Here we report the first infantile case of restrictive cardiomyopathy caused by a de novo mutation of the cardiac troponin T gene. The patient presented with an apparent life-threatening event. She developed malignant arrhythmias and hemodynamic instability, requiring initial rescue support with extracorporeal membrane oxygenation, and subsequently underwent insertion of a biventricular assist device (VAD). She successfully received an orthotopic heart transplant 172 days after VAD implantation.
- restrictive cardiomyopathy
- cardiac troponin T
- myosin binding protein C
- extracorporeal membrane oxygenation
- ventricular assist-device
- cardiac transplantation
Restrictive cardiomyopathy (RCM) is a rare cardiomyopathic disorder characterized by abnormal diastolic function resulting from impaired ventricular filling, increased ventricular end-diastolic pressures, and dilated atria. RCM is a rare cause of cardiomyopathies in children (<5%) and has a poor prognosis, with an average 2-year survival rate of <50% from time of diagnosis.1,2 The incidence of sudden cardiac death in pediatric patients with RCM with electrocardiographic evidence of ischemia has been reported as high as 23%.3 Etiologically, RCM may be classified as primary or secondary. The primary RCMs include endomyocardial fibrosis, Loeffler's endocarditis, and idiopathic RCM. The secondary forms of RCM are more common and can be subclassified as noninfiltrative (eg, anthracycline toxicity), infiltrative (eg, amyloidosis, sarcoidosis), or storage disorders (eg, glycogen-storage disease, Fabry disease). Pediatric RCM is most often idiopathic.4 Here we report a novel presentation of a mutation in the cardiac troponin T gene causing RCM in a child.
A previously healthy, 12-month-old girl with a 1-week history of nasal congestion was found limp and pale with some shaking of her extremities. Emergency medical services reported that she was cyanotic but breathing spontaneously. Evaluation at the referring hospital included normal computed head tomography and chest radiograph results and laboratory values. She was transferred to our institution for additional evaluation of an apparent life-threatening event.
An electrocardiogram (ECG) disclosed sinus rhythm with marked right atrial enlargement, left axis deviation, and nonspecific ST-T wave changes (Fig 1A). An echocardiogram (ECHO) revealed a structurally normal heart with severely dilated atria, no mitral or tricuspid valve regurgitation, mild-to-moderate left ventricular systolic dysfunction with a fractional shortening of 24% (normal for age: 33.9–43.5%), and mild right-ventricular systolic dysfunction (Fig 1B). Doppler inflow velocity pattern across the mitral valve (E/A = 0.67), tissue Doppler velocity at the mitral valve annulus (e/a = 0.5), and the mitral valve inflow velocity to tissue Doppler velocity ratio (E/e = 15) were all abnormal, consistent with left ventricular diastolic dysfunction.5 The left ventricular chamber size and wall thicknesses were normal (end-diastolic dimension: 2.57 cm [reference range: 2.38–3.28 cm]; end-diastolic wall thickness: 0.49 cm [reference range: 0.40–0.67 cm]), and no regional wall-motion abnormalities were found. The coronary artery anatomy and flow, as well as the pericardium, were normal. A presumptive diagnosis of RCM was made on the basis of the ECG and ECHO findings. She was listed emergently for a heart transplant, and additional testing for the etiology of her cardiomyopathy was pursued.
Over the next days, she experienced recurrent episodes of sinus bradycardia and tachycardia associated with diffuse ischemic ECG changes and pronounced hypotension. She was subsequently intubated for hemodynamic instability and was referred for cardiac catheterization. Normal coronary artery anatomy and flow patterns were confirmed, which eliminated anomalous coronary artery origin or coronary ostial stenosis as possible etiologies for her cardiac dysfunction. Hemodynamic evaluation confirmed the diagnosis of RCM, with typical square-root sign, and elevated atrial and right and left ventricular end-diastolic pressures. Endomyocardial specimens were obtained without complication.
The events with bradycardia, tachycardia, and hypotension became more frequent, and severe ventricular dysfunction ensued, resulting in institution of ECMO. During 10 days of ECMO support, 3 weaning trials were unsuccessful, and investigation into biventricular assist devices (VADs), as alternatives to ECMO, began.
