Abstract
Background
Patient prognosis in type 1 myotonic dystrophy (DM1) is very poor. Annual 24‐hour holter ECG monitoring is recommended but its relevance is debated. Main objective was to determine whether holter ECG parameters could predict global death in DM1 patients and secondarily to assess whether they could predict cardiovascular events and sudden cardiac death, to compare DM1 patients and healthy controls, and to assess their evolution in DM1 over a 5‐year period.
Methods
This retrospective study included genetically confirmed DM1. Primary endpoint was global death. Secondary endpoints were labeled “sudden cardiac death” which was a composite of sudden cardiac death, aborted sudden cardiac death, implantable cardioverter defibrillator therapy, sustained ventricular tachycardia, atrioventricular block grade 3, pause >3 s; and “cardiovascular events” which was a composite of all‐cause mortality, pacemaker or cardioverter defibrillator implantation, sustained ventricular tachycardia, supraventricular tachycardia, hospitalization for acute cardiac cause and heart failure.
Results
Forty‐seven patients (22 women, 40 ± 13 years old) were included. Three (7%) DM1 patients died, 9 (19%) experienced "sudden cardiac death" endpoint and 21 (45%) experienced "cardiovascular event" endpoint during mean follow‐up of 95 ± 22 months. None of holter ECG parameters were discriminant to predict death or secondary endpoints. Compared to healthy controls, DM1 patients had higher SDNN and LF/HF ratio. Finally, heart rate variability parameters remained stable over a mean interval of 61 ± 15 months excepting pNN50 which decreased significantly.
Conclusion
Results suggest that annually‐repeated holter ECG in DM1 is not useful for stratifying risk of sudden death and cardiovascular outcomes.
Keywords: ambulatory electrocardiography, autonomic nervous system, heart rate variability, Holter monitoring, myopathy, type 1 myotonic dystrophy
1. BACKGROUND
Myotonic dystrophy type 1 (DM1) is the most common genetic disease of the adult with an estimated incidence of 1 in 10,000 in Europe (Lund et al., 2014). Its prognosis is closely related to cardiac and respiratory disorders (Pelargonio, DelloRusso, Sanna, Martino, & Bellocci, 2002). Conduction disturbance is the most frequent cardiac involvement with 13% of cumulative incidence at 12 years, followed by ventricular tachycardia with 2.5% cumulative incidence (Wahbi et al., 2017).
Despite the fact that 30% of deaths are caused by arrhythmias, most of the studies seem to agree on the difficulty of predicting cardiac disorders in DM1 patients (Pelargonio et al., 2002). Several predictors of cardiovascular events have been described but DM1 patient follow‐up is not standardized (Bhakta, Groh, Shen, Pascuzzi, & Groh, 2010; Garcia, Labarre, et al., 2017; Garcia, Rehman, et al., 2017; Lau, Sy, Corbett, & Kritharides, 2015; Wahbi et al., 2017). Annual standard electrocardiogram (ECG) is admitted but annual 24‐hour Holter ECG is debated.
The aim of the study was to assess the usefulness of 24‐hour Holter ECG in DM1 patients and especially its prognostic value toward cardiovascular events. We also compared baseline 24‐hour Holter ECG parameters between DM1 and healthy controls and finally we observed the evolution of heart rate variability (HRV) over time.
2. METHODS
2.1. Patients
This observational retrospective study took place at the University Center Hospital of Poitiers, France, according to the ethical principles of the Declaration of Helsinki and the current guidelines for good clinical practice. The local ethics committee approved the study and all subjects gave written informed consent. All adults suffering from genetically confirmed DM1 with at least one 24‐hour Holter ECG carried out between 2007 and 2013 were screened for eligibility.
Exclusion criteria were defined to optimize ECG monitoring analysis. Patients with pacemaker or cardioverter defibrillator implantation, atrial fibrillation (AF), neurodegenerative diseases, cardiomyopathy, diabetes, or antiarrhythmic drugs were excluded. Socio‐demographic and clinical characteristics, ECG, and the number of Cytosine‐Thymine‐Guanine (CTG) repetitions were collected at baseline.
