Left Atrium Function Study Affection and Myocardial Deformations before and after Ventricular Septal Defect Surgical Closure

Abstract

Background:

Ventricular septal defect (VSD) is the most prevalent form of congenital heart disease (CHD) in children and the second-most prevalent form of CHD in adults, following a bicuspid aortic valve. Large VSDs can lead to significant complications, including heart failure and ventricular dysfunction. Assessing cardiac function in patients with VSD is an essential component of clinical evaluation both before and following surgery. The left atrium (LA) is a critical structure in cardiac physiology, serving as a dynamic chamber that plays a pivotal role in left ventricular filling by actively contributing during various phases of the cardiac cycle through its mechanical functions. Our aim is to validate alterations in LA function and myocardial deformation before and after surgical closure of VSD using speckle-tracking echocardiography.

Patients and Methods:

A prospective observational study with paired comparisons included 40 VSD patients who were assessed preoperatively and had short-term follow-up after 3 months at a tertiary university-based pediatric hospital.

Results:

Infants with hemodynamically significant VSD show impaired growth. Postsurgery, VSD cases demonstrated a significant elevation in weight and body surface area (P < 0.001). LA function assessment with speckle-tracking echocardiography showed a decrease in LA reservoir, LA conduit (CD), and LA contractile (CT) strain (P < 0.001, P = 0.026 and < 0.001, respectively). VSD cases showed that E/E’ and MV E/A ratios decreased postsurgery (P < 0.001 and 0.045), whereas deceleration time increased (P < 0.001), indicating improved diastolic dysfunction. VSD cases showed a systolic function decline in septal S’ (P = 0.034), and both two-dimensional (2D) and 3D echocardiography revealed significant drops in left ventricle (LV) volumes, whereas 3D echocardiography showed decreased ejection fraction postsurgery (P < 0.024 and P < 0.001, respectively). LV global longitudinal strain significantly decreased postoperatively (P < 0.001). Both left ventricular end-diastolic diameter and left ventricle end diastolic volume (LVEDV) by 2D and 3D LVEDV and LV end-systolic volume decreased significantly with surgery by 2D and 3D (P < 0.05). Right ventricle (RV) volumes by 3D conversely increased postoperatively as 3D right ventricle end diastolic volume (RVEDV) and right ventricle end systolic volume (RVESD) increased (P = 0.003 and 0.002, respectively).

Conclusion:

Preoperative VSD cases demonstrated higher LA R, CD, and CT function compared to postoperative cases.

INTRODUCTION

The primary mechanism of hemodynamic compromise in ventricular septal defect (VSD) is systemic-to-pulmonary shunting. Although many VSDs close on their own, a large one that does not close can lead to severe complications such as heart failure, pulmonary hypertension, ventricular dysfunction, and arrhythmias.[1,2]

The left atrium (LA) is a critical element of cardiac physiology that is responsible for the collection of blood into the heart and the regulation of left ventricular filling during systole and diastole.[3]

LA two-dimensional (2D) Speckle Tracking Echocardiography (STE) is a precise technique for evaluating LA function owing to its capacity to analyze myocardial deformation without angle dependency through frame-by-frame tracing of the speckles pattern produced by the interactions between ultrasound and myocardial tissue.[4]

The LA is a dynamic structure that contributes to left ventricular filling at various phases of the cardiac cycle through its mechanical function, which includes reservoir, conduit (CD), and pump functions.[5]

The atrial reservoir function is contingent upon the left ventricle (LV) end-systolic volume (LVESV) and filling pressure, functioning as a reservoir of oxygenated blood from the pulmonary veins. Consequently, any LV dysfunction will inevitably affect the atrial reservoir function.[6]

The LA functions as a CD between the LV and the pulmonary vessels during early diastole. The CD function of LA is influenced by the compliance of the atrial and ventricular chambers, which is in turn correlated with the reservoir function.[7]

Finally, the atrial booster pump function is indicative of the atrial contractile (CT) function. The degree of venous return, intrinsic LA contractility, and LV end-diastolic pressures are all factors that contribute to this phenomenon.[8]

The size of the VSD is the primary determinant of the amount of shunting that occurs across it. Hemodynamically significant shunts are typically defined as those with a pulmonary blood flow (Qp) to systemic blood flow (Qs) ratio of more than 1.5.[9]

The purpose of this study was to evaluate the alterations in LA function and other myocardial deformation changes before and after VSD surgical closure using speckle-tracking echocardiography.

