Case Reports: Emery-Dreifuss Muscular Dystrophy Presenting as a Heart Rhythm Disorders in Children

Emery-Dreifuss muscular dystrophy (EDMD) is inherited muscle dystrophy often accompanied by cardiac abnormalities in the form of supraventricular arrhythmias, conduction defects and sinus node dysfunction. Cardiac phenotype typically arises years after skeletal muscle presentation, though, could be severe and life-threatening. The defined clinical manifestation with joint contractures, progressive muscle weakness and atrophy, as well as cardiac symptoms are observed by the third decade of life. Still, clinical course and sequence of muscle and cardiac signs may be variable and depends on the genotype. Cardiac abnormalities in patients with EDMD in pediatric age are not commonly seen. Here we describe five patients with different forms of EDMD (X-linked and autosomal-dominant) caused by the mutations in EMD and LMNA genes, presented with early onset of cardiac abnormalities and no prominent skeletal muscle phenotype. The predominant forms of cardiac pathology were atrial arrhythmias and conduction disturbances that progress over time. The presented cases discussed in the light of therapeutic strategy, including radiofrequency ablation and antiarrhythmic devices implantation, and the importance of thorough neurological and genetic screening in pediatric patients presenting with complex heart rhythm disorders.


INTRODUCTION
Emery-Dreifuss muscular dystrophy (EDMD) is a group of inherited muscle-joint-cardio syndromes (1). The cardiac involvement in EDMD can be severe and life-threatening (2). Various genetic backgrounds contribute in different ways to the broad spectrum of cardiac manifestations, mainly in the form of conduction disorders, supraventricular arrhythmias and more rarely functional and structural abnormalities of the heart (3).
EDMD is a rare disease. The general prevalence of EDMD is 0.39-1 per 100,000, with significant heterogeneity of estimates; in the pediatric population −0.22 per 100,000 (4)(5)(6). The classic clinical triad includes early joint contractures (elbows, neck, ankles, and spine), slowly progressive muscle weakness/atrophy and cardiac abnormalities, but the clinical course and sequence of symptoms depend on the genotype. The severity of cardiac involvement does not correspond to the progression of muscular weakness (3). In the case of EDMD1, associated with mutations in the EMD gene, the contractures frequently emerge in the first decade of life, become more significant during the growth spurt followed by muscle atrophy and weakness in the second decade of life and usually precede cardiac phenotype. In EDMD2, caused by mutations in the LMNA gene, symptoms varied widely from a mild phenotype with a later onset and slow progression to a severe one with life-threatening complications. Lack of information for the reported cases EDMD3, EDMD4, and EDMD5 linked to the genes encoding proteins of nuclear envelope does not allow to define a unified clinical picture of each of those subtypes (7,8).
Cardiac involvement in EDMD usually follows muscle phenotypes and predominantly presents after the second decade of life (9,10). After the diagnosis is established due to typical muscular manifestation, the cardiac abnormalities are screened in the frame of expected disease phenotype (11). Several cases of early cardiac debut preceding muscle dystrophy have been reported, and even isolated cardiac involvement has been described; still, these cases remain scarce and atypical (10,(12)(13)(14)(15). The cardiac disease commonly manifests from atrial arrhythmias and conduction disturbances. Syncope and sudden cardiac death (SCD), caused by complete heart block or ventricular tachyarrhythmia, can also occur, and single cases of left ventricular non-compaction have been described (16,17). Systolic dysfunction and dilated cardiomyopathy found in a minority of patients and are mainly associated with autosomaldominant disease (AD-EDMD) due to LMNA-mutations (18)(19)(20)(21)(22). Creatine kinase (CK) level can range from normal to 15 times the upper limit, without direct correlation with muscular and cardiac involvement, so in patients with severe cardiac phenotype CK levels could remain normal (23).
The genetic spectrum of EDMD includes mutations in EMD, LMNA, SYNE1, SYNE2, FHL1, TMEM43, SUN1, SUN2, and TTN genes (5,(24)(25)(26)(27)(28). These genes mainly encode for the nuclear envelope proteins, which give rise to the term "nuclear envelopathies" (5,23,29). An exception is FHL1 protein which localizes to the sarcomere and the sarcolemma but may also shuttle between cytosolic and nucleoplasmic fraction, thus becoming a part of the nuclear envelope (30). Mutations in LMNA and EMD are the most common causes of EDMD, and together they account for about 36% of the cases (5). Thus, for EDMD, there are still a number of undetected causative genes (30). An AD-EDMD mainly arises from mutations in LMNA, which contribute to ∼28% of the cases. In LMNAassociated EDMD, males and females are equally affected and left ventricular dysfunction is more commonly observed as well as ventricular arrhythmias and SCD (9,19,20). Several cases of LMNA-associated autosomal-recessive form of EDMD presented in pediatric patients have also been described (31,32). The Xlinked form is usually caused by mutations in the EMD gene (8%) and, rarely, in the FHL1 gene (2%) and predominantly affects males with rare cases of disease manifestation in female carriers (18). Mutations in the EMD gene occur sporadically and rarely, but mutations in the LMNA gene are increasingly identified (23).
Due to the typical presentation during the second-third decades of life and a debut from muscle symptoms, there are only several reports on EDMD presented in children with the isolated cardiac phenotype (13,14,(33)(34)(35). Here we describe five cases of EDMD1 and EDMD2 with a cardiac manifestation in childhood and discuss the need for target screening of neuromuscular phenotypes in children with unexplained atrial dysfunction, conduction abnormalities, and ventricular arrhythmias.

