Myocardial strain analysis by 4D-speckle tracking echocardiography for prediction of coronary artery disease severity in patients with stable angina pectoris

Abstract

Objective

We sought to evaluate the myocardial strain by four-dimensional speckle-tracking echocardiography (4D-STE) in patients with stable angina pectoris (SAP) to determine the severity of coronary artery disease (CAD) based on the Gensini score.

Methods

The present study comprised of 150 patients with SAP. Patients with history of SAP, normal left ventricular ejection fraction, and without regional wall motion abnormalities (RWMA) were scheduled for elective coronary angiography. Based on Gensini score, there were two groups: non-critical stenosis group [Gensini score (0–19), n = 117] and critical stenosis group [Gensini score ≥20, n = 33]. Correlation between Gensini score and 4D-STE strain parameters were investigated.

Results

Out of 150 patients, critical stenosis group had significantly depressed values of all 4D-STE strain parameters than non-critical stenosis group (p < 0.001), except global radial strain (GRS) parameter. Significant positive correlation was found between Gensini score and 4D global longitudinal strain (GLS), global circumferential strain (GCS), global area strain (GAS) with Spearman's correlation coefficient (ρ) as 0.626, 0.548, and 0.631, respectively (p < 0.001), whereas significant negative correlation was found between Gensini score and GRS (ρ = −0.433, p < 0.001). A 4D GLS value of ≥ −17 had 84.9% sensitivity and 97.4% specificity, GAS ≥ −31 (90.9% sensitivity, 78.6% specificity), GCS ≥ −17 (69.7% sensitivity, 92.3% specificity), and GRS <47 (sensitivity 72.7%, specificity 76.1%) to detect critical CAD described by Gensini score ≥20.

Conclusion

The 4D-STE can aid in the assessment of severe CAD stenosis with good sensitivity and specificity in the patients with SAP without RWMA on traditional echocardiography.

1. Introduction

Coronary artery disease (CAD) is the third leading cause of death worldwide, with 17.8 million fatalities occurring per year.¹ Prompt identification and intervention (medical or invasive therapies) to maintain left ventricular (LV) systolic functions is critical in patients with CAD. Generally, the regional wall motion abnormalities (RWMA) and decreased left ventricular ejection fraction (LVEF) are the most important variables measured for the clinical diagnosis of CAD in two-dimensional (2D) echocardiography. Furthermore, despite being a gold standard parameter for the measurement of LV systolic function, LVEF is ineffective for the detection of subclinical myocardial damage due to load dependence and complex LV geometry.^2,3 Thus, developing a more sensitive index for subclinical LV dysfunction is of paramount importance.

Strain analysis by speckle-tracking technology, 2D speckle-tracking echocardiography (2D-STE), three-dimensional speckle-tracking echocardiography (3D-STE) and four-dimensional speckle-tracking echocardiography (4D-STE) accurately and thoroughly measures myocardial mechanics as well as the deformation of each myocardial segment. The 4D-STE uses LV longitudinal, circumferential and radial strains which are acquired in real time to accurately quantify the myocardial deformation. Hence, it overcomes the limitations of 2D-STE i.e., out-of-plane speckle movements, foreshortening and geometric assumptions and the need to obtain several views of the LV from different acoustic windows at different times with no specific anatomical landmarks. Moreover, it can detect strain dynamics in every myocardial segment using a 3D reconstruction of the LV and color mapping throughout the cardiac cycle. It can display fluctuations in strain in each LV segment in real time and has benefits in accessibility, feasibility and cost.3, 4, 5

Myriad of scoring systems are available for describing atherosclerosis burden i.e., the Gensini score, the Coronary Artery Surgery Study score and the Duke CAD severity index. The Gensini score is amongst the most widely utilized in clinical practice.⁶ Until now far too little attention has been paid to the relationship between 4D-STE strain parameters and CAD burden evaluated by Gensini score. Thus, the present study was designed to analyze the myocardial strain by 4D-STE in patients with stable angina pectoris (SAP) to predict severity of CAD based on the Gensini score.

2. Material and methods

2.1. Study population

This was a cross-sectional single-center study performed in 150 adult patients (aged between 18 and 80 years) with clinically suspected SAP and without RWMA on 2D echocardiography and with preserved systolic function from April 2019 to October 2020. Inclusion criteria were a) patients with angina or dyspnea on exertion b) patients without a history of CAD referred for elective coronary angiography after positive treadmill stress test, c) patients who reported symptoms for the first time and are judged to be in a chronic stable condition (medical history revealed that similar symptoms were already present for several months) and d) patients without RWMA on 2D echocardiography.