PATIENTS AND METHODS

This is a prospective observational study with paired comparisons that included 40 children diagnosed with hemodynamically significant volume overload stable VSD, who underwent surgical repair. The children’s ages varied from 0.6 to 5 years, and their weights were <15 kg. The study was performed at the pediatric cardiology department of a tertiary university-based specialized pediatric hospital from 2021 to 2024. We did not include children who had congenital heart disease (CHD), cardiac arrhythmias, any acquired heart disease, or any form of organ failure.

All cases were subjected to comprehensive history-taking and full general examination involving vital signs and anthropometric measures.

Echocardiography

Echo Doppler assessments were done utilizing a GE N95 system with probes for patient age and tissue velocity imaging. Electrocardiogram data were used to time cardiac cycle events, avoiding heart rate variability. Conventional echocardiographic measurements involved M-mode, 2D, pulsed-wave, and color Doppler blood flow velocity measurements of the mitral valve (MV), involving interventricular septal thickness, dimensions like left ventricular end-diastolic diameter (LVEDD), fractional shortening, and ejection fraction (EF).[10] All candidates were within almost the same weight and body surface area (BSA) group.

The systolic myocardial velocities (S’) at the basal segments of the lateral LV wall and septal wall were evaluated using tissue Doppler imaging measures, which involved pulsed tissue velocity imaging tests. In addition, the MV E/A ratio, deceleration time, and average septal and lateral E/e’ were calculated to estimate the LV diastolic function.

3D RV and LV volume and function assessment utilizing real-time, full-volume 3D echocardiographic images to analyze RV and LV functional indices, automatically measuring RV parameters such as end-systolic volume, end-diastolic volume, and EF.

2D B-mode multi-frame (grayscale) images were acquired in the apical four-, two, and three-chamber views for 2D speckle-tracking echocardiography. Cine-loops were implemented to archive data for offline analysis. Customized software (Echo PAC V113; GE Healthcare, Chicago, Illinois) was employed to analyze the images offline. The software automatically outlined the LV epicardial boundary; the LV endocardial boundary was manually delineated. When necessary, the breadth of the region of interest was manually adjusted. Subsequently, the software generates global and segmental longitudinal strain by dividing the LV myocardium into six segments. The longitudinal strain is lower than the baseline due to the fact that the myocardium is attenuated in the longitudinal direction during systole. For each of the myocardial segments, the maximal systolic longitudinal strain was detected from these curves. The global longitudinal strain (GLS) is calculated by recording and averaging the strain values of all segments.

The LA strain involved 2D pictures obtained from the perspectives of the apical four and two chambers. For the offline analysis, data were stored in cine-loops. A customized program, Echo PAC V204 from GE Healthcare in Chicago, Illinois, was used to analyze the pictures offline. The LA endocardial border was manually drawn using a point-and-click method. The technology generated epicardial surface tracing automatically, which generated a region of interest that was manually adjusted to encompass the entire thickness of the myocardium in the systolic frame. The program rejected segments having less-than-ideal images and weren’t included in the examination. Each atrial wall was evaluated from apical four- and two-chamber perspectives to estimate the LA segmental function. Biplane values for phasic LA strain (CD, LA R, and CT) were then recorded.

Patients were assessed preoperatively and at 3 months after surgery.

