Decoding Heart Failure in a Restrictive-Appearing Phenotype Using CMR in an Adolescent with Noonan Syndrome and a Large Atrial Septal Defect

Shaffin Siddiqui MPhil AB, Ryan J. Butts MD, Mohammad Tarique Hussain MD PhD

Clinical History

A 13-year-old girl with severe failure to thrive and chronic constipation presented with two to three weeks of progressive dyspnea, orthopnea, dizziness, and vomiting. Chest radiography demonstrated cardiomegaly with bilateral pleural effusions (Figure 1). Weight-for-age was below the first percentile (Z = −4.2), with a body surface area of 1.00 m². Central venous pressure, measured directly from a central venous catheter, was 26 mmHg, consistent with markedly elevated systemic venous pressure, and B-type natriuretic peptide (BNP) was 3,205.5 pg/mL (reference range, 10.0–120.0 pg/mL).

Figure 1. Presenting anteroposterior chest radiograph demonstrating marked cardiomegaly with pulmonary venous congestion and pleural effusion.

Electrocardiography demonstrated atrial tachycardia, refractory to esmolol and a single dose of flecainide; sinus rhythm was restored by cardioversion. Transthoracic echocardiography (TTE) demonstrated a large secundum atrial septal defect (ASD) with left-to-right shunting (Movie 1), severely dilated atria with normal ventricular size and systolic function, mitral valve prolapse (MVP) with two jets of mild mitral regurgitation (MR) (Movie 2), bilateral superior venae cavae with a persistent left superior vena cava (PLSVC) draining to a dilated coronary sinus, and severely dilated central pulmonary arteries. The combination of severe biatrial dilation, preserved biventricular systolic function, and low-output heart failure raised initial concern for restrictive cardiomyopathy.

Movie 1. Preoperative TTE, subcostal view with color Doppler, demonstrating a large secundum ASD with left-to-right shunting.

Movie 2. Preoperative TTE, four-chamber view, demonstrating severe biatrial dilation and MVP.

Genetic testing performed during the admission established a diagnosis of Noonan syndrome with a pathogenic PTPN11 variant. Following aggressive diuresis and initiation of milrinone, diagnostic cardiac catheterization under conscious sedation demonstrated right ventricular (RV) pressure of 35/8 mmHg and left ventricular (LV) pressure of 71/9 mmHg (systolic/end-diastolic), a mean pulmonary artery pressure of 24 mmHg, and a mean pulmonary capillary wedge pressure of 8 mmHg. Using measured oxygen consumption, the pulmonic flow (Qp) to systemic flow (Qs) ratio (Qp:Qs) was 4.7, indexed systemic blood flow was 2.7 L/min/m², and indexed pulmonary vascular resistance was 1.2 WU x m² (Table 1). Pulmonary vasoreactivity testing was not performed. Cardiovascular magnetic resonance (CMR) was performed the following day for anatomic and functional assessment and tissue characterization.

Table 1. Baseline cardiac catheterization hemodynamics (room air, conscious sedation). 

EDP = end-diastolic pressure; PA = pulmonary artery; SVC = superior vena cava; WU = Wood units.

CMR Findings

CMR was performed on a 1.5 T whole-body MR system (Ingenia, Philips Healthcare, Best, the Netherlands) without sedation at a heart rate of 94–96 BPM, using cine balanced steady-state free precession (bSSFP), T2-weighted imaging, phase-contrast flow (aortic valve, ascending and descending aorta, superior vena cava, main and branch pulmonary arteries), phase-sensitive inversion recovery (PSIR) late gadolinium enhancement (LGE), and respiratory-navigated three-dimensional whole-heart imaging. Motion and breathing artifact were present. Body surface area was 1.00 m² (Mosteller); absolute and index values are reported in Table 2.

There was a large secundum ASD measuring 31 × 23 mm with a deficient posterior margin, with severe biatrial enlargement (Figure 2, Movie 3). Atrial volumes were measured at maximal atrial volume (ventricular end-systole), using the biplane method for the left atrium and the four-chamber view for the right atrium; left atrial volume was 87.6 mL (87.6 mL/m²) and right atrial volume was 205 mL (205 mL/m²). A central venous catheter was visible in the right atrium, and there was no intracardiac thrombus. A small PLSVC drained to a dilated coronary sinus with a bridging vein present. The coronary origins were normal, and the aortic arch was left-sided and unobstructed with a common origin of the brachiocephalic and left carotid arteries. The main and branch pulmonary arteries were dilated.

