ELECTROCARDIOGRAPHY CHANGES IN CHILDREN WITH FEBRILE CONVULSION, BREATH-HOLDING COMPARED TO CONTROLS

Methodology: This Cross-sectional study was conducted on 270 children of 90 with febrile convulsion, 90 with breath-holding patients and 90 healthy children in 2019. Electrocardiography parameters of; QT interval, dispersion QT, corrected QT and dispersion QTc measured for participants. Consent form was asked to be signed by the parents before the study execution. Data analysis done by SPSS 20.0 considering 0.05 as significant error.


INTRODUCTION
Cardiovascular changes have a link to seizures 1 and breath holding spells (BHs). 2 QT dispersion (QTd) defined as interval between longest and the shortest QT. 3,4 Prolonging of QT intervals show a high risk of dysrhythmia and sudden death in various illnesses such as; cardiomyopathies, mitral valve prolapses, ischemic coronary heart diseases and end stage renal failure. 3 And noted as a predictor of child mortality due to various diseases 6 such as; diabetes mellitus, 7 celiac, 3 thalassemia, 4 epilepsy, 1 breath holding, 5 febrile seizure, 5 QT intervals also change in general population due to medications, electrolyte abnormalities, or endocrine disease. 8 BHs are well described phenomenon known to occur mostly in children aged 6 to 18 months even later in childhood, up to 4 years of age with 5% due to crying. 9 Cyanotic and pallid are types of BHs such that sometimes, some children have both in their lives. The mechanism of BHs is unknown and still is controversial but expressed that cerebral anoxia, anemia and genetic have impact effects. 10 A seizure is a sudden, uncontrolled electrical disturbance in the brain and occasionally occurred after losing consciousness; and then, return to normal status. 2 In this regards, febrile convulsions (FC) or febrile seizures (FS) that has been described as an event occurring during fever in children elder than one month without history and free of CNS infection, and without specific causes such as electrolyte imbalance, metabolic disorder, intoxication or trauma by ILAE. 1 FS occurring in children aged 6 months to 5 years with fever greater than 38°C (100.4°F), without intracranial cause symptom and without a history of an afebrile seizure. 7,11 In another side, FS has been defined as common neurologic disorder in children that affecting 2-5% of children peaked in the age of 12 -18 months. 7 Although FS is seen in all ethnic groups, but it is more frequent in Asian regions such as; India with 5-10% and Japan with 6-9% 7 and Middle East regions such as; 9.7% to 25-50% in Turkey. 12 The present study aimed to assess QT findings variation in children with breath-holding spells and febrile convulsion compared to healthy ones.

METHODOLOGY
This cross sectional study was conducted on children with FC or with breath holding comparing with healthy group in the pediatric neurology clinics of Zahedan University of Medical Sciences (ZaUMS) Hospitals in 2019. During the year of 2019, 101 children with BHs and 121 children with FC referred to the pediatric clinic of neurology. After considering exclusion criteria, from these children 90 and 96 referred to the unit of pediatric cardiology respectively. For having a similar size in patients and controls, the sample size of 270, 90 in each group considered for the study.
BHs diagnosed based on the criteria of; reported history of BHs by the parents (three or more spells in 1 month), and having the clinical sequences of provocation followed by crying to a point of consciousness. The BHs participants collected from those who referred to the clinics with complain of cyanotic attacks and/or fainting or by observation of the typical attacks during an examination by pediatric specialists.
The exclusion criteria for this group of patients were; epilepsy history, electrolyte imbalances, hypoglycemia, iron deficiency, impaired kidney function tests, abnormal neurological findings during examination, receiving any medications that all criteria have significant effect on QT intervals.
The children with FC collected from those who were referred to the clinics due to complain of seizure. Febrile convulsion was defined as convulsion based on the mentioned definition by ILAE with the exclusion criteria of; abnormal laboratory affecting on ECG such as calcium, potassium, magnesium. Any medications of anti-psychotics, anti-arrhythmia, and antibiotics and free of any underlying cardiovascular disease, trauma, meningitis, encephalitis, seizure-inducing syndromes, and structural disorders.
Healthy children matched in frequency and gender with the patients and selected randomly from those who came to the clinics to have a checkup routine after considering same exclusion criteria.
Electrocardiogram (ECG) changes was detected with an electrocardiogram by the Saadat device made in Iran initially from 30 minutes to 2 hours after seizure. ECG in standard scheme was obtained. Once the participants had rested for 10 minutes in a supine position in a quiet room, all 12 ECG leads were simultaneously recorded at a paper speed of 25mm/s and a voltage of 10mm/mV.
From the 12 lead of electrocardiogram that performed, at least 8 leads selected and from each lead, three complex of QT calculated. In final from these 24 calculated QT, the average considered for the study. QT interval was accepted as the distance from the beginning of the Q wave to the end of the T wave. In each lead, the duration from the beginning of the Q wave to the end of the T wave was calculated in milliseconds, the average was taken (QT average) for three consecutive beats. The maximum and minimum duration of the QT wave was selected from these 24 QT measures. The difference between maximum and minimum duration was defined as QTd.
The average QTc was calculated using the same QT interval measured using the Bazett formula (QTc¼ QT/ √RR); among all derivations, the difference between the longest and shortest QTc was calculated (QTcd). 13,14 To calculate left ventricular mass in ECG we used the following formulas: LV mass (g) =0·026 (RaVL+SV3) + 1·25 Weight + 34·4 for boys, and 0·020 (RaVL+SV3) + 1·12 Weight + 36·2 for girls. 15,16 The QT parameters of QTd, QTc and QTcd were normal when were higher than 0.05, 0.45 and 0.06 second respectively. 17 The mean of Heart rate calculated of three consecutive ventricular heart rates in lead II. Consent form was asked to be signed from the parents. The study coded 2569 and approved by ethical committee of the ZaUMS. For the frequency and percentages, Chi-square test applied for the association and for quantities variables, first normality test applied for diagnosis of variables distribution that founded all had free distribution. To find the differences in ECG findings in pairwise comparisons, unpaired Mann-Whitney Utest applied. The error P<0.05 was considered as significant level.