Statistical analysis

IBM© SPSS© Statistics version 27 (IBM© Corp., Armonk, NY, USA) was employed to conduct the statistical analysis. The independent-samples Student’s t-test is employed to compare intergroup differences, and continuous numerical variables are presented as mean and standard deviation. The paired-samples Student’s t-test is employed to compare continuous data that are paired. Pearson’s correlation is employed to evaluate correlations between continuous variables. Statistical significance is determined by P < 0.05.

RESULTS

The study involved 40 children diagnosed with VSD. They had a mean preoperative age of 13.9 months and 19.9 months after surgery. It included 24 males and 16 females.

A significant increase in weight and BSA (P < 0.001) was observed postoperatively [Table 1].

Table 1.

Anthropometric variables in ventricular septal defect cases before and after surgery

Variable	n	Preoperative			Postoperative			Difference			P †
		Mean	SD	Z score	Mean	SD	Z score	Mean	Lower 95% CI	Upper 95% CI
Weight (kg)	40	7.2	2.6	−2.8	9.5	2.5	−1.6	−2.3	−2.5	−2.0	<0.001
BSA (m2)	40	0.37	0.1	−4	0.4	0.1	−3.4	−0.1	−0.1	−0.1	<0.001

^†Paired-samples t-test. SD=Standard deviation, 95% CI=95% Confidence interval, BSA=Body surface area

VSD cases showed a significant decrease of septal S’ postoperatively (mean 5.85.1 mm/s compared to 5.8 mm/s preoperatively (P = 0.034). Furthermore, E/E’ and MV E/A decreased after surgery, with a mean around 9.3 and 1.4 compared to preoperative means of 10.9 and 1.6, respectively (P < 0.001 and P = 0.045, respectively); however, deceleration time increased after surgery with a mean around 132.6 ms compared to 102.7 ms preoperatively (P < 0.001) [Table 2].

Table 2.

Evaluation of myocardial function in ventricular septal defect cases before and after surgery

Variable	n	Preoperative			Postoperative			Difference			P †
		Mean	SD	Z score	Mean	SD	Z score	Mean	Lower 95% CI	Upper 95% CI
Septal S’ (mm/s)	40	5.8	1.5	−0.72	5.1	1.5	−2.15	0.7	0.1	1.3	0.034
Septal E’ (mm/s)	40	10.1	3.0	−0.15	9.1	1.6	−1.27	1.0	−0.1	2.0	0.062
Septal A’ (mm/s)	40	6.7	2.3	0.5	5.1	1.6	−0.48	1.6	0.7	2.5	0.101
Lateral S’ (mm/s)	40	6.2	1.4	−0.39	5.9	1.2	−1.09	0.3	−0.3	0.9	0.276
Lateral E’ (mm/s)	40	11.5	2.9	−0.67	10.7	2.3	−1.47	0.8	0.0	1.7	0.059
Lateral A’ (mm/s)	40	7.4	3.6	0.94	6.3	2.3	−3.12	1.1	−0.4	2.6	0.151
E/E’	40	10.9	2.4	1.74	9.3	1.6	1.57	1.7	0.9	2.4	<0.001
DT (ms)	40	102.7	33.4	−2.36	132.6	37.8	−0.87	−30.0	−42.2	−17.7	<0.001
MV E (m/s)	40	1.1	0.2	1.2	0.9	0.1	0.05	0.2	0.1	0.3	<0.001
MV A (m/s)	40	0.9	0.2	3.27	0.7	0.2	1.79	0.2	0.1	0.3	<0.001
MV E/A	40	1.6	0.3	−0.27	1.4	0.3	−0.76	0.2	−0.2	0.2	0.045

^†Paired-samples t-test. SD=Standard deviation, 95% CI=95% Confidence interval, DT=Deceleration time

Both LVEDD and LVEDV by 2D, and 3D LVEDV and LVESV decreased significantly with surgery, from a mean of 3.1 and 37.4 to 2.9 and 34.1, respectively by 2D (P < 0.001 and P = 0.003, respectively), and from 19.8 and 9.3 to 18.8 and 8.2 ml, respectively by 3D indicating improvement of LV dilatation and volume overload caused by the VSD [Table 3].

