Assessment of cardiac function in children with tuberculosis

ABSTRACT

Introduction:

Tuberculosis (TB) continues to be a major global health concern, emerging evidence suggests that both pulmonary and extrapulmonary TB can cause subclinical or overt cardiac dysfunction, particularly in children. Echocardiography is a sensitive tool for early detection of myocardial, pericardial, and functional abnormalities. Early recognition of cardiac involvement and its reversibility with anti-tubercular therapy (ATT) may significantly influence clinical outcomes in paediatric population.

Methods:

This observational study included 50 children diagnosed with pulmonary or extrapulmonary TB. Detailed clinical evaluation, laboratory investigations (including Troponin I and Pro-BNP), and baseline echocardiographic assessment were performed using M-mode, pulsed and tissue Doppler and 2-D speckle tracking imaging. Repeat echocardiography was done after completion of ATT to assess improvement in cardiac parameters.

Results:

Of the 50 enrolled children, 16 (32%) demonstrated cardiac dysfunction on initial echocardiography. The most common abnormalities included impaired diastolic parameters on Doppler imaging and reduced global longitudinal strain (GLS). Elevated cardiac biomarkers was also associated with diastolic dysfunction. Post-ATT follow-up was available for 47 children. Among the 14 who initially had cardiac dysfunction, 12 (85.7%) showed significant improvement after treatment, while 2 had persistent dysfunction. No new dysfunction developed in children with normal baseline echocardiography.

Conclusions:

A considerable proportion of children with tuberculosis exhibit subclinical cardiac dysfunction detectable on echocardiography. Most abnormalities showed improvement following completion of ATT, underscoring the reversible nature of of TB-related cardiac involvement. Routine echocardiographic screening, should be considered in Pediatric TB to facilitate early diagnosis, timely intervention, and monitoring of treatment response.

INTRODUCTION

Tuberculosis (TB) remains a global health problem, with its discovery dating back centuries. It has been postulated that the causal genus, Mycobacterium, emerged over 150 million years ago.[1] According to the World Health Organization Global Tuberculosis Report 2024,[2] an estimated 10.8 million people developed TB in 2023, with 1.25 million deaths, reaffirming TB as the leading cause of death from a single infectious agent globally.

Cardiovascular involvement in TB, although historically considered rare, is now increasingly recognized, especially in immunocompromised populations and in endemic regions. TB can affect various cardiac structures, most notably the pericardium, but also the myocardium, endocardium, coronary arteries, and aorta.[3,4] The frequency of pericarditis in cardiovascular TB is around 1%–2%,[5] myocarditis is 0.14%–2%,[6,7] and aortitis is 0.3%.[8,9] Tuberculous pericarditis accounts for about 85% in HIV-infected individuals in high-burden settings.[10] This condition can progress to constrictive pericarditis or lead to cardiac tamponade if not promptly diagnosed and treated. In contrast, myocarditis and aortitis are less common but are often associated with high morbidity and mortality due to their subtle presentations and rapid progression to heart failure or arrhythmia.[11]

Recent literature has begun to shed light on cardiovascular TB in children, a population in whom such complications are historically underdiagnosed. Although data are limited, emerging studies have reported pediatric cases of pericardial effusion, myopericarditis, and pulmonary hypertension associated with TB. The nonspecific clinical presentation in children, ranging from failure to thrive, dyspnea, or fever to subtle signs of right heart dysfunction, often results in diagnostic delays.

In cases of TB myocarditis, echocardiographic parameters such as reduced ejection fraction or global longitudinal strain (GLS) can provide early clues, even before overt symptoms appear. Moreover, echocardiography aids in monitoring disease progression and response to antitubercular therapy (ATT), particularly in resource-limited settings where advanced imaging modalities like cardiac MRI may not be readily available.

This study aims to bridge the knowledge gap regarding subclinical and overt cardiac involvement in pediatric TB by employing echocardiographic assessments. Through systematic evaluation before and after ATT, this research intends to quantify changes in cardiac function, thereby supporting the role of echocardiography as a screening and follow-up tool in TB management.