Exclusion criteria were a) patients with LVEF <55% on 2D echocardiography, b) patients with hypertension and left ventricular hypertrophy with interventricular septal (IVS) thickness >16 mm on 2D echocardiography, c) patients with a prior history of percutaneous coronary intervention or coronary artery bypass grafting, d) patients who presented with acute coronary syndrome confirmed by positive cardiac enzymes (serum troponin), congestive heart failure, more than trivial valvular heart disease, pathological Q-waves on resting electrocardiography or atrial fibrillation, e) patients with associated connective tissue disease or on drug therapy that affects cardiac function such as cytotoxic drugs, f) patients with poor acoustic window inappropriate for STE and low 4D image quality, g) patients with left bundle branch block or right bundle branch block, thyroid dysfunction, or chronic renal failure (glomerular filtration rate [GFR] <60 mL/min/1.73 m²).

The study was approved by the local Institutional Ethics committee and adhered to the tenets outlined in Declaration of Helsinki. Written informed consent was obtained from all the patients.

Sample size was calculated by considering specificity to be 0.93% from the previous study. The following formula was used for sample size calculation:

$$n=\frac{Z_{1-\alpha /2}^2\times S_P\times \left(1-S_P\right)}{L^2\times \left(1-Prevalence\right)}$$

where, n = required sample size, S_p = specificity = 0.93, α = size of the critical region, 1 – α = the confidence level, Z_1-α/2 = standard normal deviate corresponding to the specified size of the critical region (α), L = absolute precision desired on either side (half-width of the confidence interval) of sensitivity or specificity.

$$n=\frac{1.96\times 1.96\times 0.93\times 0.07}{0.05\times 0.05\times 0.70}$$

$$n=143$$

Total 14 (9.3%) patients who fulfilled inclusion criteria but had poor acoustic window and 4D-STE images were excluded from the analysis. Thus, we have included 150 patients for the final analysis.

2.2. Echocardiographic image acquisition and analyses

Prior to coronary angiography, baseline 2D and 4D echocardiography were performed by an experienced single operator with expertise in speckle tracking and 4D imaging using a commercially available echocardiographic system (Vivid E95; GE Healthcare, Horten, Norway) equipped with a 4 V phased array matrix transducer (1.5–4.0 MHz). In all patients, echocardiography, processing of images and 4D-STE analysis were performed either on the same day or one day prior to the coronary angiography (CAG). All images were acquired from patients in the left lateral decubitus position and connected to electrocardiography according to the echocardiography guidelines. The frame rate had to be at least 24 frames per second to work on 4D-STE. Single-beat full volume acquisition in real-time with satisfactory frame rate without stitching was acquired. LVEF, LV volumes and 4D = STE were measured automatically with 4D auto LV volume quantification. The software automatically detected the endocardial border of the LV cavity in 3D and measured the LV volumes. If the examiner found auto endocardial border detection as inaccurate, the LV endocardial border was manually adjusted in multiplanar layout (three apical and three transverse planes) with a point-click method, immediately followed by secondary automated refinement of boundary detection based on the results. Following the assessment of the LV volumes and ejection fraction, an automatic trace of the epicardial border was displayed to identify the region of interest required for LV mass and myocardial deformation measurements. This epicardial trace was manually adjusted to include the entire LV wall thickness using the same point-click method. The strain parameters were calculated as global peak systolic strain. Global longitudinal strain (GLS), global circumferential strain (GCS), global radial strain (GRS), and global area strain (GAS) of the left ventricle were used for analyses.

2.3. Coronary angiography and gensini score

CAG was performed via percutaneous femoral or radial approach. The extent and severity of CAD was assessed by the Gensini score. The Gensini score system was employed to quantitatively assess the degree of luminal narrowing and the importance of its location, which was categorized into six degrees: 1 point (<25% obstruction), 2 points (26%–50% obstruction), 4 points (51%–75% obstruction), 8 points (76%–90% obstruction), 16 points (91%–99% obstruction) and 32 points (100% complete occlusion). Then, the scores were multiplied by the factor based on the functional importance of the area given by that segment: left main coronary artery (LMCA) × 5.0; proximal segment of the left anterior descending coronary artery (LAD) × 2.5; mid segment of the left anterior descending coronary artery × 1.5; apical segment of the left anterior descending coronary artery × 1.0; first diagonal branch × 1.0; second diagonal branch × 0.5; proximal segment of the left circumflex artery (LCX) × 2.5 (if left coronary artery dominancy exist 3.5); distal segment of the left circumflex artery × 1.0 (if dominancy exist 2); obtuse marginal branch × 1.0; posterolateral branch × 0.5; proximal segment of the right coronary artery (RCA) × 1; mid segment of the right coronary artery × 1; distal segment of the right coronary artery × 1; and posterior descending artery × 1 (Fig. 1, Fig. 2). Previous research has found the Gensini score of ≥20 indicates severe atherosclerosis.^6,7 Thus, a Gensini score of ≥20 was considered as critical CAD in the present investigation.