MATERIALS AND METHODS
Patients were examined between 2009 and 2020 in tertiary pediatric cardiac care center -Almazov National Medical Research Centre, St. Petersburg, Russia. All data, including the clinical history, case notes, reports of instrumental methods and surgical protocols were extracted from paper and electronic databases. Clinical examination included physical, 12-lead electrocardiography (ECG), Holter monitoring (HM), transthoracic echocardiography, neurological examination and laboratory tests. Additionally, we performed CMR, electrophysiological study (EPS) and electroneuromyography (ENMG). One patient underwent an endomyocardial biopsy to exclude inflammatory heart disease.

GENETIC ANALYSIS
Target sequencing was performed using a panel of 108 or 172 genes, as previously described (36). For Patients 1 and 2, a targeted panel of 108 cardiomyopathy-associated genes has been initially studied using Haloplex Target Enrichment System (Agilent, Waldbronn, Germany) with an Illumina MiSeq instrument (for gene list, see Supplementary Table 1). For Patients 3-5, a targeted panel of 172 cardiomyopathy-associated genes was studied using the SureSelect Target Enrichment System (Agilent; Waldbronn, Germany) (see Supplementary Table 2). Data processing and variant calling were performed according to GATK BestPractice recommendations (Broad Institute, Cambridge, MA, USA) using hg19 and hg38 human genome references. For variant validation, bidirectional Sanger sequencing was performed using ABI 3500 machine (Applied Biosystems). All novel and previously reported variants of interest with a frequency below 0.01% were classified according to the recommendations of American College of Medical Genetics (37). The described variants were submitted to Gene bank repository under submission numbers SCV001548550 -SCV001548554.

ETHICAL CONSIDERATIONS
The study was performed according to the Declaration of Helsinki, and approval was obtained from the local ethical committee of Almazov National Medical Research Centre. Written informed consents were obtained from the parents of the minor prior to investigation.

Patient 1
The patient first presented with atrial rhythm and rare PACs on the ECG at the age of 14 years. He had no complaints, was a member of a football team and was undergoing ECG annually. No pathology was noticed on echocardiography. At 14.5 y.o. he complained on palpitations. An EPS revealed three types of arrhythmias: non-sustained mAT with heart rate (HR) 114 bpm, non-sustained AF and slow-slow AVRNT with HR 130 bpm. The radiofrequency ablation (RFA) of AVRNT was performed. Metoprolol tartrate was prescribed but canceled due to progressive AVB with pauses up to 3,261 ms. Six months later, AF with irregular AV conduction (2:1-4:1) was registered and followed by to RFA of the inferior vena cava-tricuspid isthmus. During the procedure, extensive low-amplitude areas corresponding to fibrosis fields were observed in the right atrium (Figure 1). CMR showed mild-dilated right atrium, atrial fibrosis and no acute inflammatory (Figure 1). The progression of the sinus and AV nodes dysfunction with bradycardia to 34-43 bpm and pauses to 7,104 ms (Figure 2)