METHODS

Study design and setting

This observational cohort study was conducted in the Department of Paediatrics at a tertiary care hospital from August 2023 to August 2025. Children newly diagnosed with TB (pulmonary or extrapulmonary) were enrolled after obtaining written informed consent from parents or guardians.

Study population

Children <18 years with clinically, microbiologically, or radiologically confirmed TB were included. Exclusion criteria were congenital heart disease, pre-existing cardiac disorders or refusal of consent.

Data collection

A detailed history and clinical examination were performed for all participants. Anthropometric measurements, vital signs, and systemic examination findings were recorded. Relevant blood investigations were done.

Echocardiographic assessment

Echocardiography was performed using a GE Vivid E95 3D/4D machine at baseline and after completion of ATT. M-mode echocardiography was used to assess left ventricular (LV) chamber dimensions and systolic function by measuring LV internal dimension at end-diastole, LV internal dimension at end-systole, interventricular septal and posterior wall thickness (interventricular septal thickness in diastole/interventricular septum during systole [IVSs], LV posterior wall in diastole [LVPWd]/LV posterior wall thickness in systole), and LV ejection fraction (LVEF).

Diastolic function was evaluated with pulsed-wave Doppler by recording early (E) and late (A) diastolic inflow velocities across the mitral and tricuspid valves and calculating the E/A ratio. Tissue Doppler imaging was performed at the septal, lateral LV annulus, and RV annulus to obtain early diastolic (E′), late diastolic (A′), and systolic (S′) myocardial velocities. The E/E′ ratio was calculated to estimate LV filling pressures and detect diastolic dysfunction.

Two-dimensional speckle tracking echocardiography was performed for all children to assess myocardial deformation. GLS was calculated using a 17-segment or 18-segment LV model, serving as a sensitive marker for detecting subclinical systolic dysfunction even when conventional ejection fraction remained within normal limits.

Follow-up evaluation

Echocardiography was repeated after completion of the standard course of ATT to evaluate improvement or persistence of cardiac dysfunction.

Statistical analysis

All statistical calculations were done using Statistical Package for the Social Sciences (SPSS) - IBM SPSS Statistics (Version 31). Continuous variables were expressed as means, standard deviation and frequency. All comparisons were done using paired samples t-test. To define associations between nonparametric categorical variables, Chi-square or Fisher’s exact test was used. A P < 0.05 was considered statistically significant.

RESULTS

The baseline characteristics of the study population (n = 50) revealed a mean age of 9.38 ± 3.42 years, with an average weight of 27.80 ± 9.10 kg and a mean height of 131.96 ± 18.65 cm [Table 1]. The calculated mean BMI was 15.46 ± 1.58 kg/m², indicating that the majority of children were undernourished. There was an equal number of males and females in the study population. Regarding cardiac biomarkers, the mean Troponin I level was 45.10 ± 75.63 ng/L, and the mean Pro BNP was 219.74 ± 93.39 pg/mL, indicating that some of the patients also had elevated cardiac biomarkers. In terms of diagnosis, 26 children (52%) were diagnosed with pulmonary TB, while 24 (48%) had extra-pulmonary TB. The distribution was as follows: tubercular meningitis was seen in 11 patients, tubercular lymphadenitis in 8 patients, abdominal TB in 2 patients, disseminated TB in 2 patients, and tubercular vasculitis in 1 patient. ECG was performed only at baseline before initiation of ATT. The common findings included sinus tachycardia, nonspecific ST-T changes, and low-voltage QRS complexes.

Table 1.

Sociodemographic characteristics of study population

Parameter	Value
Age (years)	9.38±3.42
Weight (kg)	27.80±9.10
Height (cm)	131.96±18.65
BMI (kg/m2)	15.46±1.58
Male, n (%)	25 (50.0)
Troponin I (ng/L)	45.10±75.63
Pro BNP (pg/mL)	219.74±93.39
Pulmonary TB, n (%)	26 (52.0)
Extra pulmonary TB, n (%)	24 (48.0)

BMI: Body mass index, TB: Tuberculosis

Hemoglobin status

The mean hemoglobin level of the study population was 10.20 ± 1.70 g/dL, reflecting a high burden of anemia. According to WHO age-specific criteria, 35 (70%) children were anemic, including 5 (10%) with mild, 24 (48%) with moderate, and 6 (12%) with severe anemia.