Table 3.

Two-dimensional three dimensional D echocardiographic measurements in ventricular septal defect cases before and after surgery

Variable	n	Preoperative			Postoperative			Difference			P †
		Mean	SD	Z score	Mean	SD	Z score	Mean	Lower 95% CI	Upper 95% CI
IVS (cm)	40	0.5	0.1	−0.5	0.5	0.1	−0.5	0.0	0.0	0.0	0.844
LVESD (cm)	40	1.8	0.4	1.1	1.8	0.3	0.6	0.0	0.0	0.1	0.298
LV EDD (cm)	40	3.1	0.5	2.5	2.9	0.4	0.96	0.2	0.1	0.2	<0.001
LV EDV (ml)	40	37.4	12.8	0.56	34.1	10.5	−1.36	3.3	1.2	5.4	0.003
LV ESV (ml)	40	11.0	4.8	0.53	10.2	3.3	−0.96	0.9	−0.2	1.9	0.096
LV PWD (cm)	40	0.5	0.1	−0.5	0.5	0.1	−0.4	0.0	0.0	0.0	0.401
EF (%)	40	72.4	7.7	1.48	70.1	5.9	1	2.3	−0.4	5.0	0.090
FS (%)	40	40.4	6.4	1.08	37.9	5.0	0.58	2.5	0.3	4.6	0.024
3D RV EDV (ml)	40	18.6	6.6	−1.28	20.7	6.1	−0.86	−2.1	−3.5	−0.8	0.003
3D RV ESV (ml)	40	10.7	3.9	−0.33	12.2	3.7	0.05	−1.5	−2.4	−0.6	0.002
3D RV EF (%)	40	42.0	7.3	−1.14	41.0	5.4	−1.29	1.0	−1.3	3.3	0.389
3D LV EDV (ml)	40	19.8	6.4	0.76	18.8	5.3	0.09	1.0	0.1	1.9	0.037
3D LV ESV (ml)	40	9.3	2.4	1.08	8.2	3.1	0.26	1.2	2.0	0.3	0.013
3D LV EF (%)	40	60.4	7.0	0.35	49.1	7.5	−2.75	11.3	8.2	14.3	<0.001

^†Paired-samples t-test. SD=Standard deviation, 95% CI=95% Confidence interval, IVS=Interventricular septum, LVESD=Left ventricular end-systolic diameter, LV EDD=Left ventricular end-diastolic dimension, LV EDV=Left ventricular end-diastolic volume, LV ESV=Left ventricular end-systolic volume, LV PWD=Left ventricular posterior wall thickness, EF=Ejection fraction, FS=Fractional shortening, RV EDV=Right ventricular EDV, RV ESV=Right ventricular ESV, RV EF=Right ventricular EF, LV EF=Left ventricular EF

The observed increase in RV volumes postoperatively may be partly attributed to reduced preload modulation following decreased diuretic therapy [Figure 1].

Figure 1.

Three-dimensional RV dimensions and function in postoperative cases

On the other hand, LV function decreased after surgery by 2D but within normal limits, where FS decreased from 40.4% to 37.9% (P = 0.024), yet 3D echocardiography showed a significant reduction in EF from 60.4% to 49.1% (P < 0.001), denoting that 3D results were more accurate [Table 3 and Figure 2].

Figure 2.

STE for LA in ventricular septal defect cases shows more reservoir function and more negative conduit and contractile function values

By STE, LV GLS significantly decreased postoperatively with a mean of −16.5% compared to −19.9% preoperatively (P < 0.001), indicating a decline in LV systolic function after surgery, which mostly reflects transient physiological remodeling after volume overload elimination [Table 4].

Table 4.