Patient 2
Patient 2 was diagnosed with atrial rhythm with 93 bpm during ECG before being brought to the orphanage at the age of 11 years. His family history is unknown. He has been abusing nicotine and alcohol since the age of 15. On ECG, at the age of 14, atrial and junctional rhythm with PACs and non-sustained mAT were registered. Six months later, he was hospitalized in the emergency unit with palpitations. AT with HR to 200 bpm were registered and aborted by overdriving stimulation. Echocardiography and CMR confirmed right chambers dilatation without myocardial dysfunction. No signs of acute myocardial damage or fibrosis have been found (Figure 1). HM identified sinus bradycardia to 35 bpm, mainly atrial and junctional rhythm, mAT with variable AV conduction (2:1-7:1) and pauses to 4,404 ms (Figure 2). The RFA of mAT was performed (Figure 1)

Patient 4
The patient was first referred to a neurologist at 1.

Patient 5
The patient was first examined by a cardiologist at nine y.o. due to palpitations, but ECG revealed no pathology, HM and echocardiography were not performed. At the age of 11 episodes of mAT, PACs, and transient AVB I were detected by HM. Concurrently, elbow contractures and muscle weakness were noted. ENMG identified a moderate myopathic pattern in the lower limbs. Increase in CK by 6.5 times was detected. Within the next several years, arrhythmias progressed with the sustained episodes of mAT, AFib, deterioration of AV conduction and decrease of HR. Specific changes in the P-wave were detected (Figure 3). No structural abnormalities were detected by echocardiography. Metoprolol tartrate was up-titrated with a good but temporary effect. The genetic study identified an earlier reported variant in the LMNA gene (NM_170707.3):c.305T>C, (p.Leu102Pro). Family history was unremarkable. PM was implanted along with AAT (Propafenone) at the age of 15 years.

