Author + information
- Philippe Maury, MD∗ (, )
- Estelle Gandjbakhch, MD, PhD,
- Alban-Elouen Baruteau, MD, PhD,
- Francis Bessière, MD,
- Florence Kyndt, PharmD, PhD,
- Patrice Bouvagnet, MD, PhD,
- Anne Rollin, MD,
- Damien Bonnet, MD, PhD,
- Vincent Probst, MD, PhD and
- Alice Maltret, MD
- ↵∗Division of Cardiology, University Hospital Rangueil, 1 Avenue Jean Pouhes, 31059 Toulouse Cedex 09, France
One of the essential homeobox transcription factors orchestrating cardiac embryologic development is NKX2-5 (1). Mutations in NKX2-5 were initially discovered in inherited atrial septal defect (ASD) associated with atrioventricular block (AVB) (2), but were further found linked to other congenital heart defects and cardiomyopathies. However, it remains difficult to have a clear idea of the cardiac abnormalities caused by NKX2-5 mutations in large populations.
We describe the cardiac phenotype and long-term prognosis of a nationwide cohort of NKX2-5 mutated patients. This cohort included 50 patients (23 males) from 19 families. Cardiac phenotype was diagnosed at a median of 5 years of age (range: birth to 54 years) and median follow-up was 15 years (range: 0 years to 50 years). During the same period, a total of 450 NKX2.5 gene screenings were performed. Thus, approximately 90% of the genetic analyses were negative.
None of the 50 patients had a normal phenotype, although no cardiac or conduction defect was observed in 14 nonmutated other kindreds, suggesting high penetrance. The most frequent congenital heart defect was ostium secundum ASD (38 patients, 76%). Muscular or perimembraneous ventricular septal defect (VSD) was present in 11 patients (22%; 88% with ASD and/or VSD). No other relevant congenital heart disease was found in our population. Structural cardiomyopathy was observed in 14 patients (28%) (7 left ventricular noncompaction, 4 hypertrophic cardiomyopathy, 2 dilated cardiomyopathy [DCM], 1 restrictive/constrictive) with a preserved left ventricular ejection fraction in 44 of 48 patients.
AVB was observed in 45 patients (94%), either at first observation or during follow-up (20 first degree, 12 second degree, and 13 third degree AVB). AVB was not related to ASD, surgery, age, or follow-up duration. Significant sinus bradycardia/pauses were observed in 23%. Electro-physiological investigation (18 patients) revealed suprahisian conduction disturbances (AH and HV intervals 167 ± 48 and 47 ± 22 ms).
Atrial fibrillation (AF)/flutter was documented in 35% and was not related to ASD or surgery but was more frequent in case of cardiomyopathy (58% vs. 28%; p = 0.05). Sustained or nonsustained ventricular arrhythmias (VAs) were observed in 10 patients whereas sudden cardiac death (SCD) happened in 2 patients, and were more common in case of cardiomyopathy (64% vs. 19%; p = 0.005).
Twenty-one patients (42%) have been implanted during the follow-up (16 pacemakers and 5 implantable cardioverter-defibrillators [ICDs]) at 24 ± 14 years. Four of 5 ICD patients presented with cardiomyopathy. After 12 ± 11 years, 43% were found dependent on the device. Four patients have been implanted with a loop recorder, leading to documentation of asymptomatic paroxysmal high-degree AVB/asystole in 3 patients.
Four patients died in follow-up: 1 case of meningitis in a 19-month-old infant, 1 47-year-old man with SCD and DCM at necropsy, 1 SD in a 20-year-old woman with ASD and first-degree AVB (without cardiomyopathy or VA), and 1 right heart endocarditis in an implanted 56-year-old man.
The vast majority of NKX2-5–mutated patients in this population presented with ostium secundum ASD or perimembranous/muscular VSD. NKX2-5 is involved in atrial/ventricular septation and morphogenesis of the septum secundum and probably accounts for the observed cardiac defects (1). Various other congenital heart diseases have been linked to mutations in the NKX2-5 gene (2). The very upstream role of NKX2-5 in embryologic development may advocate various and unpredictable consequences of mutations, but other such defects were lacking in our population, although possibly caused by referral bias.
Various cardiomyopathies were detected in 25% of patients with preserved left ventricular ejection fraction in most, mainly left ventricular noncompaction, but also hypertrophic cardiomyopathy and DCM. The role of NKX2-5 in cellular lineage, myogenesis, trabeculations, hypertrophy, and transcriptional regulation of the ventricular myocardium (1,3), may explain these findings.
NKX2-5 gene is known to be also involved in the development and further maturation of the cardiac conduction system (2,3). Conduction disturbances emerge or progress during post-natal development in NKX2-5 mutated humans or knockout animals due to a progressive alteration of the conduction system (1–3). We also find a major part of atrioventricular conduction disturbances. This study also allows further evaluation of cardiac conduction in NKX2-5–mutated patients: AVB was suprahisian in the vast majority of investigated cases, a significant number of patients will finally present with high degree permanent AVB, whereas manifestations of sinus node disease were observed in 23% of cases.
One of the most striking finding is the high prevalence of SCD in relatives and the existence of VA in 30% of patients. SCD has been reported in NKX2-5 mutations (2), suspected to be related to AVB because initially only reported in patients without pacemaker (2). However, SCDs in implanted patients were noted in some family members. All this suggests that SD might not be secondary only to conduction disturbance but also to VA. NKX2-5 being involved in ventricular architecture, cell differentiation, expression of connexins and channels, disruptions of the ventricular architecture/electrical properties may underlie the substrate for VA.
Apart from ASD, NKX2-5 mutations may lead to a variety of arrhythmia and/or cardiomyopathy prompting for a continuous follow-up during adulthood (Figure 1). Indications for pacemaker/ICD implantation are not defined to date but should probably be large.
Please note: Dr. Baruteau is supported by a research grant from the European Society of Cardiology. Dr. Gandjbakhch has received consultant fees from Bayer, Livanova, Medtronic, and Boston Scientific. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Drs. Maury and Gandjbakhch contributed equally to this paper.
The authors thank Dr. F. Heitz, N. Combes, N. Blot-Souletie, M. Gouton, C. Marchal, F. Wiart, C. Thauvin, F. Marcon, C. Selton-Suty, S. Reuter, A. Khemache, M. Levy, RG. Huguet, G. Duthoit, F. Hidden-Lucet, P. Acar, Y. Dulac, S. Hascoet, P. Chevalier, P. Charron, V. Fressart and F. Salerno, for their help in genetics and clinical follow-up of the patients.
This work is written on behalf of the Réseau Francophone de Rythmologie pédiatrique et congénital.
- American College of Cardiology Foundation