Evaluation of systolic and diastolic cardiac functions and heart rate variability in patients with juvenile myoclonic epilepsy

Summary

Aim:

To explore if patients with juvenile myoclonic epilepsy (JME) have dif­ferences in heart rate variability (HRV) as­sociated with risk of ar­rhythmias and in systolic and diastolic functions.

Patients and methods:

This single-centre prospective study included 50 patients with JME fol­lowed up at the epilepsy outpatient clinic within the Antalya Train­­ing and Research Hospital’s Neurology Department (34 women, mean age 26 ± 7.58 years) and 45 healthy controls (30 women, mean age 26.71 ± 5.14 years). Two patients were excluded since they had arterial hypertension, one patient was excluded due to diabetes mel­litus and one patient was excluded due to rheumatic mitral stenosis. Final­ly, 46 patients were included in the study. All patients and controls were evaluated by conventional echocardiography and tis­sue Doppler imag­­ing for systolic and diastolic functions (e. g. left ventricular ejection fraction, left ventricle diameters and volumes, deceleration time, pulmonary forward flows) and perform­­ing 24-h Holter monitor­­ing to explore time domain (e. g. standard deviation of the normal-to-normal interval, the square root of the mean squared dif­ferences of succes­sive normal-to-normal intervals) and frequency domain parameters of HRV.

Results:

There were no significant dif­ferences between the Holter parameters of the two groups with respect to HRV. Echocardiographic investigation did not reveal any significant dif­ferences except for the ratio of pulmonary venous systolic and diastolic (PVS/ PVD) forward flows to one another (p = 0.008).

Conclusion:

In this study, we did not find anyth­­ing else about cardiac involvement other than increased ratio of PVS/ PVD and we did not find autonomic dysfunction in patients with JME. This may be due to good seizure control.

Key words:

heart rate variability – juvenile myoclonic epilepsy – cardiac functions – echocardiography

The authors declare they have no potential conflicts of interest concerning drugs, products, or services used in the study.

The Editorial Board declares that the manu­script met the ICMJE “uniform requirements” for biomedical papers.

Chinese summary - 摘要

青少年肌阵挛性癫痫患者收缩期和舒张期心功能及心率变异性的评估

目标：

探讨青少年肌阵挛性癫痫（JME）患者心率变异性（HRV）与心律失常风险以及收缩和舒张功能的差异。

患者和方法：

这项单中心前瞻性研究纳入了50名JME患者，随后在安塔利亚培训和研究医院神经内科的癫痫门诊纳入（34名女性，平均年龄26±7.58岁）和45名健康对照（30名女性，平均年龄26.71± 5.14岁）。两名患者因动脉高血压被排除在外，一名患者因糖尿病而被排除，另一名患者由于风湿性二尖瓣狭窄而被排除。最后，46名患者被纳入研究。通过常规超声心动图和组织多普勒成像评估所有患者和对照的收缩和舒张功能（例如左心室射血分数，左心室直径和体积，减速时间，肺前向流动）并进行24小时动态心电图监测以探索时域（例如，正常到正常间隔的标准偏差，连续正常到正常间隔的均方差的平方根）和HRV的频域参数。

结果：

两组的Holter参数与HRV无显著差异。除了肺静脉收缩压和舒张压（PVS / PVD）相互之间的流量比（p = 0.008），超声心动图检查未发现任何显著差异。

结论：

在这项研究中，除了增加PVS / PVD的比例之外，我们没有发现任何关于心脏牵涉的信息，我们在JME患者中没有发现自主神经功能障碍。 这可能是由于良好的癫痫发作控制。