Fig. 1.

Normal strain parameters; Upper panel Gensini score −0, Lower panel Gensini score −3, respective strain parameters (from present study).

Fig. 2.

Upper Panel-Angiogram showing critical stenosis in LAD, RCA, and depressed strain parameters. Middle panel-Triple vessel disease, Lower panel-LM- Triple vessel disease and depressed strain parameters (from present study).

2.4. Statistical analysis

Continuous variables are described as mean and standard deviation. Categorical variables are represented as number and percentages. Unpaired t-test was used to compare the mean strain parameters with Gensini score and critical stenosis and non-critical stenosis of coronary arteries (LMCA, LAD, LCX, and RCA). Spearman's correlation coefficient (ρ) was calculated to determine correlation between Gensini score with 4D-STE parameters. Receiver operating characteristics (ROC) curve analyses of strain parameters were performed for prediction of critical CAD (Gensini score ≥20). A p-value of <0.05 was considered statistically significant. The statistical analyses were performed using the SPSS statistical software, version 25.0 (Statistical Package for the Social Sciences, Inc., Chicago, Illinois, USA).

3. Results

Majority of the patients were in the sixth decade of life (55.3%) with male preponderance (68.7%). The most common co-morbidity in our patients was hypertension (58.7%) followed by diabetes mellitus (32.7%). All the patients were stratified into two groups based on Gensini score: a) non-critical stenosis (Gensini score: 0–19, n = 117) and b) critical stenosis (Gensini score ≥20, n = 33). The 4D-STE parameters significantly reduced in patients with critical stenosis than those with non-critical stenosis (Table 1).

Table 1.

Comparison of 4DSTE strain parameters between Gensini score ≥20 (critical stenosis) and Gensini score <20 (non-critical stenosis).

4DSTE strain parameters	Gensini score		p
	≥20 (n = 33)	<20 (n = 117)
Global longitudinal strain	−15.61 ± 1.84	−19.29 ± 1.61	< 0.001a
Global circumferential strain	−16.00 ± 2.02	−19.09 ± 2.12	< 0.001a
Global area strain	−26.52 ± 3.55	−33.09 ± 3.06	< 0.001a
Global radial strain	41.70 ± 7.97	53.45 ± 9.65	< 0.001a

Data are presented as mean ± S.D.

^aUnpaired t-test used.

A significant positive linear correlation was found between Gensini score with GLS (ρ = 0.626, p < 0.001), GCS (ρ = 0.548, p < 0.001)), GAS (ρ = 0.631, p < 0.001)., A significant negative linear correlation was found between GRS and Gensini score (ρ = −0.433, p < 0.001) as shown in Fig. 3.

Fig. 3.

Correlation between Gensini score and 4DSTE strain parameters.

The GLS, GAS, GCS, and GRS significantly predicted the critical CAD (Gensini score ≥20) with an area under the ROC of 0.951 (95% CI: 0.903–0.980; p < 0.001), 0.933 (95% CI: 0.880–0.967; p < 0.001), 0.870 (95% CI: 0.806–0.920; p < 0.001), and 0.822 (95% CI: 0.751–0.880; p < 0.001), respectively. The optimal cut-off value of GLS, GAS, GCS, and GRS was ≥-17 (84.9% sensitivity and 97.4% specificity), ≥-31 (90.9% sensitivity and 78.6% specificity), ≥-17 (69.7% sensitivity and 92.3% specificity), and <47 (72.7% sensitivity and 76.1% specificity), respectively (Fig. 4).

Fig. 4.

Receiver operating characteristic curve of 4DTSE strain parameters for prediction of critical CAD (Gensini score ≥20).