Steroid use

Corticosteroids were administered in children with tuberculous meningitis, disseminated TB, and tubercular vasculitis in accordance with standard treatment guidelines. However, steroids were not prescribed specifically for the management of cardiac manifestations observed in the study.

Following ATT, there was a statistically significant reduction in LV diastolic dimension, and a significant increase was observed in the IVSs [Table 2]. The LVEF, though in a normal range, showed a marked improvement (P < 0.001), highlighting a significant enhancement in global systolic function following ATT.

Table 2.

Comparison of M mode parameters before and after antitubercular therapy

M mode	Before ATT, Mean±SD	After ATT, Mean±SD	P
LVIDd	3.63±0.44	3.30±0.46	<0.001*
LVIDs	2.38±0.44	2.38±0.41	0.452
IVSs	0.55±0.10	0.63±0.12	<0.001*
IVSd	0.50±0.09	0.54±0.08	0.019
LVPWd	0.45±0.08	0.49±0.07	<0.001*
LVPWs	0.42±0.23	0.43±0.24	0.715
LVEF (%)	61.55±6.27	66.34±3.74	<0.001*

*Statistically significant. Parameters measured in cm. ATT: Anti-tubercular therapy, IVSs: Interventricular septum during systole, LVEF: Left ventricular ejection fraction, LVPWd: Left ventricular posterior wall in diastole, IVSd: Interventricular septal thickness in diastole, LVPWs: Left ventricular posterior wall thickness in systole, LVIDd: Left ventricular internal dimension at end-diastole, LVIDs: Left ventricular internal dimension at end-systole, SD: Standard deviation

As shown in Table 3, only the tricuspid early diastolic velocity (T-E) showed a statistically significant improvement, indicating enhanced myocardial relaxation and improved early diastolic function following ATT. In contrast, the mitral E wave (M-E), A wave (M-A), their ratio (M E/A), Tricuspid A velocity (T-A), and tricuspid E/A ratio (T E/A) all showed minor changes post-therapy, but these were not statistically significant.

Table 3.

Comparison of pulsed wave Doppler parameters before and after antitubercular therapy

Pulse Doppler	Before ATT, Mean±SD	After ATT, Mean±SD	P
M-E	0.84±0.11	0.87±0.12	0.082
M-A	0.74±0.10	0.76±0.11	0.113
M E/A	1.22±0.27	1.20±0.18	0.493
T-E	0.69±0.16	0.78±0.10	<0.001*
T-A	0.71±0.15	0.74±0.14	0.059
T E/A	1.08±0.19	1.01±0.29	0.069

*Statistically significant. Parameters measured in m/s. SD: Standard deviation, ATT: Anti-tubercular therapy

Table 4 shows a statistically significant increase in LV early (LVE′) diastolic velocity, the right ventricular early diastolic (RV-E′) and systolic (RV-S′) velocities also increased significantly, suggesting improved right heart compliance and contractility. Although interventricular septal velocities and LV systolic velocity (LV-S′) had not shown any significant changes, the overall findings reflect notable recovery of both left and right ventricular (RV) function post-ATT.

Table 4.

Comparison of tissue Doppler parameters before and after antitubercular therapy

Tissue Doppler	Before ATT, Mean±SD	After ATT, Mean±SD	P
IVS-E’	0.09±0.03	0.11±0.03	0.090
IVS-S’	0.09±0.03	0.10±0.02	0.351
LV-E’	0.12±0.04	0.14±0.04	<0.001*
LV-A’	0.14±0.03	0.15±0.03	0.014
LV-S’	0.09±0.02	0.09±0.02	0.081
RV-E’	0.12±0.04	0.13±0.04	<0.001*
RV-A’	0.18±0.21	0.18±0.20	<0.001*
RV-S’	0.11±0.02	0.12±0.02	<0.001*

*Statistically significant. Parameters measured in m/s. SD: Standard deviation, IVS-E: Interventricular septum early diastolic, IVS-S: Interventricular septum Systolic, LV: Left ventricle, RV: Right ventricular, ATT: Anti-tubercular therapy

The GLS values showed a mild improvement after ATT from −16.62 ± 1.99 to −17.35 ± 7.76; however, the change was not statistically significant (P = 0.502) [Table 5].