关键词：

心率变异性 - 青少年肌阵挛性癫痫 - 心脏功能 - 超声心动图

Introduction

Epilepsy often af­fects the autonomic fun­ct­­ions dur­­ing ictal, inter-ictal and postictalperiods. Epileptic discharges are inculpated for impair­­ing or alter­­ing the central auto­nomic pathway and normal autonomic cardiac functions in patients hav­­ing seizures. Sudden unexpected death in epilepsy (SUDEP) is the lead­­ing cause of death in patients with epilepsy and accounts for about 17% of these deaths [1]. However, the mechanisms involved in SUDEP are not well understood; fatal cardiac ar­rhythmias caused by the activation of the autonomic nervous system (ANS) are also inculpated. Measurement of heart rate variability (HRV) is one of the ways to measure the ef­fects of ANS on the heart. The intervals between the heart beats in the sinus rhythm are dif­ferent in healthy individuals, which is a normal physiological proces­s [2]. HRV refers to the mathematical formulation of the fluctuations in heart rate around the average heart rate within a certain time [2]. Physical and metabolic factors lead to variations in the heart rate as­sociated with autonomic tonus. HRV seems to be very promis­­ing for the investigation of cardiovascular response to the variations in the autonomic tonus [2,3].Increased activation of the sympathetic system and suppres­sion of the para­sym­pathetic system leads to a decline in the HRV [2]. HRV is evaluated in two ways: time and frequency analysis. The intervals between the normal pulses in the 24-h electrocardiography (ECG) records are analyzed dur­­ing the time domain measure­ment, while the heart rate signals are grouped accord­­ing to frequencies and intensities and information is obtained about the amount of heart rate variation on the basis of the periodical cardiac oscil­lations at dif­ferent frequencies dur­­ing the frequency domain spectral analysis [2,4]. In our study, we investigated autonomic dysfunction through time domain and frequency domain HRV analysis in patients dia­gnosed with juvenile myoclonic epilepsy (JME) and healthy controls. Recent studies found that epilepsy patients had cardiac systolic and diastolic dysfunctions, while the ef­fect of ANS on the heart are inculpated for such abnormalities [5].

Increased sympathetic cardiac stimulation may lead to cardiac ar­rhythmias but also repetitive sympathetic stimulation may cause structural damage in the heart, which means that there is a high predisposition to ar­rhythmia and ischaemia. Myocardial fibrosis may also lead to left ventricular systolic and diastolic dysfunction. We evaluated the presence of such dysfunction through conventional echocardiographic investigation and tis­sue Doppler imag­­ing (TDI).

Patients and methods

This single-centre prospective study enrol­led 50 patients with JME fol­lowed up at the epilepsy outpatient clinic within the Antalya Train­­ing and Research Hospital’s Neurology Department (34 women, mean age 26 ± 7.58 years) and 45 healthy controls (30 women, mean age 26.71 ± 5.14 years). Patients were dia­gnosed by three neurologists who were specialized in epilepsy. The neurological examination of the patients and controls was performed by that team of neurologists and their cardiologic examination was performed by a cardiologist.

Exclusion criteria included pregnancy, known coronary artery dis­ease, hepatic andrenal dysfunction, arterial hyper­tension, val­vular heart dis­ease, left ventri­cular systolic dysfunction (ejection fraction <50%), restric­tive, hypertrophic or dilated cardiomyopathy, congenital cardiac dis­ease, pulmonary patho­logies, diabetes mel­litus, malignancies, se­vere alcohol consumption, hyperlipidaemia, previous cardiac surgery, severe mitral an­nular calcification, atrial fibril­lation or other severe ar­rhythmias, use of pacemaker or im­plantable cardioverter defibril­lator, poor echocardiographic image quality. Two pa­tients were excluded since they had arterial hypertension, one patient was excluded due to diabetes mel­litus and one patient was excluded due to rheumatic mitral stenosis and thus 46 patients were included in the study and 40 of them were seizure-free.

The epilepsy dia­gnosis was established on the basis of the Guidelines of the International League Against Epilepsy, and the patients were dia­gnosed accord­­ing to the medical history of the patients and their relatives, physical and neurological examination, as well as the EEG and neuro­radiological findings [6].