Speckle tracking echocardiography indices in ventricular septal defect cases before and after surgery

Variable	n	Preoperative		Postoperative		Mean	95% CI		P †
		Mean	SD	Mean	SD		Lower	Upper
LV GLS (%)	40	−19.9	2.7	−16.5	2.0	−3.3	−4.4	−2.3	<0.001
LA-S R (%)	40	31.2	6.3	25.9	4.4	5.3	3.7	6.9	<0.001
LA-S CD (%)	40	−19.2	5.0	−17.7	3.7	−1.4	−2.7	−0.2	0.026
LA-S CT (%)	40	−12.1	3.3	−8.1	1.8	−4.0	−5.0	−2.9	<0.001

^†Paired-samples t-test. SD=Standard deviation, 95% CI=95% Confidence interval, LV GLS=Left ventricular global longitudinal strain, LA-S R=Left atrial strain – reservoir, LA- S CD=Left atrial strain – conduit, LA-S CT=Left atrial strain - contraction

LA strain values with speckle-tracking echocardiography showed that LA R decreased postoperatively with a mean of 25.9% compared to 31.2% preoperatively (P < 0.001), as LA distention is higher in VSD physiology to adapt to the volume overload.

LA CD and LA CT strains show less negative values in the postoperative state because of the elimination of volume overload, with mean values of −19.2% and −12.1%, respectively, compared to −17.7% and −8.1% preoperatively, respectively, P = 0.026 and P < 0.001, respectively) [Table 4 and Figures 2, 3].

Figure 3.

STE for LA in ventricular septal defect cases after surgery shows less reservoir function and less negative conduit and contractile function values

DISCUSSION

This is a prospective observational study of cases diagnosed with VSD who were candidates for open heart surgery and VSD closure; a follow-up study was done for the same patients at least 3 months after the operation.

The current study showed statistically significant increase in anthropometric measurements with an increase in weight and BSA (P < 0.001), denoting improvement of the physical developmental and nutritional condition of the VSD cases after the VSD surgical closure with the advancement of age.

This is consistent with the findings of recent studies conducted in 2022, which demonstrated that children with CHD have a high prevalence of preoperative malnutrition and a variety of delays in weight gain and growth and may exhibit catch-up growth postoperatively.[11,12]

We also found that the LA R strain was significantly higher in the preoperative state, consistent with increased atrial adaptation to volume overload.

While LA CD and LA CT strains show more negative values in preoperative assessment, with means of −19.2% and −12.1%, respectively, whereas the postoperative assessment shows less negative values, with means of −17.7% and −8.1%, respectively, with significant P = 0.026 and < 0.001.

This comes in agreement with Montaser et al. who assessed LA function in VSD patients by 2D STE and compared it with normal healthy controls.[13] He discovered that the mechanical function of the left atrium is impacted, particularly the reservoir and contractility, in hemodynamically significant VSD that results in LA volume excess. He also discovered that VSD cases exhibited a statistically significant increase in LA volume (reservoir) and a statistically highly significant decrease in LA global CT strain compared to controls.

Regarding diastolic function, we found that VSD cases have higher E/E’ and MV E/A before surgery with means around 10.9 and 1.6, respectively, whereas after surgery have lower E/E’ and MV E/A with means 9.3 and 1.4 with significant P < 0.001 and 0.045 respectively, however, deceleration time was lower in preoperative cases with mean around 102.7 ms but higher in postoperative cases with mean 132.6 ms with significant P < 0.001. E/E’, MV E/A, and deceleration time (DT) change indicate that patients with VSD show improvement in diastolic dysfunction after surgery

Fukushima et al. also studied aortic root dilatation (ARD), which was reported in postoperative patients with VSDs, and its effect on diastolic dysfunction. They discovered that the existence of ARD among individuals with postoperative VSD was significantly correlated with LV diastolic dysfunction.[14]

In patients with congenital cardiac disease, diastolic dysfunction or abnormal loading conditions are the primary causes of congestion, as discovered by Masutani and Senzaki.[15]