2.2. 24‐hour ambulatory ECG monitoring
Livanova Spiderview™ holter recorders (Milan, Italy) with 3‐channel recording were used for 24‐hour ambulatory Holter ECG monitoring in each patient. Each beat was automatically classified and labeled by Livanova Synescope™ software using the template‐matching technique. Then, a single senior cardiologist retrospectively read over all the recordings and manually classified ventricular beats, nonventricular beats, and artifacts. Nonsustained ventricular tachycardia, sustained ventricular tachycardia, AF, atrial flutter, sinus node dysfunction, 2nddegree atrioventricular block, 3rddegree atrioventricular block, pauses >3 s, and number of ventricular extrabeats were reported.
Heart rate variability was computed over a 24‐hr interval. Periods with AF or ventricular extrabeats were excluded from HRV analysis. Time domain method was explored with pNN50 (%), SDNN (ms), and rMSSD (ms). pNN50 is defined as the number of pairs of adjacent NN intervals differing by more than 50 ms in the entire recording divided by the total number of all NN intervals. SDNN is the standard deviation of all NN intervals. RMSSD is defined as the square root of the mean of the sum of the squares of differences between adjacent NN Intervals. HRV was also assessed with spectral analysis. Ultra‐low frequency (ms2) was determined by frequency band <0.04 Hz, low frequency (ms2) between 0.04 and 0.15 Hz, and high frequency (ms2) between 0.15 and 0.4 Hz allowing for calculation of power in low‐frequency range (LF)/power in high‐frequency range (HF) ratio (Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology, 1996).
2.3. Twenty‐four‐hour Holter ECG parameter comparison and HRV evolution over time
The first step was the comparison of 24‐hour holter ECG parameters between DM1 and healthy controls. Healthy controls were recruited among the hospital staff and were matched to case subjects according to sex and age (±3 years). Controls were asymptomatic and free of any medical history or medication. Then, in order to assess intra‐individual HRV evolution over time, a second 24‐hour Holter ECG, carried out at least 5 years after the first one, was re‐analyzed.
2.4. Follow‐up
Patients were followed up from the time of their inclusion, that is to say from the first 24‐hour Holter ECG, until January 1, 2017. The primary endpoint was death. The secondary endpoints were labeled “sudden cardiac death” which was a composite of sudden cardiac death, aborted sudden cardiac death, appropriate implantable cardioverter defibrillator therapy, sustained ventricular tachycardia, atrioventricular block type 3, pause >3 s; and “cardiovascular events” which was a composite of all‐cause mortality, pacemaker or cardioverter defibrillator implantation, sustained ventricular tachycardia, supraventricular tachycardia, hospitalization for acute cardiac cause and heart failure. Data were collected using hospital archives and the hospital software archiving medical data. All the hospitalizations and consultation reports were examined. If necessary, general practitioners and referring cardiologists were contacted to provide previously missing information.
2.5. Statistical analysis
Continuous data were described by mean and standard deviation, and groups compared by Student's t test or Mann‐Whitney U test if necessary. Categorical data were described by numbers and corresponding percentages, and groups compared by chi2 test or by Fisher's test if necessary. Correlation between CTG repetition number, age, and HRV parameters was assessed with Spearman's test. To evaluate intra‐individual HRV evolution over time, paired t test was used.
Statistical analysis was performed with SPSS 22 (SPSS, Inc., Chicago, IL, USA). Two‐sided p‐values of <0.05 were considered statistically significant.
3. RESULTS
3.1. Study population
Of 55 eligible DM1 patients, eight were not included: one because of uninterpretable recording, five because of pre‐existing pacemaker or implantable cardioverter defibrillator, one because of diabetes, and one because of antiarrhythmic drug (Figure 1). Forty‐seven DM1 patients (22 women, 40 ± 13 years old) were included in the study. Four (8.5%) patients had first‐degree family history of sudden death. Mean number of CTG repetitions was 377 ± 314. Regarding cardiovascular risk factor exposure, five (10.6%) patients had hypertension, five (10.6%) patients had dyslipidemia, and nine (19.1%) patients were smokers. On 12‐lead ECG, mean PR interval and mean QRS duration were 217 ± 44 and 97 ± 21 ms, respectively. Mean value of N‐terminal pro brain natriuretic peptide, left ventricular ejection fraction, and left atrial volume were 81 ± 58 pg/ml, 61% ± 8.7%, and 24 ± 7.4 ml/m2, respectively. Data are presented in Table 1.
Figure 1.
Flow chart. DM1, myotonic dystrophy type 1; ICD, implantable cardioverter defibrillator; Clinical follow‐up was from first 24‐hour Holter ECG until January 1, 2017
Table 1.