Blood pres­sure of both groups was measured in sitt­­ing position after 5-min rest. Height, weight and waist circumference of both groups were measured and their body mass index and body surface area were calculated. Potential SUDEP risk in patients was estimated us­­ing an inventory of seven validated SUDEP risk factors (SUDEP-7), which was as­sembled from a large prospective cohort study of SUDEP reported by Walczak et al. [7]. Based on this study, Di Giorgo et al. validated a SUDEP-7 inventory [8]. This inventory consists of seven items in which scores were based on the log of the odds ratio of the main risk factors [7,8]. The score on the SUDEP-7 ranged from 1 to 7, out of a maximum pos­sible score of 10.

Electrocardiographic study

Standard 12-lead ECG was recorded in all participants to study the fol­low­­ing variables: heart rate, PR interval (the time between the begin­n­­ing of P-wave and the begin­n­­ing of QRS complex), QRS complex duration, QT interval (the time between the begin­n­­ing of QRS complex and the end of T-wave), QTc interval (the QT interval after cor­rection for heart rate, which is the QT interval divided by the square root of RR interval [Bazett’s formula]).

Echocardiography study

The echocardiographic investigation of all patients and the control group was performed using Philips IE33 xMatric (Philips, Andover, MA, USA) with 5-1MHz matrix transducer us­­ing transthoracic approach. The images of the patients were obtained in left lateral decubitus position, while measurements were performed in 2D, M Mode, continuous-wave Doppler, pulse-wave Doppler and TDI echocardiography us­­ing parasternal long and short axis, apical four-chamber and five-chamber views. Left ventricular and right atrium diameters as well as interventricular septum and posterior wall thickness were measured us­­ing M mode technique in line with the recom­mendations of the American Society of Echocardiography [9]. The early (E) and late (A) peak velocities of diastolic fil­ling, deceleration time (DT) and isovolumetric relaxation time (IVRT) were obtained from mitral inflow and aortic outflow Doppler records [10]. The filter sett­­ing was reduced and Nyquist limit was adjusted for TDI measurement (range 15– 20 cm/ s). The gain of the device was reduced in order to decrease the noise in the background and obtain a clear tis­sue signal. For TDI measurements to be performed from the apical window, 5- mm sample volume was placed lateral to the mitral valvular an­nulus on 4-chamber view [10]. Measurements were performed simultaneously with ECG at a rate of 50– 100 m­m/ s. Peak systolic velocity (s’), early (e’) and late (a’) diastolic velocities and ratio of e’/ a’ were measured. For record­­ing pulmonary venous flow velocities, right superior pulmonary venous flow in the apical four-chamber view was used and sample volume was placed 1– 2 cm into the orifice. Here, pulmonary venous systolic forward flow (PVS) and pulmonary venous diastolic forward flow (PVD) were recorded. The ratio of E/ e’ was obtained by divid­­ing transmitral E peak to e’. The dia­gnosis and stag­­ing of left ventricular diastolic dysfunction (LVDD) were done accord­­ing to the recom­mendations of the European Society of Echocardiography [11]. LVDD was dia­gnosed when lateral mitral an­nular e’ velocity was lower than 10 cm/ s or left atrium volume index was greater than 34 mL/ m². LVDD has 3 stages (I, II and III). When LVDD was dia­gnosed, Doppler parameters such as E/ A ratio, DT, IVRT, E/ e’ ratio and PVS/ PVD ratio were used for staging [11]. Ejection fraction was quantified accord­­ing to Teicholz formula [12]. All echocardiographic investigations were recorded digital­ly so that they could be examined of­f-line later. All measurements were performed by calculat­­ing the average of three cardiac cycles in order to decrease the respiratory variation.

Heart rate variability

In order to evaluate HRV of the patients and controls, 24-h ECG records were performed us­­ing 3-chan­nel digital Holter record­­ing device DMS300-3A (DM Software Inc., Stateline, NV, USA). Holter records were analyzed us­­ing the software Cardioscan 12,0 (DM Software Inc., Stateline, NV, US) and the records were evaluated manual­ly in order to rule out the artifacts (ectopic beats, ar­rhythmic events, mis­s­­ing data and noise ef­fects).