RESULTS
Gender distribution and abnormality of ECG parameters in Febrile Convulsion, Breath Holding, and Controls are presented in Table 1. The gender distributed similarly in groups of participants (p=0.233). Table 1 showed that QTc was normal in 97.8%, 76.7% and 71.1% of controls, febrile convulsion and breath holding respectively with significant association (p<0.001). QTd was normal in 100.00%, 98.9% and 96.7% of controls, febrile convulsion and breath holding respectively with a non-significant association (p=0.196). QTcd was normal in 95.6%, 72.2% and 85.6% of controls, febrile convulsion and breath holding respectively with a significant association (p<0.001).

DISCUSSION
From the present study resulted that QTc was abnormal in 2.2%, 23.3% and 28.9% and QTcd was abnormal in 4.4%, 27.8% and 14.4% in healthy individuals, FC and BHs in the order given.Also resulted that QTd, QTc min, QTc max and QTcd were similar in groups of FC and controls. QT min and QTc min were similar in BHs and in healthy individuals. QTc min and QTcd were same in BH and FC children. But QTd were higher significantly in the FC compared to children with BH. And QT min, QT max and QTc max were higher in the children with BHs. Olsen et al. 18 showed that in children with BHs, the QT intervals parameters did not change compared to controls. Akalın et al. 19 concluded that QT, QTc were similar in BHs patients and in healthy individuals. But QTd and QTcd were significantly lower in healthy ones compared to BHs children.
Noori et al. 20 conducted a study on children to evaluate electrocardiography findings in BHs compared to in healthy individuals. They found that QT max, QTd, QTc max, and QTcd were different in patients compared to healthy individuals significantly similar to the present study findings. Amoozgar et al. 21 reported that QTcd was higher in BHs compared to healthy individuals when QTd was similar. Therefore, in comparing with our results, it seems that there is a confusing of using the ECG parameters for assessing cardiac involvements in BHs patients because of contradiction in reports. Noori et al. 17 evaluated electrocardiography findings in FC compared to healthy individuals and found that QTc, QTd and QTcd were higher in FC children significantly compared to healthy children. They also found that QT min and QT max did not changed in FC children. Tomoum et al. 22 concluded that a higher QTd in BHs patients compared to healthy individuals. They also observed that during BHs attacks the bradycardia had longer time and more observations of dysrhythmia during cyanotic spells in patients with BHs. These results are importance in clinical status and may led to the effective autonomic dysfunction in nervous system. The link between these findings and dysrhythmia need to be clarified. The knowledge and awareness about ion channel disorders causing paroxysmal symptoms are increasing in various systems due to impaired electrical activity. Previously thoughts of BHs were; being self-limited, benign and resolving spontaneously.
The complications and prolonged QTd revealed more investigations about arrhythmia and ANS dysfunctions in BHs. Kandler et al. 23 showed 23% QTc prolongation in FC within two hours after seizures. Sadrnia et al. 24 resulted a significant prolongation QT in patients with seizure compared to healthy individuals. Brotherstone et al. 25 showed a mild or a short increase in QTc during seizures when El Amrousy 26 found 55% of the patients with convulsion had ECG changes in epileptic seizures during the first six hours. From their study also resulted that, the most frequent abnormalities in ECG were conduction abnormalities, ischemic change, and arrhythmias. The shortening of the QT interval that detected in El Amrousy et al. 26 can be induced by catecholamine release, acidosis, and hyperkalemia which are very common during or shortly after seizures. Yilmaz et al. 10 considered two groups of children with BHs and FC. They observed similar values of QTc in the groups. They also received to the conclusion that in BHs, QTc were similar in cyanotic and pallid types of BHs. In this regards, the present study received to an opposite direction.
Overall, the studies that above mentioned, showed that both BHs and FC are causes of long QT. Some of ECG parameters such as QTd and QTc were different among these two types of diseases such that QTd were higher in the FC compared to BHs and QT and QTc were higher in BHs compared to FC. The electrocardiography parameters need to be assessed for cardiac dysrhythmia that increases the life-threatening particularly in BHs 22 and FCs. 23

Limitations
The study was limited in design and generalizability as it was done in a single center.

CONCLUSION
From the present study concluded that QTc prolonged in breath-holding spells more than febrile convulsion as well more that healthy individuals. QTcd and QTd were prolonged more in febrile convulsion than breath-holding spells and also than healthy individuals. The results of the study suggest an ECG performance for children who suffered from seizures or breath-holding spells to control their cardiac dysrhythmia.