;25:333–41. [DOI] [PubMed] [Google Scholar]
- [15].Pellikka PA, Arruda-Olson A, Chaudhry FA, Chen MH, Marshall JE, Porter TR, et al. Guidelines for Performance, Interpretation, and Application of Stress Echocardiography in Ischemic Heart Disease: From the American Society of Echocardiography. J Am Soc Echocardiogr. 2020;33:1–41.e8. [DOI] [PubMed] [Google Scholar]
- [16].Shaw LJ, Berman DS, Picard MH, Friedrich MG, Kwong RY, Stone GW, et al. Comparative definitions for moderate-severe ischemia in stress nuclear, echocardiography, and magnetic resonance imaging. JACC Cardiovascular imaging. 2014;7:593–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Medvedofsky D, Lang RM, Kruse E, Guile B, Weinert L, Ciszek B, et al. Feasibility of Left Ventricular Global Longitudinal Strain Measurements from Contrast-Enhanced Echocardiographic Images. J Am Soc Echocardiogr. 2018;31:297–303. [DOI] [PubMed] [Google Scholar]
- [18].Mirea O, Pagourelias ED, Duchenne J, Bogaert J, Thomas JD, Badano LP, et al. Variability and Reproducibility of Segmental Longitudinal Strain Measurement: A Report From the EACVI-ASE Strain Standardization Task Force. JACC Cardiovascular imaging. 2018;11:15–24. [DOI] [PubMed] [Google Scholar]
- [19].Yingchoncharoen T, Agarwal S, Popović ZB, Marwick TH. Normal ranges of left ventricular strain: a meta-analysis. J Am Soc Echocardiogr. 2013;26:185–91. [DOI] [PubMed] [Google Scholar]
- [20].Yang H, Wright L, Negishi T, Negishi K, Liu J, Marwick TH. Research to Practice: Assessment of Left Ventricular Global Longitudinal Strain for Surveillance of Cancer Chemotherapeutic-Related Cardiac Dysfunction. JACC: Cardiovascular Imaging. 2018;11:1196–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Plana JC, Galderisi M, Barac A, Ewer MS, Ky B, Scherrer-Crosbie M, et al. Expert consensus for multimodality imaging evaluation of adult patients during and after cancer therapy: a report from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2014;27:911–39. [DOI] [PubMed] [Google Scholar]
- [22].Wang LP, Maxwell SE. On disaggregating between-person and within-person effects with longitudinal data using multilevel models. Psychol Methods. 2015;20:63–83. [DOI] [PubMed] [Google Scholar]
- [23].Herring AH. Applied Longitudinal Analysis, 2nd Edition, by Fitzmaurice Garrett M., Laird Nan M., and Ware James H., John Wiley & Sons, 2011. Journal of Biopharmaceutical Statistics. 2013;23:940–1. [Google Scholar]
- [24].Potter E, Marwick TH. Assessment of Left Ventricular Function by Echocardiography: The Case for Routinely Adding Global Longitudinal Strain to Ejection Fraction. JACC: Cardiovascular Imaging. 2018;11:260–74. [DOI] [PubMed] [Google Scholar]
- [25].Sugimoto T, Dulgheru R, Bernard A, Ilardi F, Contu L, Addetia K, et al. Echocardiographic reference ranges for normal left ventricular 2D strain: results from the EACVI NORRE study. Eur Heart J Cardiovasc Imaging. 2017;18:833–40. [DOI] [PubMed] [Google Scholar]
- [26].Khuddus MA, Pepine CJ, Handberg EM, Bairey Merz CN, Sopko G, Bavry AA, et al. An intravascular ultrasound analysis in women experiencing chest pain in the absence of obstructive coronary artery disease: a substudy from the National Heart, Lung and Blood Institute-Sponsored Women’s Ischemia Syndrome Evaluation (WISE). Journal of interventional cardiology. 2010;23:511–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Taqueti VR, Solomon SD, Shah AM, Desai AS, Groarke JD, Osborne MT, et al. Coronary microvascular dysfunction and future risk of heart failure with preserved ejection fraction. European Heart Journal. 2017;39:840–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Rahman H, Ryan M, Lumley M, Modi B, McConkey H, Ellis H, et al. Coronary Microvascular Dysfunction Is Associated With Myocardial Ischemia and Abnormal Coronary Perfusion During Exercise. Circulation. 2019;140:1805–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Saghari M, Assadi M, Eftekhari M, Yaghoubi M, Fard-Esfahani A, Malekzadeh JM, et al. Frequency and severity of myocardial perfusion abnormalities using Tc-99m MIBI SPECT in cardiac syndrome X. BMC nuclear medicine. 2006;6:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Murthy VL, Naya M, Taqueti VR, Foster CR, Gaber M, Hainer J, et al. Effects of sex on coronary microvascular dysfunction and cardiac outcomes. Circulation. 2014;129:2518–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Panting JR, Gatehouse PD, Yang GZ, Grothues F, Firmin DN, Collins P, et al. Abnormal subendocardial perfusion in cardiac syndrome X detected by cardiovascular magnetic resonance imaging. N Engl J Med. 2002;346:1948–53. [DOI] [PubMed] [Google Scholar]
- [32].Thomson LEJ, Wei J, Agarwal M, Haft-Baradaran A, Shufelt C, Mehta PK, et al. Cardiac Magnetic Resonance Myocardial Perfusion Reserve Index Is Reduced in Women With Coronary Microvascular Dysfunction. Circulation: Cardiovascular Imaging. 2015;8:e002481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].Rodriguez-Zanella H, Arbucci R, Fritche-Salazar JF, Ortiz-Leon XA, Tuttolomondo D, Lowenstein DH, et al. Vasodilator Strain Stress Echocardiography in Suspected Coronary Microvascular Angina. J Clin Med. 2022;11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Galderisi M, Cosyns B, Edvardsen T, Cardim N, Delgado V, Di Salvo G, et al. Standardization of adult transthoracic echocardiography reporting in agreement with recent chamber quantification, diastolic function, and heart valve disease recommendations: an expert consensus document of the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2017;18:1301–10. [DOI] [PubMed] [Google Scholar]
- [35].Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2015;28:1–39.e14. [DOI] [PubMed] [Google Scholar]
- [36].Farsalinos KE, Daraban AM, Ünlü S, Thomas JD, Badano LP, Voigt JU. Head-to-Head Comparison of Global Longitudinal Strain Measurements among Nine Different Vendors: The EACVI/ASE Inter-Vendor Comparison Study. J Am Soc Echocardiogr. 2015;28:1171–81, e2. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.