Figure 2. Axial bSSFP view at end-diastole demonstrating severe biatrial enlargement and severe RV dilation. The small round density in the right atrium is a central venous catheter.

Movie 3. Axial cine bSSFP demonstrating a large ASD, severe biatrial enlargement, and RV chamber dilation.

Phase-contrast flow demonstrated a net antegrade ascending aortic flow of 1.8 L/min/m² against a net antegrade main pulmonary artery flow of 10.6 L/min/m², yielding a Qp:Qs of 5.9 in the setting of low cardiac output (Figure 3, Table 2). This exceeded the catheter-derived Qp:Qs of 4.7 obtained the previous day, a difference likely reflecting the different methods (Fick principle versus direct phase-contrast flow measurement) and differing loading conditions between the two studies. The RV was severely dilated (RV end-diastolic volume index 307 mL/m²) with normal systolic contractility and a RV ejection fraction of 52%; the LV was mildly dilated (LV end-diastolic volume index 103 mL/m²) with normal contractility and a LV ejection fraction of 59%, giving a right-to-left ventricular end-diastolic volume ratio of 3.0 (Movie 4).

Figure 3. Through-plane phase-contrast flow analysis with regions of interest at the ascending aorta (left) and main pulmonary artery (right), and the corresponding flow-versus-time curves. The pulmonary flow curve (blue) greatly exceeds the systemic flow curve (red).

Figure 4. Four-chamber CMR image demonstrating anterior MVP.

Movie 4. Four chamber cine bSSFP showing a large ASD, severe biatrial dilation, severe RV chamber dilation, normal biventricular systolic function, anterior leaflet MVP, and moderate tricuspid regurgitation (TR).

The mitral valve demonstrated anterior leaflet prolapse producing severe regurgitation (Figure 4, Movie 4); derived from the volumetric and flow data, the regurgitant volume was 42.6 mL/beat against a LV stroke volume of 61 mL/beat, corresponding to a regurgitant fraction of approximately 70% (Table 2). The tricuspid valve was thin and mobile with moderate central regurgitation (regurgitant fraction 29%). The aortic valve was competent. There was no LGE and no vascular wall enhancement, and myocardial native T1, T2, and extracellular volume were within normal limits (Figure 5).

Table 2. CMR ventricular volumetry, phase-contrast flow, and derived shunt and regurgitant volumes. Heart rate 94–96 bpm. Papillary muscles and trabeculations were included in the blood pool volumes. 

LV = left ventricle; MR = mitral regurgitation; RF = regurgitant fraction; RV = right ventricle; TR = tricuspid regurgitation.

Figure 5. Four chamber (A), three chamber (B), and mid short axis (C) PSIR LGE images demonstrating absence of myocardial enhancement.

Conclusion

CMR characterized a large secundum ASD with severe mitral regurgitation due to anterior leaflet prolapse, producing a Qp:Qs of 5.9 with a systemic output of 1.8 L/min/m² and severe RV volume loading, in the presence of normal biventricular contractility and no LGE. TR, graded moderate-to-severe by TTE, was moderate by CMR at 29%. Normal parametric mapping and the absence of LGE argued against an infiltrative or fibrotic cardiomyopathy and reframed the presentation as a reversible mechanical loading lesion.

Following nutritional rehabilitation via nasogastric feeding and heart failure therapy (intravenous milrinone and furosemide during admission, followed by oral furosemide, spironolactone, sacubitril/valsartan, and dapagliflozin, with atenolol and flecainide for rhythm control), the patient underwent surgical repair. Intraoperative inspection confirmed prolapse of the entire anterior mitral leaflet with elongated chordae. The mitral valve was repaired by chordal shortening at four sites, plicating the elongated chordae onto the papillary muscle heads, with a reinforcing posterior suture annuloplasty sized to 22 mm (Z-score 0). The massively dilated tricuspid valve was repaired by closure of the anteroseptal commissure and a reinforcing inferior suture annuloplasty sized to 22 mm (Z-score 0). Given concern for elevated left-sided filling pressures after closure, the ASD was closed with a bovine pericardial patch with a 4 mm fenestration. Post-bypass transesophageal echocardiography demonstrated trivial MR and TR and mild mitral valve stenosis (gradient 3–5 mmHg). Post-operative TTE (Movie 5) demonstrated a small residual ASD with entirely left-to-right shunting, mild MR with persistent prolapse, mild TR, normal LV systolic function, and a mildly hypertrophied RV. The TR peak gradient decreased from 39 to 26 mmHg, consistent with, although not by itself establishing, a reduction in RV systolic pressure. BNP had decreased from 3,205.5 pg/mL at presentation to 536.2 pg/mL at hospital discharge, 12 days postoperatively. She was weaned to oral heart failure therapy, which was able to be discontinued as an outpatient. At one-month follow-up she was clinically well, without dyspnea, edema, or palpitations. 