Baseline characteristics of DM1 patients and controls
|
DM1 n = 47 |
Controls n = 47 |
P‐value | |
|---|---|---|---|
| Age, years | 40 ± 13.6 | 40 ± 13.4 | – |
| Women | 22 (46.8%) | 22 (46.8%) | – |
| SD family history | 4 (8.5%) | 0 | <0.001 |
| CTG repeats | 377 ± 314 | – | – |
| Hypertension | 5 (10.6%) | 4 (8.5%) | 0.73 |
| Dyslipidemia | 5 (10.6%) | 4 (8.5%) | 0.73 |
| Current smoking | 9 (19.1%) | 8 (17%) | 0.79 |
| Cardiomyopathy | 0 | 0 | – |
| Nocturnal ventilator support | 17 (36.2%) | 0 | <0.001 |
| PR interval, ms | 217 ± 44 | 175 ± 17 | <0.001 |
| QRS duration, ms | 97 ± 21.5 | 87 ± 13.3 | 0.01 |
| Left atrial volume, ml/m2 | 24 ± 7.4 | – | – |
| Left ventricular ejection fraction, % | 61 ± 8.7 | – | – |
| NT‐proBNP, pg/ml | 81 ± 58 | – | – |
Data are expressed as mean ± SD or number (%).
CTG: cytosine thiamine guanidine; DM1: myotonic dystrophy type 1; NT‐proBNP: N‐terminal pro brain natriuretic peptide; SD: sudden death.
3.2. Baseline 24‐hour Holter ECG comparison between DM1 and control group
Arrhythmias and conduction disturbance parameters were not statistically different between DM1 group and healthy controls, as mean heart rate. Regarding frequency domain HRV analysis, DM1 patients had higher LF/HF ratio values compared to controls (3.5 ± 2.2 vs. 1.6 ± 0.5, respectively; p < 0.001). With the time domain method, SDNN was significantly higher in DM1 patients compared to healthy controls (132 ± 42 ms vs. 119 ± 19 ms, respectively; p = 0.04) (Table 2). None of the HRV parameters were correlated with the number of CTG repetitions: ρ = 0.08 and p = 0.6 for pNN50; ρ = 0.08 and p = 0.59 for SDNN; ρ = 210.04 and p = 0.8 for LF; ρ = 0.05 and p = 0.76 for HF; ρ = −0.05 and p = 0.73 for LF/HF. Age was significantly correlated with pNN50 and LF (ρ = −0.24; p = 0.02 and ρ = −0.21; p = 0.04 respectively) but not correlated with SDNN, HF and LF/HF ratio (ρ = −0.15; p = 0.16; ρ = −0.13; p = 0.22 and ρ = −0.15; p = 0.16, respectively).
Table 2.
Baseline 24‐hour Holter ECG parameter comparison between DM1 and healthy controls
| DM1 (n = 47) | Controls (n = 47) | p‐value | |
|---|---|---|---|
| Tachyarrhythmias | |||
| Atrial fibrillation | 2 (4.3%) | 1 (2.1%) | 0.56 |
| Atrial flutter | 0 | 0 | – |
| Non‐sustained ventricular tachycardia | 0 | 0 | – |
| Sustained ventricular tachycardia | 0 | 0 | – |
| Ventricular extrabeats | 275 ± 1669 | 14 ± 43 | 0.29 |
| Conduction disturbances | |||
| Node sinus dysfunction | 0 | 1 (2.1%) | 0.31 |
| Second or third degree AV block | 0 | 0 | – |
| Pause | 1 (2.1%) | 0 | 0.31 |
| Heart rate variability | |||
| Time domain methods | |||
| Mean NN interval, ms | 895 ± 123 | 886 ± 77 | 0.99 |
| pNN50, % | 16 ± 16.6 | 12 ± 5.2 | 0.11 |
| SDNN, ms | 132 ± 42 | 119 ± 19 | 0.04 |
| rMSSD, ms | 33 ± 5.2 | 34 ± 3.6 | 0.07 |
| Frequency domain methods | |||
| LF, ms2 | 1,490 ± 2,186 | 1,021 ± 242 | 0.15 |
| HF, ms2 | 954 ± 3,186 | 669 ± 183 | 0.54 |
| LF/HF | 3.5 ± 2.2 | 1.6 ± 0.5 | <0.001 |
AV: atrioventricular; DM1: myotonic dystrophy type 1; HF: high frequency; LF: low frequency.