Of 150 patients, 16.6% of patients had single vessel disease (SVD), 7.3% had double vessel disease (DVD), 0.6% had triple vessel disease (TVD), 3.3% had left main and TVD, and 72% had non-critical or normal CAG. Thirty patients had stenosis of ≥75% in LAD, 14 had ≥75% stenosis in LCX, 19 had ≥75% stenosis in RCA, and 5 had ≥50% stenosis in LMCA The 4D-STE strain parameters significantly reduced in patients with ≥75% critical stenosis in LAD, LCX, RCA and ≥50% in LMCA, except GRS (Table 2, Table 3).

Table 2.

Comparison of 4DSTE strain parameters in vessel involved with ≥75% stenosis vs. <75% stenosis in any vessel and <50% stenosis in LMCA.

	LAD			LCX			RCA
	Critical stenosis (n = 30)b	Non-critical stenosisc (n = 108)	p	Critical stenosis (n = 14)b	Non-critical stenosisc (n = 108)	p	Critical stenosis (n = 9)b	Non-critical stenosisc (n = 108)	p
Global longitudinal strain	−15.87 ± 1.68	−19.35 ± 1.61	< 0.001a	−15.93 ± 2.13	−19.35 ± 1.61	< 0.001a	−16.21 ± 2.62	−19.35 ± 1.61	< 0.001a
Global circumferential strain	−16.13 ± 2.00	−19.19 ± 2.06	< 0.001a	−15.50 ± 1.45	−19.19 ± 2.06	< 0.001a	−16.79 ± 2.78	−19.19 ± 2.06	< 0.001a
Global area strain	−27.53 ± 3.16	−33.19 ± 3.15	< 0.001a	−26.93 ± 4.60	−33.19 ± 3.15	0.001a	−28.00 ± 4.06	−33.19 ± 3.15	< 0.001a
Global radial strain	42.27 ± 7.61	53.67 ± 9.43	< 0.001a	44.36 ± 8.88	53.67 ± 9.43	< 0.001a	45.42 ± 11.17	53.67 ± 9.43	< 0.001a

Data are presented as mean ± S.D.

^aUnpaired t-test used.

^b75% stenosis in LAD or LCX or RCA vessel.

^c<75% stenosis in any vessel and <50% stenosis in LMCA.

Table 3.

Comparison of 4DSTE strain parameters in LMCA ≥50% stenosis vs. <75% in any vessel and <50%in LMCA.

	Critical stenosis (n = 5)	Non-critical stenosis (n = 108)	p
Global longitudinal strain	−14.60 ± 1.67	−19.35 ± 1.61	<0.001a
Global circumferential strain	−15.60 ± 1.67	−19.19 ± 2.06	<0.001a
Global area strain	−25.20 ± 3.83	−33.19 ± 3.15	<0.001a
Global radial strain	39.60 ± 8.08	53.67 ± 9.43	<0.001a

Data are presented as mean ± S.D.

§ <75% stenosis in any vessel and <50% stenosis in LMCA.

¶ ≥50 %stenosis in LMCA.

^aUnpaired t-test used.

Stress test positive patient analysis revealed that seven patients had normal CAG, 23 had non-critical CAD, 13 had SVD, 1 had DVD, 7 had TVD and 5 had LM with TVD. Gensini score ≥20 was present in 27.4% of 51 patients.

4. Discussion

Presently, speckle tracking echocardiography allows the objective and noninvasive measurement of global and localized cardiac distortion.³ The noteworthy findings of this study were as follows: first, we discovered that the strains of left ventricular myocardium from 4D-STE were reduced in conjunction with the worsening of CAD (Gensini score ≥20). Second, the critical stenosis group had significantly lower GLS, GCS, GRS, and GAS than non-critical stenosis group except GRS. Third, there was a significant positive linear correlation between Gensini score and GLS, GCS, GAS whereas a significant negative correlation was found between Gensini score and GRS. Fourth, ROC analysis revealed that we could accurately assess CAD severity using strain parameters and the Gensini score with good sensitivity and acceptable specificity.

The 2D-STE is the most commonly utilized imaging modality for subclinical LV dysfunction although it has certain drawbacks. The 3D-STE has the potential to overcome some of the inherent limitations of 2D-STE in the assessment of complex LV myocardial mechanics by providing extra deformation parameters like area strain, a new deformation parameter that reflects the relative area change that incorporates both longitudinal and circumferential shortening effects.⁴ The examination of the present study revealed that all the strains of left ventricular myocardium from 4D-STE were depressed along with the aggravation of coronary artery stenosis i.e., increased Gensini score. Gensini score could be used to predict the cardiovascular prognosis during a long-term follow-up of CAD patients. A comparison of strain parameters between the present study and previous studies are depicted in Table 4.^6,7 However, all these studies used 3D-STE to assess the stenosis of coronary arteries in CAD patients. Furthermore, recent 2D-STE investigations revealed that lower values of strain parameters were associated with worsening of CAD. However, these investigations segregated the patients depending on the severity of individual coronary artery stenosis and the syntax score, rather than using the Gensini score.8, 9, 10, 11, 12, 13, 14, 15

Table 4.