Movie 5. Postoperative TTE, four-chamber view, following fenestrated patch closure of the ASD and mitral and tricuspid valve repair.

Perspective

Initial TTE findings and hemodynamics raised concern for a primary cardiomyopathic process. Further, because cardiomyopathy is part of the Noonan/RASopathy spectrum, an intrinsic myopathy was a real consideration in this child.[1] However, several features prompted reconsideration of this interpretation. The presence of normal ventricular end-diastolic pressures after aggressive diuresis (LV end-diastolic pressure 9 mmHg, RV end-diastolic pressure 8 mmHg), marked biventricular dilation, an exceptionally large left-to-right shunt, and severe MR were difficult to reconcile with advanced myocardial disease as the primary pathology. The absence of LGE and normal native T1 values provided additional evidence against a diffuse or infiltrative cardiomyopathic process and supported the conclusion that the ventricular remodeling was primarily hemodynamic in origin.

The clinical presentation was also informative. The onset of atrial arrhythmia accompanied by heart failure symptoms is a classic manifestation of long-standing ASD in later life, reflecting the natural history described by Campbell and colleagues many decades ago.[2] However, atrial arrhythmias are also frequently present in restrictive cardiomyopathy due to atrial stretch. Symptomatic heart failure is uncommon in children with isolated ASD. In this patient, the severity of the presentation was likely amplified substantially by the coexisting severe MR, which imposed an additional volume load on the left heart and accelerated the development of heart failure symptoms.

The magnitude of the shunt likely reflects severe MR raising left atrial pressure across a large, unrestrictive defect with low pulmonary vascular resistance (1.2 WU x m²). It is reasonable to conjecture that the Qp:Qs was further exaggerated by increased LV stiffness, which may be the result of prolonged tachyarrhythmia or chronic MR.

An important physiological consideration is that a large ASD may partially decompress the left atrium and thereby limit the degree of LV enlargement that would otherwise be expected in severe MR. This atrial “pop-off” vents the left atrium, masks the severity of the left-sided lesion, and volume-loads the right heart. Consequently, the absence of profound LV dilation should not be interpreted as evidence against severe regurgitation in this setting.[3] Moreover, the severity of the MR had initially been underestimated because the regurgitant jet was highly eccentric and associated with MVP, a well-recognized circumstance in which conventional TTE assessment may underestimate lesion severity.[4] Because MR is load dependent, differences in rhythm and loading conditions between studies, including atrial tachycardia before cardioversion and subsequent diuresis and milrinone, may also have contributed to the discordant grading. Taken together, these findings support the view that the ventricular enlargement and heart failure phenotype were driven predominantly by the combined hemodynamic effects of a large ASD and severe MR rather than by an underlying cardiomyopathy.

Click here to view the entire CMR on CloudCMR. 

References

1. Reynolds G, Gazzin A, Carli D, et al. Update on the clinical and molecular characterization of Noonan syndrome and other RASopathies: a retrospective study and systematic review. Int J Mol Sci. 2025;26(8):3515. 
2. Campbell M. Natural history of atrial septal defect. Br Heart J. 1970;32(6):820-826.
3. Ghazal SN. Chronic severe mitral regurgitation with normal left ventricular size — a case with coexisting atrial septal defect. J Cardiovasc Echography. 2018;28(2):133-137.
4. Baessato F, Fusini L, Muratori M, et al. Echocardiography vs. CMR in the quantification of chronic mitral regurgitation: a happy marriage or stormy divorce? J Cardiovasc Dev Dis. 2023;10(4):150.