Table 5.

Comparison of 2D Speclke tracking parameters before and after antitubercular therapy

2D speckle tracking	Before ATT, Mean±SD	After ATT, Mean±SD	P
GLS (%)	−16.62±1.99	−17.35±7.76	0.502

SD: Standard deviation, ATT: Anti-tubercular therapy, GLS: Global longitudinal strain

Assessment of cardiac function in study population

Diastolic dysfunction criteria

On Pulsed wave Doppler, Diastolic dysfunction was categorized as per the following:[12]

• Grade 1 diastolic dysfunction E/A <0.8 and E/E’ <8–10

• Grade 2 diastolic dysfunction E/A 0.8–2 and E/E’ 10–14

• Grade 3 diastolic dysfunction E/A >2 and E/E’ >14.

Out of the 50 children enrolled in the study, 16 patients were found to have diastolic dysfunction before initiation of ATT, with 6 having Grade 1 and 10 having Grade 2 dysfunction. Following ATT, diastolic function normalized in the majority of patients, with only 2 children continuing to have Grade 2 dysfunction [Table 6]. Importantly, no cases of Grade 3 dysfunction were observed either before or after therapy. Three patients died during the course of the study and so post-ATT cardiac assessment could not be done. Among them, 2 had diastolic dysfunction while 1 had normal diastolic function.

Table 6.

Assessment of diastolic function in study population

	Number of patients before ATT	Number of patients after ATT	Number of deceased
Grade 1 diastolic dysfuntion	6	0	1
Grade 2 diastolic dysfunction	10	2	1
Grade 3 diastolic dysfunction	0	0	0
Normal	34	45	1
Total	50	47	3

ATT: Anti-tubercular therapy

Systolic dysfunction

Systolic dysfunction was categorized as per following:

• LVEF < 55%[13]

• GLS >-16.[14]

Out of the total 50 patients, 3 patients (6%) exhibited systolic dysfunction before ATT, and the rest had normal systolic function. 3 patients (2 with systolic dysfunction and 1 with normal function) died before ATT completion. 1 patient with systolic dysfunction before ATT improved after the therapy [Table 7].

Table 7.

Assessment of systolic function in study population

	Number of patients before ATT	Number of patients after ATT	Number of deceased
Systolic dysfunction	3	0	2
Normal systolic function	47	47	1
Total	50	47	3

75% of children with elevated Troponin I had diastolic dysfunction, while only 23.8% of those with normal Troponin I had dysfunction [Table 8a]. The Pearson Chi-square test yielded a value of 8.093, and Fisher’s Exact Test showed a P = 0.009. These results suggest that elevated Troponin I is strongly associated with the presence of diastolic dysfunction in this study.

Table 8.

Assessment of diastolic dysfunction with cardiac enzymes

a: Association between diastolic dysfunction and troponin i levels
Troponin I level	Diastolic dysfunction present, n (%)	Diastolic dysfunction absent, n (%)	P (Fischer’s exact test)
Elevated	6 (75)	2 (25)	0.009
Normal	10 (23.8)	32 (76.2)
b: Association between diastolic dysfunction and pro BNP levels
Pro BNP level	Diastolic dysfunction present, n (%)	Diastolic dysfunction absent, n (%)	P (Fischer’s exact test)
Elevated	3 (30)	7 (70)	1
Normal	13 (32.5)	27 (67.5)

Among children with elevated Pro BNP levels, 30% had diastolic dysfunction, while 70% did not. 32.5% of those with normal Pro BNP levels had diastolic dysfunction, and 67.5% had no dysfunction [Table 8b].

The Pearson Chi-square test showed a value of 0.023, and Fisher’s Exact Test showed a P = 1.000, indicating no statistically significant association between elevated Pro BNP levels and diastolic dysfunction.