During HRV evaluation, standard deviation (SD) of the normal-to-normal (NN) interval (SDNN), the square root of the mean squared differences of successive NN intervals (RMSSD), the division of the number of interval differences of successive NN intervals of more than 50 ms by the total number of NN intervals (pNN50) were analyzed for the time domain parameters. SDNN represents a global measure of HRV. RMSSD is considered a powerful measure of high frequency (HF) variations in short-term recording, as it provides a useful evaluation of HF and vagal tone [2]. The fast Fourier transformation (FFT) has been used to calculate the power spectrum density. Total power (S POWER) in the range 0.0033–0.4 Hz, power in very low frequency (VLF) in the range 0.0033–0.04 Hz, power in low frequency (LF) in the range 0.04–0.15 Hz and power in high frequency (HF) in the range 0.15–0.4 Hz. were analyzed for frequency domain parameters. VLF was evaluated although this component does not have a wel­l-defined physiological explanation [2]. The HF was regarded as a measure of solely parasympathetic activity. The LF was considered to be a measure of mainly sympathetic activity that was modulated by the influence of the parasympathetic system. LF/ HF ratio expres­ses the balance between sympathetic and parasympathetic nervous system activity. All analyses were performed accord­­ing to the standards set by the Task Force of the European Society of Cardiology and the North American Society of Pac­­ing and Electrophysiology [2].

This study was conducted upon the local approval no 46/ 12 and dated September 11, 2014 of the ethics board of Antalya Train­­ing and Research Hospital, and written consent forms were obtained from everyone included in the study. This study was conducted in compliance with the Declaration of Helsinki.

Statistical analysis

All of the statistical analyses were performed us­­ing the SPSS software package, version 21.0 (SPSS Inc., Chicago, IL, USA). The continuous variables were expres­sed as mean ± SD and the categorical variables as numbers and percentages. The Kol­mo­gorov- Smirnov test was used to determine whether the data were normal­ly distributed. The Man­n- Whitney U test was used for variables that were not normal­ly distributed. Statistical significance was defined as p < 0.05.

Results

All participants were Caucasians. Thirty-two patients (69.56%) out of 46 included in the study were female and the mean age was 25.30 ± 6.18 years. In the control group, 30 individuals (66.66%) were female and the mean age was 26.71 ± 5.14 years while the control group included 45 healthy individuals. There was no dif­ference between the two groups with respect to height, weight, body mass index, body surface area, waist circumference, systolic and diastolic blood pres­sure and rest­­ing heart rate (Tab. 1).

In JME group, 24 patients did not have any is­sues in their history; however, 6 patients had a history of febrile convulsion, one had a history of hypoxic delivery, one had a history of CNS infection, 11 had a head trauma and three had a history of febrile convulsion and head trauma. Thirty-two patients did not have any is­sues in their family history, whereas 13 patients had a family history of epilepsy and the parents of one patient had consanguineous mar­riage. The mean epilepsy duration was 10.43 ± 6.67 years while the mean duration of drug use was 8.43 (max. 27, min. 1) years, the ave­rage frequency of seizures was 0.25 ± 0.7(max. 4, min. 0) per month. For the epilep­tic treatment, 20 patients were tak­­ing val­proic acid, 10 patients were tak­­ing lamotrigine, 7 patients were tak­­ing levetiracetam, 1 patient was tak­­ing topi­ramate, 6 patients were tak­­ing dual com­bination of these medications while 2 pa­tients were on triple combination, mean SUDEP-7 inventory score of patients was 0.19 (max. 2, min. 0). Tab. 2 shows the com­ponents of the SUDEP-7 inventory and the number of sub­jects with each factor.