Data are expressed as mean ± SD and number (%).
3.3. Evolution of heart rate variability and other ECG parameters over time in DM1 patients
Among the 47 patients included in the DM1 population, a second 24‐hour holter carried out at least 5 years after the first could be re‐analyzed in 32 patients. Fifteen patients could not be included in the analysis because of absence of recording, permanent AF or presence of pacemaker. The mean time interval between first and second 24‐hour holter ECG was 61 ± 15 months. None of the HRV parameters, whether analyzed with spectral or by time domain method, changed significantly over time (NN interval: 895 ± 123 ms vs. 898 ± 181 ms; pNN50%: 14% ± 12.6% vs. 10% ± 11.7%, p = 0.07; SDNN: 130 ± 40 ms vs. 136 ± 61 ms, p = 0.5; rMSSD: 33 ± 5.2 vs. 31 ± 4.7, p = 0.67; LF: 1,232 ± 1,076 ms2 vs. 1,009 ± 1,522 ms2, p = 0.36; HF: 427 ± 449 ms2 and 362 ± 646 ms2, p = 0.5; LF/HF: 3.9 ± 2.3 and 3.7 ± 1.8, p = 0.49 for first and second Holter, respectively).
Regarding other Holter ECG parameters, the number of ventricular extrabeats did not change significantly over time (7 ± 18 vs. 62 ± 188, p = 0.11 for first and second holter, respectively). There was no difference between the two groups regarding nonsustained ventricular tachycardia, AF, pause >3 s, type 3 atrioventricular block (non‐sustained ventricular tachycardia: 0 vs. 0, AF: 0 vs. 0, pause >3 s: 0 (0%) vs. 3 (9%), p = 0.24; type 3 atrioventricular block: 0 (0%) vs. 1 (3%), p = 0.5 for first and second Holter, respectively).
3.4. Death and cardiovascular events
Median follow‐up duration was 95 ± 22 months. During follow‐up, 3 (7%) patients died, nine (19%) experienced the "sudden cardiac death" endpoint (sudden death n = 3, ventricular tachycardia n = 3, pause >3 s n = 2, atrioventricular block type 3 n = 1) and 21 (45%) experienced the "cardiovascular event" endpoint (sudden death n = 3, pacemaker n = 8, ventricular tachycardia n = 3, atrial fibrillation n = 4, acute heart failure n = 3). None of the 24‐hour Holter ECG parameters were associated with death or with occurrence of sudden cardiac death and cardiovascular events (Table 3).
Table 3.
Baseline 24‐hour Holter ECG parameter comparison between DM1 patients alive and dead and between DM1 patients who reached the secondary endpoints and those who did not during follow‐up
| All causes of death |
Combined endpoint "Cardiovascular events" |
Combined endpoint "Sudden Cardiac Death" |
|||||||
|---|---|---|---|---|---|---|---|---|---|
| Yes | No | p‐value | Yes | No | p‐value | Yes | No | p‐value | |
| n = 3 (7%) | n = 44 (93%) | n = 21 (45%) | n = 26 (55%) | n = 9 (19%) | n = 38 (81%) | ||||
| Tachyarrhythmias | |||||||||
| Atrial fibrillation | 0 | 2 (4.3%) | NS | 1 (4.7%) | 1 (3.9%) | NS | 0 | 2 (5.2%) | NS |
| Atrial flutter | 0 | 0 | – | 0 | 0 | – | 0 | 0 | – |
| Non‐sustained ventricular tachycardia | 0 | 0 | – | 0 | 0 | – | 0 | 0 | – |
| Sustained ventricular tachycardia | 0 | 0 | – | 0 | 0 | – | 0 | 0 | – |
| Conduction disturbances | |||||||||
| Node sinus dysfunction | 0 | 0 | – | 0 | 0 | – | 0 | 0 | – |
| Second or third degree AV block | 0 | 0 | – | 0 | 0 | – | 0 | 0 | – |
| Pause | 0 | 1 (2.1%) | NS | 0 | 1 (3.9%) | NS | 1 (11.1%) | 0 | NS |
| Time domain methods | |||||||||
| Mean NN interval, ms | 873 ± 69 | 897 ± 126 | 0.46 | 914 ± 111 | 880 ± 132 | 0.83 | 907 ± 93 | 892 ± 130 | 0.33 |
| pNN50, % | 4 ± 4 | 17 ± 17 | 0.19 | 18 ± 20 | 15 ± 13 | 0.6 | 11 ± 11 | 17 ± 18 | 0.35 |