In this study, 25% of children with elevated Troponin I levels had systolic dysfunction, while 75% did not. In contrast, among children with normal Troponin I levels, only 2.4% had systolic dysfunction, and the remaining 97.6% had no dysfunction [Table 9a].

Table 9.

Assessment of systolic dysfunction with cardiac enzymes

a: Association between systolic dysfunction and troponin I levels
Troponin I levels	Systolic dysfunction present, n (%)	Systolic dysfunction absent, n (%)	P (Fischer’s exact test)
Elevated	2 (25)	6 (75)	0.063
Normal	1 (2.4)	41 (97.6)
b: Association between systolic dysfunction and pro BNP levels
Pro BNP level	Systolic dysfunction present, n (%)	Systolic dysfunction absent, n (%)	P (Fischer’s exact test)
Elevated	2 (20)	8 (80)	0.098
Normal	1 (2.5)	39 (97.5)

The Pearson Chi-square value was 6.096, with a P = 0.014, indicating a statistically significant association between elevated Troponin I levels and the presence of systolic dysfunction. However, it was observed that 2 out of 4 cells (50%) had expected counts < 5, with the minimum expected count being 0.48, thus violating the assumptions of the Pearson Chi-square test. As a result, Fisher’s Exact Test was also considered, which yielded a borderline nonsignificant P = 0.063.

Of the three deaths observed during the study, two were attributable to severe cardiac dysfunction, while the third occurred in a child with Tubercular Aortitis complicated by uncontrolled hypertension and massive pulmonary hemorrhage. 20% of children with elevated Pro BNP had systolic dysfunction, compared to 2.5% among those with normal Pro BNP. The Pearson Chi-square test yielded a value of 4.344 with a P = 0.037, indicating a statistically significant association between elevated Pro BNP and systolic dysfunction; Fisher’s Exact Test showed a P = 0.098, indicating a non-significant association [Table 9b].

DISCUSSION

This study was conducted over a period of 2 years in 50 children with TB (Pulmonary and extrapulmonary). Cardiac function was assessed using three Echocardiography modalities, i.e. M-mode, 2D Doppler, and 2D speckle tracking before and after ATT. The objective of the study was to assess cardiac function in children with TB and also observe the changes in cardiac parameters after completion of ATT.

M mode

Among the M-mode echocardiographic parameters assessed before and after ATT, a significant increase was observed in the interventricular septum during systole (IVSs) suggesting enhanced myocardial contractility or reduced septal wall stress. Additionally, the LVPWd increased indicating possible myocardial remodeling or resolution of myocardial involvement. LVEF showed a marked improvement, highlighting a significant enhancement in global systolic function following ATT.

A case series showed 3 confirmed cases of tuberculous myocarditis in young adults, all of whom exhibited acute LV systolic dysfunction. Over the course of treatment, follow-up ECHOs in all patients demonstrated normalized cardiac function with EFs >55%, indicating full recovery of myocardial performance.[15]

Pulsed wave Doppler

Among the Pulsed wave Doppler parameters, the Tricuspid diastolic velocity (T-E) was the only parameter that demonstrated a statistically significant improvement following ATT.

In contrast, mitral inflow velocities (M-E and M-A) and their derived ratio (M E/A), while showing numerical improvements, did not reach statistical significance. The E/A ratio, which is often used to grade diastolic dysfunction, slightly declined, a change that was not significant and could reflect improved balance between early and late filling phases rather than worsening. A similar study in 2025[16] reported that an elevated E/A ratio (>1.5) with a short deceleration time (<160 ms), alongside reduced IVC respiratory variability, indicative of constrictive physiology.

Tissue Doppler

Among the LV parameters of Tissue Doppler Imaging, a statistically significant improvement was observed in early diastolic velocity (LV-E′) reflecting enhanced myocardial relaxation and improved early diastolic function of the left ventricle post-therapy. In contrast, systolic velocity (LV-S′) did not change significantly, indicating that LV systolic function remained relatively preserved or showed minimal detectable improvement post-ATT.

