European Journal of Cardiovascular Medicine

Congenital heart disease (CHD) represents the most frequent congenital anomaly, with an incidence ranging from 4 to 10 per 1,000 live births. The survival rate of patients with CHD has increased remarkably due to advancements in diagnostic imaging and surgical interventions, with nearly 90% of affected infants surviving into adulthood. However, the accurate anatomical and functional delineation of both intracardiac and extracardiac structures remains critical for effective pre-surgical planning and postoperative management.^[1]

Historically, cardiac catheter angiography was the mainstay for diagnosing congenital heart defects, but its invasive nature and associated risks prompted the search for non-invasive alternatives. Two-dimensional transthoracic echocardiography (2D-TTE) has emerged as the first-line diagnostic tool because of its accessibility, non-invasiveness, portability, and ability to assess cardiac morphology and hemodynamics using Doppler techniques. Despite these advantages, echocardiography is operator-dependent and limited by acoustic window constraints, particularly in assessing complex CHDs and extracardiac vascular anomalies such as anomalies of the great arteries and pulmonary veins.^[2]

Magnetic resonance imaging (MRI) provides detailed structural and functional evaluation without radiation exposure, but its prolonged scan duration, higher cost, and lower spatial resolution make it less practical in emergency or pediatric settings. Multidetector computed tomography (MDCT)—particularly 64-slice and 128-slice scanners—has revolutionized cardiac imaging by offering rapid acquisition times, superior spatial resolution, and 3D reconstruction capabilities. These features enable accurate visualization of intricate cardiovascular anatomy, including extracardiac vessels, airways, and pulmonary parenchyma, with minimal sedation requirements in children. The primary limitation of CT lies in the exposure to ionizing radiation and the use of iodinated contrast, necessitating careful dose optimization.^[3]

With recent advancements such as 128-slice MDCT and ECG-gated acquisition, it is possible to achieve high-quality images at significantly reduced radiation doses. Several studies have demonstrated the potential of MDCT in providing diagnostic information comparable or superior to echocardiography, especially in delineating complex structural relationships and post-operative anatomy. However, echocardiography remains invaluable for functional and hemodynamic assessment. Therefore, combining both modalities offers a comprehensive evaluation, enhancing diagnostic accuracy, reducing diagnostic uncertainty, and guiding therapeutic decisions.^[4]

Aim

To evaluate the diagnostic correlation between three-dimensional computed tomography (3D-CT) and two-dimensional echocardiography (2D-Echo) in detecting congenital heart diseases at a tertiary care hospital.

Objectives

1. To assess the diagnostic capability and advantages of 3D-CT and 2D-Echocardiography in various congenital cardiac anomalies.
2. To correlate the findings of 3D-CT with 2D-Echocardiography for both intracardiac and extracardiac structural anomalies.
3. To determine the complementary role of both imaging modalities in the preoperative assessment and management of congenital heart diseases

Source of Data: The study included pediatric and adult patients clinically suspected or previously diagnosed with congenital heart disease who were referred for both echocardiographic and CT evaluation at the Department of Radiodiagnosis, at tertiary care hospital.

Study Design: Hospital-based analytical cross-sectional study.

Study Location: Department of Radiodiagnosis, at tertiary care hospital.

Study Duration: Two years.

Sample Size: 200 patients.

Inclusion Criteria:

• Patients (neonates to adults) clinically suspected or confirmed to have congenital heart disease.
• Patients who underwent both 2D-Echocardiography and 3D-CT evaluation.
• Patients providing informed consent (or parental consent for minors).

Exclusion Criteria:

• Patients with contraindications to iodinated contrast (renal dysfunction, allergy).
• Hemodynamically unstable patients not suitable for CT examination.
• Incomplete or poor-quality imaging data.

Procedure and Methodology: All participants underwent 2D-Echocardiography using standard parasternal, apical, subcostal, and suprasternal views. Cardiac chambers, septal defects, outflow tracts, and major vessels were assessed.

Subsequently, 128-slice MDCT angiography was performed using ECG-gated acquisition. Non-ionic contrast (1–2 ml/kg) was injected via peripheral IV line with bolus tracking at 100–120 HU threshold in the ascending aorta. Images were reconstructed with slice thickness 0.625 mm and analyzed on dedicated cardiac workstations. 3D-volume rendering, multiplanar reformations (MPR), and maximum intensity projections (MIP) were used to delineate intracardiac and extracardiac anomalies.