Echocardiography records of both groups and 24-h rhythm Holter were analyzed and registered. There was no significant dif­ference between the JME group and control group with respect to time domain (SDNN, RMSSD, pNN50) and frequency domain (S POWER, VLF, LF, HF, LF/ HF) parameters of HRV. No statistical­ly significant dif­ference was found with respect to left atrium diameter, area and volume index (p = 0.374, p = 0.256 and p = 0.398, resp.); likewise, interventricular septum and posterior wall thickness were similar in both groups (p = 0.703 and p = 0.554, resp.); moreover, there was no statistical dif­ference between the other anatomic echocardiographic values. Furthermore, there was no dif­ference between the Doppler and TDI parameters such as E wave velocity, E/ A ratio, DT, IVRT, e’/ a’ ratio, E/ e’ ratio except PVS/ PVD (p = 0.008). Analysis showed that JME patients had a significantly lower mean of heart rate, QRS duration, QT and QTc interval (p = 0.007, p = 0.032, p = 0.027, p = 0.021 resp.). Tab. 3 presents the HRV, echocardiographic and ECG data of the patient and control groups.

Discus­sion

Sinus ar­rhythmia refers to the respiratory cyclic changes in the heart rate and is as­sociated with the cardio-respiratory con­nection in the brain stem. Physiological HRV is the indication of cardiac health while its absence or reduction may be a sign of a dis­ease or ag­­ing proces­s [13]. HRV analysis is a strong tool that gives semi-quantitative information about the relationship between the cardiovascular sympathetic and parasympathetic modula­tion under several physiological and patho­physiological conditions [14].

Compared to the healthy controls, patients with chronic temporal lobe epilepsy were found to have declined HRV. Although the results suggest that there is a decrease in theparasympathetic tonus and/ or increase in the sympathetic tonus, the data about the re-lationship between epilepsy and the variations in the sympathetic and parasympathetic tonus are still insuf­ficient [15]. We did not find HRV changes in our JME group which may be due to dif­ferences in anti-epileptic drugs (AEDs) such as clas­sical sodium blockers that are used in temporal lobe epilepsy but are not used in JME. Increased sympathetic tone of the heart rate may play an important role in the development of ventricular tachycardia that may be as­sociated with the high incidence of SUDEP in generalized tonic clonic seizure (GTCS) patients compared to the control group. Recently, a meta-analysis of HRV in epilepsy reported a lack of significant alterations of LF in epilepsy patients compared to controls; however, this parameter was found to be lower in patients with AEDs when compared to drug-free subjects [16]. As mentioned in the SUDEP-7 inventory that was designed to analyze risk factors identified by Walczak et al., seizure frequency is a wel­l-known risk factor for SUDEP, but recent studies reported that seizure frequency of GTCS is much more important than for patients with any other type of seizure [7,17]. In the SUDEP-7 inventory, results indicated that older age, longer duration of epilepsy, and presence of developmental disability had direct influence on vagus-mediated HRV and thus increased SUDEP risk. The higher the SUDEP-7 inventory score, the higher the risk of SUDEP. Lower RMSSD values were as­sociated with higher risk scores on the new SUDEP risk inventory. This provides new evidence that HRV (specifical­ly RMSSD) is a marker of SUDEP risk [8].

Heart rate variability is observed to vary across the treated and untreated patients. Haliloğlu et al. found in their study that children tak­­ing valproic acid, oxcarbazepine or phenobarbital had better HRV results compared to the untreated ones [18]. Contrary to this finding, we did not find any dif­ference with respect to heart rate and HRV parameters, which might be due to the fact that the seizures were control­led wel­l. In a recent trial, comparison of changes in HRV parameters in the peri-apnea / hypopnea period of patients with JME and healthy controls did not show any significant difference except for SDNN [19]. However we did not find any significant difference between JME patients and healthy controls in none of HRV parameters. Further studies us­­ing a larger sample size and in patients with newly dia­gnosed and/ or drug-resistant epilepsy may prove beneficial in understand­­ing the mechanism.