Regarding systolic function, VSD cases showed a decrease of septal S’ postoperatively compared to preoperatively from 5.8 to 5.1 (P = 0.034). Furthermore, 2D and 3D echocardiography showed a statistically significant drop after surgery as 2D FS decreased after surgery within normal limits, from 40.4% to 37.9% (P = 0.024). In comparison, 3D echocardiography showed a significant decline in EF from 60.4% to 49.1% (P < 0.001), denoting that 3D results were more accurate. This comes in agreement with Dorosz et al. and Turton and Ender who found that 3D echocardiography can assess the LV and RV function and volumes better than 2D conventional echocardiography in cardiac surgery.[16,17]

LV GLS significantly decreased postoperatively from −19.9 to −16.5 (P < 0.001). This reduction in systolic function can be attributed to the short-term follow-up period (3 months) used in our study. The decline likely reflects a transient effect of physiological remodeling following the elimination of volume overload, rather than a true impairment of myocardial function.

Cantinotti et al. documented postoperative LV systolic impairment by STE following pediatric cardiac surgery. They conducted a global and segmental strain analysis using STE and discovered that the LV systolic function was substantially impaired after pediatric cardiac surgery. This was evidenced by the substantial reduction in global and segmental LV strain parameters in comparison to preoperative and healthy subjects.[18]

This is in agreement with Adamson et al., who assessed LV function by conventional echocardiography in VSD cases before and after surgery. They realized that one-third of infants have left ventricular dysfunction after VSD surgery, and this LV dysfunction uniformly resolves within 9 months.[19]

Matsuhisa et al. also employed 2D STE and discovered that patients with membranous VSDs experienced a decrease in postoperative global ventricular and ventricular septal functions.[20]

Regarding LV and RV volumes, by 2D echocardiography, LVEDD and LVESV decreased significantly with surgery, from a mean of 3.1 to 2.9 cm, and from 11 to 10.2 ml, respectively (P < 0.001 and P = 0.003, respectively), indicating improvement of LV dilatation and volume overload caused by the VSD. RV volumes conversely increased with surgery, as mean RVEDV and RVESD increased from 18.6 and 10.7 to 20.7 and 12.2 ml postoperatively, respectively (P = 0.003 and P = 0.002, respectively. These results indicate that RV volumes increase with the decrease or discontinuation of diuretic therapy.

Cho et al. assessed the LV dimensions in VSD patients in preoperative and postoperative states. They found significant decreases in LVEDV and LVEDD at one, three, and 12-month assessments postoperatively.[21] Conversely, Amoozgar observed that a substantial number of patients continued to experience left ventricular hypertrophy and dilation in the follow-up period following percutaneous peri-membranous VSD closure.[22]

Limitations

Our study’s validity is restricted by the relatively limited number of patients. Therefore, these findings have to be confirmed in larger prospective studies.

CONCLUSION

By LA speckle-tracking echocardiography, we found that VSD cases have more reservoir, CD, and CT LA strain function than postoperative cases. After surgery, reservoir, CD, and CT function had lower values than VSD cases in short-term follow-up, mostly due to the elimination of volume overload effect after surgery.

Surgical closure of hemodynamically significant VSD cases showed improved remodeling in LV dimensions, an increase in RV dimensions, and a decrease in LV systolic function by 2D and 3D echocardiography. However, 3D was more accurate than 2D in measurements and assessment. LV GLS showed a significant drop in STE after surgery in our short-term follow-up study, mostly related to transient postoperative remodeling. Furthermore, we found a significant improvement in anthropometric measurements after surgery.

Ethical statement

The study protocol was approved by the Ethical Scientific Committee of the Faculty of Medicine, Cairo University (code: MD-353-2021) and per the Declaration of Helsinki for medical research involving human subjects. Informed consent was obtained from patients’ legal guardians before enrolling patients in the study.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.