CT and echocardiography findings were independently interpreted by radiologists and cardiologists blinded to each other’s results. Diagnostic correlation was established by comparing concordant and discordant findings across modalities, with surgical or clinical diagnosis as the reference standard wherever available.

Sample Processing: CT datasets were anonymized and analyzed for anatomical details—chamber morphology, septal defects, valve abnormalities, great vessel anomalies, and extracardiac connections. Echocardiography results were similarly tabulated for structural and flow findings.

Statistical Methods: Descriptive statistics (frequency, percentage) were used for lesion distribution. Concordance between CT and echocardiography was analyzed using Cohen’s kappa coefficient (κ). Sensitivity, specificity, and accuracy were calculated against operative/surgical findings where available. A p-value <0.05 was considered statistically significant. Statistical analysis was performed using SPSS version 25.0.

Data Collection: Patient data including demographic details, clinical diagnosis, echocardiographic findings, and CT results were recorded in a structured proforma. All imaging findings were reviewed jointly for final consensus interpretation

Table 1: Overall diagnostic correlation between 3D-CT and 2D-Echo (N = 200)

Table 1 presents the overall diagnostic agreement between three-dimensional computed tomography (3D-CT) and two-dimensional echocardiography (2D-Echo) in detecting congenital heart diseases among 200 patients. Out of the total, 3D-CT identified 170 cases (85.0%) as positive for CHD, while 2D-Echo detected 165 cases (82.5%). The negative cases constituted 15.0% for CT and 17.5% for Echo. The overall proportion agreement between the two modalities was 92.5% (95% CI: 88.2–95.5), suggesting a high level of concordance. The Cohen’s kappa coefficient (κ = 0.73; 95% CI: 0.62–0.83; p < 0.001) indicated substantial agreement between the two diagnostic tools. The McNemar test (χ² = 1.67, p = 0.196) revealed no statistically significant difference, implying that both modalities performed comparably in the overall detection of CHD. The positive predictive value (PPV) of echocardiography against CT was 96.97%, while the negative predictive value (NPV) was 71.43%, showing that 2D-Echo accurately detected most true positive cases, but occasionally missed certain lesions identified on CT.

Table 2: Diagnostic capability by anomaly class

Table 2 evaluates the sensitivity of 3D-CT and 2D-Echo across different classes of congenital cardiac anomalies, benchmarked against clinico-surgical findings. In septal defects (VSD, ASD, AVSD; n = 80), CT achieved 96.3% sensitivity versus 90.0% for Echo (χ² = 4.17, p = 0.041), signifying a statistically significant difference. For outflow tract and great-vessel anomalies (TOF, TGA, CoA; n = 40), CT showed 95.0% sensitivity compared to 85.0% for Echo (p = 0.046).
The greatest disparity appeared in pulmonary venous anomalies (TAPVR, PAPVR; n = 20), where CT reached 95.0% sensitivity versus 65.0% for Echo (χ² = 6.13, p = 0.013), reflecting the advantage of CT in visualizing extracardiac venous pathways. For complex cyanotic CHD (DORV, Truncus arteriosus; n = 30), sensitivities were 93.3% for CT and 86.7% for Echo, but the difference was not statistically significant (p = 0.317). When extracardiac vascular anomalies were pooled (n = 50), CT outperformed Echo (96.0% vs. 74.0%; p = 0.004).

Table 3: Correlation for intracardiac vs extracardiac structures

Table 3 compares the correlation between CT and Echo in assessing intracardiac and extracardiac structures. For intracardiac lesions involving chambers, septa, and valves (n = 120), CT detected 94.2% (113/120) and Echo 91.7% (110/120). The overall agreement between modalities for these structures was 98.0%, with a Cohen’s κ = 0.82 (95% CI: 0.73–0.90), indicating excellent agreement. The McNemar test (χ² = 0.50, p = 0.480) suggested no significant difference, confirming comparable accuracy in intracardiac assessments. In contrast, for extracardiac anomalies (aortic arch, pulmonary arteries, and systemic veins; n = 80), CT detected 96.3% (77/80) versus 76.3% (61/80) by Echo. The agreement was slightly lower (93.0%), and the kappa value of 0.68 (95% CI: 0.55–0.79) denoted moderate agreement. The McNemar test (χ² = 9.00, p = 0.003) revealed a significant difference, confirming that 3D-CT is markedly superior in identifying extracardiac and vascular anomalies.