Approval for implantation of a Berlin heart pediatric paracorporeal VAD (Berlin Heart, Berlin, Germany) was granted on a compassionate-use basis by the US Food and Drug Administration and the Johns Hopkins Institutional Review Board. Intraoperatively, severe diastolic dysfunction and hypokinesia of the left ventricular apex were found, with a corrugated and discolored appearance of the ventricular epicardium. The child returned to the operating room 36 hours later for delayed primary chest closure and was extubated on postoperative day 11. She was weaned off all intensive care monitoring. She successfully received an orthotopic heart transplant 172 days after VAD implantation.
Etiologic investigations of her RCM included normal karyotype, plasma amino acids, and urine organic acids. Endomyocardial biopsy revealed moderate hypertrophy of the myocytes with patchy mild interstitial fibrosis but no evidence of an inflammatory infiltrate, glycogen deposits, fibroelastosis, iron, or giant cells. Abnormally shaped mitochondria and mild dilation of the sarcoplasmic reticulum were noted on ultrastructural examination of the specimens, with no viral inclusions (Fig 2). Histologic examination of the explanted heart showed significant myocyte disarray and multiple areas of replacement fibrosis.
Genetic testing was performed by direct sequencing of the coding regions of the 8 genes most commonly implicated in hypertrophic cardiomyopathy (HCM) (β-cardiac myosin heavy chain [MYH7], cardiac myosin-binding protein C [MYBPC3], cardiac troponin T gene [TNNT2], cardiac troponin I [TNNI3], α-tropomyosin [TPM1], α-cardiac actin [ACTC], regulatory myosin light chain [MYL2], and essential myosin light chain [MYL3]). She was found to have a novel in-frame 285-to-287 GGA deletion in exon 9 of the TNNT2 leading to the deletion of glutamine in amino acid position 96, as well as an MYBPC3 polymorphism (c.2686G→A in exon 27, p.Val896Met). It was subsequently shown that the TNNT2 mutation was absent in both parents, confirming the de novo occurrence of this change, whereas the MYBPC3 variant was also found in the father. Both parents had normal ECHOs at ages 28 and 34, respectively.
The importance of monogenic disorders in the etiology of cardiomyopathies has been recognized over the last decade.6 Mutations in genes encoding for several proteins of the sarcomere, the contractile unit of cardiac muscle, have been identified in both HCM and dilated cardiomyopathy (DCM), classifying these cardiomyopathies as “sarcomeric diseases.”6 Mutations in 11 different genes have been associated with HCM, including genes for α-cardiac and β-cardiac myosin heavy chains (MYH6 and MYH7), TNNT2, TPM1, MYBPC3, essential (MYL3) or regulatory (MYL2) myosin light chain, TNNI3, cardiac troponin C (TNNC1), ACTC, and titin (TTN).7 DCM has been linked to at least 16 different genes, including TNNT2.8 Recently, 6 TNNI3 mutations were found to be associated with idiopathic RCM in families that also segregated HCM, which supports the notion that this phenotype may also be part of the spectrum of hereditary sarcomeric diseases.9
Stimulated by these observations, we obtained genetic testing of all genes most commonly involved in HCM in our patient. This led to the identification of a de novo deletion (c.285-287GGA deletion, p.96delGlu) in the TNNT2 gene, a gene previously not associated with RCM. As such, TNNT2 mutations have now been associated with all 3 types of cardiomyopathy. It is interesting to note that we also found a missense change (c.2686G→A, p.Val896Met) in the MYBPC3 gene. This mutation was originally thought to be disease-causing10 and then subsequently believed to be a polymorphism.11,12 More recently, this missense change has been considered to be a modifier that may contribute to phenotype severity when found in association with a second mutation.12–14 There is increasing evidence that compound heterozygosity for mutations in the same or different sarcomeric protein–encoding genes leads to more severe and earlier-onset cardiomyopathy.15,16 Recently, Blok et al15 reported the case of a 9-year-old girl with severe RCM resulting from 2 mutations in TNNI3; in addition, a severe case of neonatal HCM was caused by compound heterozygous mutations in the MYBPC3.16 In our patient, the presence of the MYBPC3 mutation may have aggravated the phenotypic expression of the TNNT2 mutation, perhaps causing an unusual early onset of RCM. However, without functional analysis of the MYBPC3 protein, this hypothesis remains to be proven.