RV function, however, exhibited notable and consistent improvements across all three measured TDI parameters. RV-E′ increased significantly, denoting improved RV relaxation. RV-A′, though unchanged in mean value, showed a significant reduction in variability. Importantly, RV-S′ also improved significantly, indicating enhanced systolic performance of the right ventricle after ATT. These improvements suggest that the right heart, potentially affected due to TB-related pulmonary pathology or systemic inflammatory effects, benefits significantly from effective anti-tubercular treatment.

In a study RV involvement was evaluated through parameters such as tricuspid annular plane systolic excursion and RV systolic velocity (TV annulus S’), which showed preserved median values, indicating largely preserved RV function. However, RV dilatation and RV longitudinal dysfunction were observed in a minority of patients (6% and 4% respectively), suggesting subclinical RV involvement in a subset of TB cases.[16]

2D Speckle tracking

In our study, the mean GLS value before ATT showed mild improvement numerically post-therapy. Although this numeric improvement in GLS suggests a trend toward enhanced myocardial recovery, the change was not statistically significant and therefore not conclusive.

In a prospective observational study[17] involving 15 patients with tuberculous CP and 15 healthy controls, a significantly reduced GLS and global circumferential strain was found in the TB group.

Diastolic dysfunction

Among the 50 children enrolled in the study, diastolic dysfunction was observed in 16 patients (32%) prior to ATT initiation. Based on established Doppler echocardiographic criteria,[12] 6 patients had Grade 1 dysfunction while 10 had Grade 2 dysfunction. No patients fulfilled the criteria for Grade 3 dysfunction, suggesting the absence of severe diastolic abnormalities in this cohort. The dysfunction can be due to chronic inflammation, cytokine-mediated myocardial injury, Pulmonary hypertension, Subclinical myocarditis, Nutritional deficiencies or due to myopericarditis leading to fluid retention.

Following ATT, there was a marked improvement in diastolic function. Only 2 patients (4.2% of the total cohort and 14.3% of those with pre-existing dysfunction) continued to exhibit Grade 2 dysfunction. All patients with Grade 1 dysfunction showed complete resolution.

In a large study[18] conducted on 800 patients with active pulmonary TB, LV diastolic dysfunction was observed in 75% of patients, and 77.4% showed improvement after ATT. There was an association of diastolic dysfunction with elevated Troponin I levels, but the numbers are not large enough to arrive at a definitive conclusion.

Systolic dysfunction

In our study, 3 out of 50 children (6%) demonstrated evidence of systolic dysfunction at baseline, highlighting the substantial burden of myocardial involvement in TB. Notably, all patients who survived and were followed up after completion of ATT showed complete recovery, with normalization of both LVEF and GLS. The fact that 2 of the 3 children who died in the current study had systolic dysfunction before ATT adds clinical weight to its prognostic implications. Although causality cannot be definitively established, this association suggests that early identification and close monitoring of cardiac involvement – particularly systolic dysfunction – may have important therapeutic and prognostic value.

However no significant association was found with the level of biomarkers and systolic dysfunction due to limited sample size.

These findings align well with previous research that has highlighted reversible myocardial dysfunction in TB which found LV systolic dysfunction in 16.34% of patients, with significant improvement post-ATT in over 77% of cases; however, another study found full normalization of ejection fraction (>55%) within 8–12 weeks of ATT.[15]

Limitations

The study was limited to 50 children, which may not provide sufficient statistical power to generalize findings to the broader pediatric TB populations. Cardiac function was assessed only at baseline and after completion of ATT, without long-term follow-up to evaluate sustained recovery or late sequelae.

CONCLUSION

In this study, comprehensive echocardiographic evaluation, including M-mode, Doppler, tissue Doppler imaging, and GLS, revealed that a significant proportion of children with TB exhibit subclinical cardiac dysfunction at baseline. Most echocardiographic parameters, particularly tissue Doppler velocities and LVEF, improved after completion of ATT, indicating largely reversible myocardial involvement. These findings highlight the importance of incorporating detailed cardiac assessment in the evaluation of pediatric TB.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.