Table 4: Complementary role in pre-operative assessment and management (N = 200)

Table 4 demonstrates how combining 3D-CT and 2D-Echo enhances preoperative planning and management of CHD. A surgically adequate anatomical roadmap was achieved in 95.0% of cases with the combined approach, compared to 88.0% with CT alone and 82.0% with Echo (z = 3.10, p = 0.002). Additionally, 44 patients (22.0%) had actionable extracardiac findings—such as anomalous pulmonary venous return, arch hypoplasia, or airway anomalies—identified exclusively by CT (p < 0.001). For hemodynamic assessment essential in anesthesia risk evaluation, the combined use of both modalities yielded 95.5% adequacy, significantly higher than Echo alone (89.0%; p = 0.002). The requirement for invasive angiography dropped from 12.0% (Echo alone) and 9.0% (CT alone) to 2.0% with the combined approach (χ² = 10.4, p = 0.001), showing that integrating the two techniques reduced diagnostic uncertainty. Diagnostic confidence, measured on a Likert scale, also improved marginally (88.0% vs. 87.0% for CT, 84.0% for Echo).

Cohort shows high overall agreement (92.5%) between 3D-CT and 2D-Echo for detecting CHD with substantial concordance (κ=0.73) and a non-significant McNemar test—indicating no systematic bias in overall detection. This aligns with multi-modality reviews and inter-modality series reporting strong but not perfect agreement, especially when extracardiac structures are considered. In particular, pediatric CT reviews emphasize that while Echo is first-line, CT provides decisive anatomic mapping that raises concordance and confidence in complex cases—consistent with our κ in the “substantial” range. Zhang K et al.(2024)^[6]

By anomaly class (Table 2), CT sensitivity exceeded Echo for septal defects (Δ=+6.3%), outflow/great-vessel anomalies (Δ=+10%), and most notably pulmonary venous anomalies (Δ=+30%; p=0.013). These patterns mirror the literature: multi-detector CT (64–320 row) consistently excels at delineating great-vessel relationships, arch hypoplasia/coarctation extent, and anomalous pulmonary venous connections, where Echo’s acoustic windows can be limiting. Comparative studies of ECG-gated MDCT vs Echo report significantly better detection of extracardiac findings and surgical detail with CT—matching our higher CT sensitivities in outflow and venous categories. Moceri P et al.(2021)^[7]

For TAPVR/PAPVR, our CT sensitivity of 95% vs Echo 65% echoes prior observations that echocardiography can miss mixed/obstructed pulmonary venous drainage patterns, whereas CT reliably shows the entire venous course and systemic connections. Reports document lower Echo performance in complex/mixed TAPVR and higher accuracy with CT angiography for preoperative definition, concordant with our paired analysis. Xie WH et al.(2022)^[8]

When we separated intracardiac from extracardiac assessments (Table 3), both modalities agreed excellently for intracardiac lesions (κ=0.82; McNemar NS), but CT was significantly superior for extracardiac mapping (κ=0.68; McNemar p=0.003). This is precisely the domain where reviews and guideline-style statements place CT’s greatest value—comprehensive 3D depiction of aortic arch variants, pulmonary arteries/veins, systemic venous anomalies, airway adjacency, and post-operative pathways. Echo’s known limitations in arch and branch pulmonary visualization explain our lower extracardiac Echo yield. Sachdeva R et al.(2024)^[9]

The complementary clinical impact (Table 4) is also supported by literature. We observed that combining CT+Echo increased the rate of “surgically adequate roadmaps” to 95% and reduced “indeterminate/needs invasive angiography” labels to 2%, while uncovering actionable extracardiac findings in 22%—all consistent with contemporary CCTA/MDCT reviews highlighting CT’s role in pre-operative decision-making, especially in conotruncal lesions and complex cyanotic CHD. Such studies describe how CT reduces uncertainty, guides operative approach, and may obviate invasive angiography in many patients—closely paralleling our risk-difference and risk-ratio effects. Majid A et al.(2025)^[10]

The present study demonstrates that three-dimensional computed tomography (3D-CT) and two-dimensional echocardiography (2D-Echo) show a high degree of diagnostic concordance in the evaluation of congenital heart diseases (CHD), with an overall agreement of 92.5% and a substantial Cohen’s κ value of 0.73 (p < 0.001). While 2D-Echo remains the primary, non-invasive modality for initial screening and functional assessment, 3D-CT significantly enhances anatomical delineation, particularly for extracardiac vascular and complex structural anomalies. CT exhibited higher sensitivity in identifying outflow tract, pulmonary venous, and vascular anomalies, areas where echocardiography was limited by acoustic window and operator dependency.
The integration of both modalities provided the most comprehensive diagnostic accuracy, improving surgical planning adequacy to 95% and reducing the need for invasive angiography to 2%. Thus, a combined imaging approach using 3D-CT and 2D-Echo is recommended for complete preoperative assessment and management of complex CHD cases in tertiary care settings.