Contraction of cardiomyocytes is regulated by calcium via its binding to a specific regulatory protein complex, troponin. Troponin is a complex of 3 different proteins: troponin T (TnT; tropomyosin-binding component), troponin I (TnI; inhibitory component), and troponin C (TnC; calcium-binding component). After calcium binding to TnC, the inhibitory action of TnI exerted on the thin filament is relieved, enabling the myosin head to interact with actin in the thin filament and generate force. A mutation involving cardiac troponin T mutation, such that our patient was found to have, likely leads to altered calcium sensitivity of the troponin complex, as previously shown for DCM and HCM,17 contributing to altered relaxation of the cardiac muscle, which is the hallmark of RCM.
Idiopathic RCM is noninfiltrative, and interstitial fibrosis is often the only detectable histologic abnormality of the myocardium.4 However, several groups have reported varying degrees of hypertrophy with disorganization and even myocardial fiber disarray on histologic specimens of patients with RCM.18–20 The histology of our patient's native heart is consistent with these findings. The biopsy finding of abnormally shaped mitochondria has not been associated with RCM previously. Given this distinctive pattern of mitochondria morphology, it is unlikely that it merely represents secondary abnormalities of a failing heart. Rather, it is possible that mutations in TNNT2 and/or MYBPC3 cause morphologic abnormalities of mitochondria by altering the cytoskeletal architecture, leading to perturbation of mitochondrial energy metabolism and contributing to the pathogenesis of RCM. We are currently investigating the potential impact of cytoskeletal abnormalities on mitochondrial morphology and function.
Our patient's continued hemodynamic instability, associated with her identified adverse risk factors for sudden cardiac death resulting from her RCM (young age,1 female gender,3 and continual signs and symptoms of ischemia), led us to institute mechanical circulatory support quickly and early in her hospital course. Very few successful interventions in the pediatric patient with end-stage RCM have been reported in the literature. Medical treatment commonly includes β-blocking agents or angiotensin-converting enzyme inhibitors, but because of the lack of efficacy,3 cardiac transplantation is often considered the treatment of choice.1,2 We chose ECMO as our first line of support because of the rapidity with which it could be instituted and our experience with this modality. When it became evident that she could not be weaned successfully from ECMO, we switched from this short-term therapy to a longer-term bridging device (VAD) as she waited for a suitable donor heart.
We present a novel presentation of a mutation in the TNNT2 gene causing RCM in a child. These findings, together with a possibly aggravating mutation in MYBPC3, further expand the phenotypic spectrum caused by TNNT2 and MYBPC3 mutations. Cardiac transplantation is the treatment of choice for RCM, but because the pool of available pediatric donor hearts remains small, long-term mechanical-support modes should be thought of and instituted early.
We acknowledge the services of the Laboratory for Molecular Medicine for Genetics and Genomics (H.L. Rehm, PhD, and R. Kucherlapati, PhD), Harvard Medical School (Boston, MA) and also appreciate their useful discussions.
- Accepted October 26, 2005.
- Address correspondence to Luca A. Vricella, MD, Department of Cardiac Surgery, Blalock 618, Johns Hopkins University School of Medicine, 600 N Wolfe St, Baltimore, MD 21287. E-mail: firstname.lastname@example.org
The authors have indicated they have no financial relationships relevant to this article to disclose.
Drs Peddy and Vricella contributed equally to this work.
- ↵Rivenes SM, Kearney DL, Smith EO, Towbin JA, Denfield SW. Sudden death and cardiovascular collapse in children with restrictive cardiomyopathy. Circulation.2000;102 :876– 882
- ↵Blok R, van den Wijngaard A, Merckx D, et al. Two novel TNNI3 mutations in restrictive cardiomyopathy [poster P0135]. Poster presented at: European Human Genetics Conference 2005; May 7–10, 2005; Prague, Czech Republic. Available at: www.eshg.org/eshg2005/index1.htm. Accessed March 15, 2005
- ↵Baars MJH, Muurling-Vlietman JJ, Hruda J, et al. Severe neonatal hypertrophic cardiomyopathy caused by compound heterozygous mutations in MYBPC3 [poster P088]. Poster presented at: European Human Genetics Conference 2005; May 7–10, 2005; Prague, Czech Republic. Available at: www.eshg.org/eshg2005/index1.htm. Accessed March 15, 2005
- ↵Robinson P, Mirza M, Knott A, et al. Alterations in thin filament regulation induced by a human cardiac troponin T mutant that causes dilated cardiomyopathy are distinct from those induced by troponin T mutants that cause hypertrophic cardiomyopathy. J Biol Chem.2002;277 :40710– 40716
- Copyright © 2006 by the American Academy of Pediatrics