| SDNN, ms | 111 ± 23 | 134 ± 43 | 0.37 | 133 ± 46 | 131 ± 38 | 0.9 | 129 ± 34 | 133 ± 43 | 0.77 |
| rMSSD, ms | 35 ± 4.6 | 33 ± 5.3 | 0.46 | 33 ± 5.4 | 33 ± 5.1 | 0.8 | 35 ± 6.0 | 33 ± 5.0 | 0.33 |
| Frequency domain methods | |||||||||
| LF, ms2 | 532 ± 192 | 1555 ± 2,245 | 0.44 | 1821 ± 3,147 | 1,222 ± 837 | 0.36 | 891 ± 767 | 1632 ± 2,389 | 0.37 |
| HF, ms2 | 162 ± 65 | 1,009 ± 3,288 | 0.66 | 1,458 ± 4,712 | 547 ± 727 | 0.33 | 295 ± 305 | 1,111 ± 3,531 | 0.5 |
| LF/HF | 3.8 ± 2.1 | 3.5 ± 2.2 | 0.81 | 3.3 ± 2.1 | 3.7 ± 2.3 | 0.51 | 3.8 ± 2.1 | 3.5 ± 2.2 | 0.7 |
Data are expressed as mean ± SD and number (%).
AV: atrioventricular; HF: high frequency; LF: low frequency.
4. DISCUSSION
This study aimed to assess 24‐hour holter ECG usefulness in DM1 patient follow‐up. Two ascertainments can be made: 1/None of the 24‐hour holter ECG parameters, even HRV parameters, were prognostic of long‐term sudden cardiac death or cardiovascular events; 2/Compared to controls, DM1 patients had higher SDNN and LF/HF ratio.
Pathophysiological mechanisms of arrhythmias and conduction disturbances in DM1 are related to ribonucleic acid toxicity resulting in myocyte hypertrophy, fatty infiltration, interstitial fibrosis, myofibrillar degeneration, and prominent I bands (Motta, Guilleminault, Billingham, Barry, & Mason, 1979; Nguyen, Wolfe, Holmes, & Edwards, 1988). After 12 years of follow‐up, these abnormalities are responsible for a 19.3% cumulative incidence rate of major conduction defects and for a 2.3% cumulative incidence rate of ventricular arrhythmias (Wahbi et al., 2017). Despite more and more studies in the literature, cardiac management of asymptomatic DM1 is not codified and identification of tools enabling to predict these complications is crucial (Lau et al., 2015). Several parameters have been described as predictors of cardiovascular events: age, left bundle branch block, syncope, atrial fibrillation, longitudinal strain, left ventricular ejection fraction (Bhakta et al., 2010; Garcia, Labarre, et al., 2017; Garcia, Rehman, et al., 2017; Lau et al., 2015; Wahbi et al., 2017). Nevertheless, the relevancy of 24‐hour holter ECG, which is currently recommended, has never been assessed (Lau et al., 2015). That is the reason why we carried out a stepwise approach by 1/Comparing DM1 24‐hour holter ECG parameters with healthy matched controls; 2/Assessing the evolution over time of these parameters; and finally 3/Evaluating the prognostic value of 24‐hour holter ECG parameters toward global death, sudden cardiac death, and cardiovascular events.
Generally speaking, the most striking difference compared to other studies is that we included only DM1 patients in whom HRV was interpretable, that is to say without atrial fibrillation, ventricular tachycardia, pacemaker, or implantable cardioverter defibrillator history. While Brembilla‐Perrot found a mean SDNN of 128 ± 43 ms, the other HRV parameters were not studied (Brembilla‐Perrot, Luporsi, Louis, & Kaminsky, 2011). We present very similar values of SDNN but, possibly due to a lack of power, did not find an association with death. Compared to Hardin's study, we had comparable frequency analysis data, even though the recording duration was different (Hardin, Lowe, Bhakta, & Groh, 2003). Combined basic holter ECG abnormalities were not predictive of death in Brembilla‐Perrot's study, whereas abnormal ECG was predictive, but no specific analysis regarding sudden cardiac death was carried out (Brembilla‐Perrot et al., 2011).