Comparison of 4DSTE strain parameters with previous investigations.

	Present study		Yehia et al 20218		Dogdus et al 20187
Speckle-tracking echocardiography	4D		3D		3D
	Critical stenosis (n=5)	Non-critical stenosis (n=108)	Critical stenosis (n=23	Non-critical stenosis (n=37)	Critical stenosis (n=36)	Non-critical stenosis (n=84)
Global longitudinal strain	−14.60 ± 1.67	−19.35 ± 1.61	−9.65 ± 2.04	−14.03 ± 1.36	−8±2.37	−14 ± 3.48
Global circumferential strain	−15.60 ± 1.67	−19.19 ± 2.06	−15.13 ± 1.63	−21.78 ± 3.08	−9±3.65	−24 ± 4.77
Global area strain	−25.20 ± 3.83	−33.19 ± 3.15	−17.52 ± 3.12	−24.81 ± 1.29	−15 ± 2.14	−37.5 ± 3.94
Global radial strain	39.60 ± 8.08	53.67 ± 9.43	26.00 ± 2.30	21.92 ± 2.87	25.6 ± 6.53	36.7 ± 9.27

§ Data are presented as mean ± S.D.

§ Data are significant (p < 0.001).

What is noteworthy is that the 4D-STE strain parameters were significantly lower in patients with critical stenosis than non-critical stenosis in all coronary arteries except GRS. An increase in the number of coexisting risk factors for atherosclerosis in patients with CAD considerably worsened the degree of myocardial damage in the present study which was consistent with findings of Dogdus et al.⁶ Moreover, the current investigation demonstrated that all 4D-STE strain parameters has significantly high sensitivity and specificity for the detection of CAD severity. The GAS outperformed other strain parameters such as GCS, GRS, and GLS, and it had the highest diagnostic value for detection of critical CAD with high sensitivity (84.9%) and specificity (97.4%). This finding was similar with that of Dogdus et al, who had discovered that a GAS value of ≥ -21 had 97.2% sensitivity and 88.1% specificity for the assessment of severe CAD using 3D-STE.⁶ Similarly, Yehia et al found GAS value of ≥ -23 with higher sensitivity (95.7%) and specificity (86.5%).⁷

Few studies also reported the importance of 4D-STE in patients with suspected myocardial ischemia, heart failure or ST-segment elevation myocardial infarction. Fang et al discovered that GLS, GRS, and GCS had significant independent prognostic utility in patients with suspected heart failure and preserved ejection fraction.¹⁶ Similarly, Arbeille et al (19) revealed that the 4D-STE showed considerable improvements in left ventricular contractility in over 70% of the downstream locations from the percutaneous transluminal coronary angioplasty segment in patients with proven or suspected myocardial ischemia.¹⁷ Furthermore, Ali et al demonstrated the value of the novel 4D-STE in assessing the unfavorable clinical outcomes in patients with ST-segment elevation myocardial infarction who were managed by successful primary percutaneous coronary intervention.³ In line with these findings, 4D-STE can be used to diagnose severe coronary artery stenosis in patients with normal ejection fraction.

4.1. Study limitations

There are certain limitations of the present study. This study is not the representative for the general population as it was a single-center study with a small sample size and uneven distribution. In addition, because 4D-STE is a novel technology, standard strain range values are yet to be developed. This technology requires a good acoustic window, low frame rates, acceptable quality pictures, a steady heart rhythm, an appropriate temporal resolution for good tracking, and patient cooperation for breath holding. Lastly, 4D strain parameter measurements, normal limits, and cut-off values are currently vendor-specific and largely reliant on the 4D ultrasound equipment employed. Further studies with large sample size are warranted.

5. Conclusion

The 4D-STE can facilitate the detection of severe coronary artery stenosis with good sensitivity and specificity in the patients with SAP without RWMA on traditional echocardiography. Thus, it can significantly improve the utility of echocardiography in the early diagnosis of CAD in clinical practice.

Sources of support/funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.