DISCUSSION
Supraventricular tachycardia and conduction disorders with sinus and AV node dysfunction are the most common conditions in pediatric cardiology practice. The thorough diagnostic workup and precise nosology definition are important for the optimal therapeutic strategy. The presented cases as well as several earlier reports ( Table 2) provide the evidence that EDMD has to be considered in children presenting with conduction disorders or atrial arrhythmias even without overt signs of muscular dystrophy and joint contractions. The natural cause of EDMD leads to the requirement of PM in 56% of the patients, AF/flutter developed in 61%, atria standstill in 45%, and embolic stroke in 36% of the patients (18). Therefore, identification of EDMDunderlying genetic background is important for following genetic cascade screening first-degree relatives of patients, that is recommended and includes ECG, HM, echocardiography, and CMR (43). The precise molecular mechanisms underlying predominantly atrial and conduction pathology in EDMD are largely unknown.
Most of the genes and genetic loci identified by GWAS linked to AFib represent transcriptional factors and signaling molecules responsible for cardiac development and cardiomyocyte differentiation (44). In this context, the ability of nuclear envelope proteins to modify chromatin organization and transcriptional activity could explain the effect of EMD deficiency on myocardial cell fate (45,46). A deficiency or defects of this proteins could lead to nuclear instability in tissues undergoing mechanical stress, including cardiac and skeletal muscle. Of note, another nuclear envelope protein -nesprin 2 encoded by the SYNE2 gene and reported as a rare cause of EDMD was also identified among GWAS loci in association with AF (47,48). In addition, Shimojima et al. reported that mutant forms of emerin caused abnormalities in nuclear Ca ++ transients, which may further modulate nuclear transcriptional pathways in response to mechanical stress (49). As a result, alterations in cardiomyocyte differentiation gradually led to replacing the normal myocardial cells with fibrotic tissue leading to sinus node dysfunction, ectopic loci, and conduction defects. The question remains why these processes mainly start in the atria, often involve the atrioventricular node and only eventually affect the ventricles. This question is further difficult to address because of the unavailability of atrial tissue for morphological examination and limitation of CMR for atrial imaging because of difficulties in achieving adequate image resolution in thin-walled atria (50). Cardiac structural abnormalities, ventricular arrhythmias and dysfunction are more common in AD-EDMD and are not typical of X-linked EDMD. Of note, in the presented case series, the only patient who demonstrated rare PVCs was the Patient 4 with LMNA-associated EDMD. Together with other reports on LMNA-associated SCD in patients with cardiac and muscle pathologies, this further draws attention for the more thorough follow-up of this group in terms of different arrhythmias (51).
It is commonly accepted that cardiac symptoms of EDMD follows the neuromuscular phenotypes and becomes evident in the second-third decade of life (7,11,23). Cases of isolated cardiac manifestations of EDMD remain rarely reported, especially in pediatric patients (7,33,40). However, such cases, together with our study, emphasize the importance of genetic testing and target search for EDMD in cases of progressive refractory arrhythmias with or without specific neurological and laboratory findings. Three patients with EDMD1 we observed did not have a typical clinical course with early contractures or muscle weakness, but two patients had severe cardiac abnormalities requiring medication and intervention treatment. Patients 4 and 5 with LMNA-genetic variants demonstrated the clinical variability of EDMD ranging from mild disease course and later cardiac debut to early manifestation and fastprogression that led to PM and AAT. Thus, the results of genetic testing helped us draw up a plan of management and further follow-up, including annual cardiological and neurological examinations.
The main therapeutic problem we faced was coexistence of symptomatic high-frequency atrial arrhythmias and conduction disorders, which limited the prescribing of AAT. The use of RFA in EDMD is a debatable issue (52). The literature search resulted in four cases of RFA in EDMD described in details: three with poor outcomes [1 -RFA of AF with recurrence of arrhythmia, leading to systolic dysfunction and heart transplantation (53); 2 -recurrent SVT with PM after AV nodal ablation, accession of ventricular tachycardia and death from irrecoverable asystole (54); 3 -repeated unsuccessful RFA of AFib leading to embolic stroke (42)] and one successful: RFA of AF by employing a threedimensional mapping (52). In our study, RFA was performed in two patients: successful treatment of AVNRT and AF in Patient 1 and unsuccessful RFA of mAT in Patient 2.
Since sudden death in X-linked EDMD is primarily caused by a complete heart block, it can be reliably averted by PM implantation (3). The ICD, cardiac resynchronization therapy, mechanical circulatory support, and heart transplantation remain a rare but potentially applicable strategies in EDMD patients (21,43,55). In the long-term longitudinal study performed by Boriani et al., heart failure requiring transplantation occurred in 6% and asymptomatic LV dysfunction in 17% of patients (18). ICD is more often required for LMNA mutation carriers and should be considered with sustained or non-sustained ventricular tachyarrhythmias, especially in those with LVEF < 45% or in patients with indications for PM implantations (41,56). In line with that, three of five described patients had I class indications for PM, and none had indications for ICD implantation.

CONCLUSION
In conclusion, while being rare cases, heart rhythm disorders can represent the first and for a long time, the only clinical symptom of EDMD even in the pediatric group of patients. Therefore, thorough laboratory and neurological screening along with genetic studies are of importance in each pediatric patient presenting with complex arrhythmias of primary supraventricular origin to exclude EDMD or other neuromuscular disorders. Consideration of EDMD in diagnostic workup can facilitate the optimal strategy and personalized follow-up of this group of patients.

DATA AVAILABILITY STATEMENT
The data presented in the study are deposited in the Gene bank repository, accession numbers SCV001548550-SCV001548554 and publicly available.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Almazov National Medical Research Centre Institute Ethical Review Boards. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

AUTHOR CONTRIBUTIONS
EY, SF, and TL contributed to the conception and design of the study, analysis, and interpretation of the data and drafting of the manuscript. TK contributed to the study concept and research design and wrote the manuscript. AKos, TP, and EV made contributions to the conception and design of the study and revision of the manuscript critically. VL, TV, DL, and LM took part in the analysis and interpretation of the data and have been involved in revising the manuscript critically. AR, PS, YF, AKoz, SZ, NS, and AZ conducted the experiments and performed the analysis and interpretation of the data. All authors have read and agreed to the published version of the manuscript.