Mechanisms underly­­ing SUDEP may lead to fatal ar­rhythmia or ischaemia in patients with or without predispos­­ing functional or structural cardiac abnormalities. However, malignant cardiac ar­rhythmias have ra­rely been seen dur­­ing recorded SUDEP cases [20]. Increased sympathetic cardiac stimulation may lead to cardiac ar­rhythmias but also repetitive stimulation may cause structural damage in the heart, which means that there is a high predisposition to ar­rhythmia and ischaemia. It was found previously that patients with SUDEP had myocardial fibrosis [21]. Myocardial fibrosis may lead to left ventricular systolic and diastolic dysfunction. Troponin does not rise in the post-ictal period in patients with non-complicated seizures, while ischaemia findings on ECG and elevated cardiac enzymes suggest that epileptic patients may have secondary cardiac damage [22,23]. Alehan et al. showed that brain natriuretic peptide (BMP) and creatine kinase-muscle/brain (CK-MB) were elevated in postictal period in patients with seizures and they obtained the evidence proving that epileptic patients may had mild cardiac dysfunction which was difficult to detect [24]. Elevated sympathetic activity may lead to left ventricular dilatation which may result in Takotsubo and stres­s-induced cardiomyopathy reported in GTCS. This may af­fect the cardiac output and lead to insuf­ficient peri-ictal oxygen supply [25,26]. Two studies report­­ing that recur­rent Takotsubo cardiomyopathy developed due to GTCS and convulsive status epilepticus were published [27,28]. The cited authors stated that this occur­red due to the sympathetic fluctuations. Bilgi et al. reported that patients with GTCS had not only systolic and diastolic dysfunction but had also impaired end-systolic left ventricular diameter and volume [5]. In a recently published study, increased left atrial diameter was reported in patients with GTCS compared to the control group. The authors argued that this might be caused by the increased left ventricular end-systolic pres­sure due to the sympathetic activity [29]. Pulmonary vein Doppler method contains load-dependent parameters and thus can be used for the estimation of left ventricular fil­l­­ing pres­sures [30]. Kuecherer et al. thought that increased left atrial pres­sure was negatively cor­related with PVS and PVS2/ PVD ratio in patients with pseudo normal or restrictive pattern [31]. In our study, we did not find a negative change in PVS/ PVD either, similar to other echocardiographic parameters.

Cardiac ion chan­nels are subdivided as de­polariz­­ing and repolariz­­ing ones. Prolo­n­­ged repolarization can cause triggered activity, while rapid repolarization may lead to reentrant excitation. Eric et al. reported decreased heart rate from 80 to 67 and increased QRS intervals in Lamotrigine users [32]. In our study, we could have detected decreased heart rate in the patient group because 10 of our patients were us­­ing LTG but decreased QRS intervals could be due to other AEDs ef­fects. In the line of our study, Ramadan et al. reported shorten­­ing of QT and QTc intervals in newly dia­gnosed GTCS patients without medication [29]. There are, however, conflict­­ing data on whether QTc intervals in epileptic patients dif­fer from controls [33,34]. A potential confounder in these studies could be, at least partly, the use of dif­ferent AEDs.

Further studies are needed us­­ing a larger sample size and in patients with newly dia­g­nosed and/ or drug-resistant JME.

Conclusion

Epilepsy patients may be predisposed to autonomic dysfunctions which may be as­sociated with cardiac ar­rhythmias due to the ef­fects of recur­rent seizures on the cardiac microstructure. There are several studies demonstrat­­ing the ef­fect of epilepsy on cardiac and autonomic functions. However, studies on JME patients, which represents a specific patient group, are very limited. Our study did not show any find­­ing within echocardiographic and HRV parameters suggest­­ing cardiac involvement or auto­nomic dysfunction in patients with JME who were undergo­­ing treatment.

Accepted for review: 5. 12. 2017

Accepted for print: 10. 10. 2018

Fatma Genç, MD

Department of Neurology

Antalya Training and Research Hospital

Varlık Mahallesi

Kazım Karabekir Cd.

07100 Muratpaşa/Antalya

Turkey

e-mail: sanivardr@yahoo.com