LIMITATIONS OF THE STUDY

1. The study was single-centered and conducted in a tertiary care hospital, which may limit generalizability to broader or primary care populations.
2. The sample size (n = 200), though adequate for statistical analysis, may not fully represent the entire spectrum of congenital heart anomalies.
3. CT imaging involves ionizing radiation and contrast administration, which restricts its routine use in all pediatric patients, despite dose-reduction techniques.
4. Echocardiographic accuracy was operator-dependent, and inter-observer variability was not evaluated in this study.
5. Surgical or catheterization confirmation was available only in a subset of cases, which could have introduced bias in diagnostic accuracy comparison.
6. Functional and hemodynamic data were better captured by echocardiography; CT was limited to anatomical assessment only.
7. The study did not incorporate cardiac MRI correlation, which could have provided a radiation-free comparative modality.

1. Surkova E, Brida M, Muraru D, van den Bosch A, Elsayed HM, Li W, Gatzoulis MA, Di Salvo G, Bohbot Y, Grapsa J, Keenan N. Three-dimensional echocardiography in adults with congenital heart disease: a scientific statement of the European Society of Cardiology Working Group on Adult Congenital Heart Disease and the European Association of Cardiovascular Imaging of the European Society of Cardiology. European Heart Journal-Cardiovascular Imaging. 2025 Mar 28:jeaf105.
2. Fournier E, Batteux C, Mostefa-Kara M, Valdeolmillos E, Maltret A, Cohen S, Van Aerschot I, Guirgis L, Azarine A, Sigal-Cinqualbre A, Provost B. Cardiac tomography-echocardiography imaging fusion: a new approach to congenital heart disease. Revista Española de Cardiología (English Edition). 2023 Jan 1;76(1):10-8.
3. Nussbaumer C, Bouchardy J, Blanche C, Piccini D, Pavon AG, Monney P, Stuber M, Schwitter J, Rutz T. 2D cine vs. 3D self-navigated free-breathing high-resolution whole heart cardiovascular magnetic resonance for aortic root measurements in congenital heart disease. Journal of Cardiovascular Magnetic Resonance. 2021 Mar 1;23(1):65.
4. Shiina Y, Inai K, Takahashi T, Shimomiya Y, Nagao M. Clinical impact of cardiac computed tomography derived three-dimensional strain for adult congenital heart disease: a pilot study. The international journal of cardiovascular imaging. 2020 Jan;36(1):131-40.
5. D’Alberti E, Mappa I, Rizzo G. Accuracy of 3D Ultrasound in the Prenatal Diagnosis of Congenital Heart Diseases. In3D Ultrasound of Fetal Heart 2025 Jul 16; pp. 35-48.
6. Zhang K, Zhong J. Bio inspired technological performance in color Doppler ultrasonography and echocardiography for enhanced diagnostic precision in fetal congenital heart disease. SLAS technology. 2024 Dec 1;29(6):100207.
7. Moceri P, Duchateau N, Gillon S, Jaunay L, Baudouy D, Squara F, Ferrari E, Sermesant M. Three-dimensional right ventricular shape and strain in congenital heart disease patients with right ventricular chronic volume loading. European Heart Journal-Cardiovascular Imaging. 2021 Oct 1;22(10):1174-81.
8. Xie WH, Chen LJ, Hu LW, Ouyang RZ, Guo C, Sun AM, Wang Q, Qiu HS, Zhang YQ, Zhang H, Fu QH. Cardiac computed tomography-derived left atrial strain and volume in pediatric patients with congenital heart disease: a comparative analysis with transthoracic echocardiography. Frontiers in Cardiovascular Medicine. 2022 Jun 20;9:870014.
9. Sachdeva R, Armstrong AK, Arnaout R, Grosse-Wortmann L, Han BK, Mertens L, Moore RA, Olivieri LJ, Parthiban A, Powell AJ. Novel techniques in imaging congenital heart disease: JACC scientific statement. Journal of the American College of Cardiology. 2024 Jan 2;83(1):63-81.
10. Majid A, Obungoloch J, Enywaku A, Francis O, Jjuuko D, Bizimana E, Joshua B, William W. Diagnosis of Congenital Heart Diseases in Children from 2D and 3D Sonography using Convolutional Neural Networks: A Scoping Literature Review. WFUMB Ultrasound Open. 2025 Oct 8:100096