Two points were particularly interesting. The first one is that standard deviation was much higher in DM1 patients compared to controls. This finding was also present in Hardin's work (Hardin et al., 2003). The second new feature was that DM1 patients' LF/HF ratio was higher compared to healthy controls and that ratio did not change over time in DM1 patients. Previous studies comparing spectral analysis in healthy controls and DM1 were carried out on very small samples, used different methods and had contradictory results (DiLeo et al., 2004; Fregonezi et al., 2012; Inoue et al., 1995; Rakocević‐Stojanović et al., 2007).The largest study among them also found autonomic dysfunction despite heterogeneous results (DiLeo et al., 2004). It has been established that HF reflects vagal tone that LF correlates to sympathetic or both sympathetic and vagal tone (Bigger et al., 1992; Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology, 1996). We believe that this increased sympathetic modulation might be explained by hypoxemia, heart failure, or vasoplegia. To support our findings, assay of plasma catecholamine could have been interesting. DM1 patients are at risk of arrhythmia and conduction disturbance, to which autonomic dysfunction may be a major contributor.
Correlations of HRV parameters with age and CTG repetitions were previously assessed. Hardin et al. (2003) found a correlation between the number of CTG repetitions, age, and HRV parameters. In the present work, we did not find any relationship with CTG repetitions, and age was correlated with pNN50 and LF. These discrepancies can be explained by different reasons. We had a small number of DM1 patients. Our population was more selected than in Hardin's study, and finally, DM1 is an incomplete penetrance with variable expression.
The main limitation to this work was that the number of patients was restricted. Noninclusion of patients presenting diabetes, AF, or cardiomyopathy was essential to analyze HRV but our cohort might not be representative of the whole DM1 population. Nevertheless, DM1 is a rare disease and this work is the first to propose a stepwise approach to assess the usefulness of 24‐hour holter ECG in this disease.
5. CONCLUSION
Stratifying risk of sudden death is one of the major challenges in management of patients suffering from DM1. Wide use of 24‐hour holter ECG monitoring does not appear to be accurate as a means of identifying predictive factors of sudden cardiac death and cardiovascular outcomes, perhaps because of the slow evolution of cardiac involvement. However, the HRV impairments and sympatho‐vagal imbalance observed in our study could in future studies be a key to understanding the mechanisms of sudden death and vulnerability to ventricular arrhythmias.
CONFLICT OF INTEREST
None.
ACKNOWLEDGMENTS
We thank Jeffrey Arsham for rereading this manuscript.
Gamet A, Degand B, Le Gal F, et al. Twenty‐four‐hour ambulatory ECG monitoring relevancy in myotonic dystrophy type 1 follow‐up: Prognostic value and heart rate variability evolution. Ann Noninvasive Electrocardiol. 2019;24:e12587 10.1111/anec.12587
REFERENCES
- Bhakta, D. , Groh, M. R. , Shen, C. , Pascuzzi, R. M. , & Groh, W. J. (2010). Increased mortality with left ventricular systolic dysfunction and heart failure in adults with myotonic dystrophy type 1. American Heart Journal, 160, 1137–1141. 10.1016/j.ahj.2010.07.032 [DOI] [PubMed] [Google Scholar]
- Bigger, J. T. Jr , Fleiss, J. L. , Steinman, R. C. , Rolnitzky, L. M. , Kleiger, R. E. , & Rottman, J. N. (1992). Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation, 85, 164–171. 10.1161/01.CIR.85.1.164 [DOI] [PubMed] [Google Scholar]
- Brembilla‐Perrot, B. , Luporsi, J. D. , Louis, S. , & Kaminsky, P. (2011). Long‐term follow‐up of patients with myotonic dystrophy: An electrocardiogram every year is not necessary. Europace, 13, 251–257. 10.1093/europace/euq423 [DOI] [PubMed] [Google Scholar]
- DiLeo, R. , Rodolico, C. , DeGregorio, C. , Recupero, A. , Coglitore, S. , Annesi, G. , … Vita, G. (2004). Cardiovascular autonomic control in myotonic dystrophy type 1: A correlative study with clinical and genetic data. Neuromuscular Disorders, 14, 136–141. 10.1016/j.nmd.2003.11.002 [DOI] [PubMed] [Google Scholar]
- Fregonezi, G. , Araújo, T. , Dourado Junior, M. E. , Ferezini, J. , Silva, E. , & Resqueti, V. (2012). Heart rate variability in myotonic dystrophy type 1 patients. Arquivos Brasileiros De Cardiologia, 98, 353–360. [DOI] [PubMed] [Google Scholar]
- Garcia, R. , Labarre, Q. , Degand, B. , Ingrand, P. , LeGal, F. , Bonnet, B. , … Christiaens, L. (2017). Apical left ventricular myocardial dysfunction is an early feature of cardiac involvement in myotonic dystrophy type 1. Echocardiography, 34, 184–190. 10.1111/echo.13426 [DOI] [PubMed] [Google Scholar]
- Garcia, R. , Rehman, M. , Goujeau, C. , Degand, B. , LeGal, F. , Stordeur, B. , … Bouleti, C. (2017). Left ventricular longitudinal strain impairment predicts cardiovascular events in asymptomatic type 1 myotonic dystrophy. International Journal of Cardiology, 19, 32086–32087. 10.1016/j.ijcard.2017.05.061 [DOI] [PubMed] [Google Scholar]
- Hardin, B. A. , Lowe, M. R. , Bhakta, D. , & Groh, W. J. (2003). Heart rate variability declines with increasing age and CTG repeat length in patients with myotonic dystrophy type 1. Annals of Noninvasive Electrocardiology, 8, 227–232. 10.1046/j.1542-474x.2003.08310.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Inoue, K. , Ogata, H. , Matsui, M. , Hayano, J. , Miyake, S. , Kumashiro, M. , … Akatsu, J. (1995). Assessment of autonomic function in myotonic dystrophy by spectral analysis of heart‐rate variability. Journal of the Autonomic Nervous System, 55, 131–134. 10.1016/0165-1838(95)00040-5 [DOI] [PubMed] [Google Scholar]
- Lau, J. K. , Sy, R. W. , Corbett, A. , & Kritharides, L. (2015). Myotonic dystrophy and the heart: A systematic review of evaluation and management. International Journal of Cardiology, 184, 600–608. 10.1016/j.ijcard.2015.03.069 [DOI] [PubMed] [Google Scholar]
- Lund, M. , Diaz, L. J. , Ranthe, M. F. , Petri, H. , Duno, M. , Juncker, I. , … Melbye, M. (2014). Cardiac involvement in myotonic dystrophy: A nationwide cohort study. European Heart Journal, 35(32), 2158–2164. 10.1093/eurheartj/ehu157 [DOI] [PubMed] [Google Scholar]
- Motta, J. , Guilleminault, C. , Billingham, M. , Barry, W. , & Mason, J. (1979). Cardiac abnormalities in myotonic dystrophy. Electrophysiologic and histopathologic studies. American Journal of Medicine, 67, 467–473. 10.1016/0002-9343(79)90795-2 [DOI] [PubMed] [Google Scholar]
- Nguyen, H. H. , Wolfe, J. T. 3rd , Holmes, D. R. Jr , & Edwards, W. D. (1988). Pathology of the cardiac conduction system in myotonic dystrophy: A study of 12 cases. Journal of the American College of Cardiology, 11, 662–671. 10.1016/0735-1097(88)91547-1 [DOI] [PubMed] [Google Scholar]
- Pelargonio, G. , DelloRusso, A. , Sanna, T. , De Martino, G. , & Bellocci, F. (2002). Myotonic dystrophy and the heart. Heart, 88, 665–670. 10.1136/heart.88.6.665 10.1136/heart.88.6.665 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rakocević‐Stojanović, V. , Milovanović, B. , Ivić, N. , Ille, T. , Marjanović, I. , Stević, Z. , … Lavrnić, D. (2007). Cardiac autonomic nervous system in patients with myotonic dystrophy type 1. Acta Myologica, 26, 112–114. [PMC free article] [PubMed] [Google Scholar]
- Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology (1996). Heart rate variability: Standards of measurement, physiological interpretation and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Circulation, 93, 1043–1065. 10.1161/01.cir.93.5.1043 [DOI] [PubMed] [Google Scholar]
- Wahbi, K. , Babuty, D. , Probst, V. , Wissocque, L. , Labombarda, F. , Porcher, R. , … Duboc, D. (2017). Incidence and predictors of sudden death, major conduction defects and sustained ventricular tachyarrhythmias in 1388 patients with myotonic dystrophy type 1. European Heart Journal, 38(10), 751–758. 10.1093/eurheartj/ehw569 [DOI] [PubMed] [Google Scholar]