Materıals and methods

Our study consists of patient and control groups. Fifty-seven coeliac patients and 57 healthy patients were included in the study. Coeliac patients without additional chronic diseases who applied to the Paediatric Gastroenterology Unit were included as patient group. The cases who applied to the Paediatric Cardiology Unit for sports report without coeliac disease and additional chronic diseases were included as the control group. Cases similar to the patient group in terms of age and gender distribution were included in the study.

Physical examination and anthropometric measurements of coeliac patients and control group were performed. Routine blood tests (complete blood count, alanine aminotransferase, aspartate aminotransferase, urea, creatinine, ferritin, vitamin B12, calcium, magnesium, folic acid, C-reactive protein, TSH, ft4, IgA, glucose, and coeliac autoantibodies), age of diagnosis, complaints, dietary compliance, endoscopy and pathology findings, electrocardiogram, and echocardiography findings were recorded with a prospectively standardised form. The families were informed about the study, and written consent was obtained from each individual.

Electrocardiogram evaluation was done by the same physician. After all patients were rested for 10 minutes, 12-lead electrocardiogram records of 25 mm / sec velocity and 10 mm/mV amplitude were obtained. The data on the standard 12-lead electrocardiogram derivations were examined. The ventricular rate of three consecutive beats in the DII derivation was calculated and averaged, and the average heart rate of each patient was determined. In addition, P wave, PR interval, QRS complex, QT interval, QTc, Tp-e, Tp-e / QT, Tp-e / QTc, QT dispersion, and QTc dispersion calculations were made. Echocardiographic examinations were performed with two-dimensional and colour Doppler echocardiography device. Echocardiographic evaluation was performed by the same paediatric cardiologist. As classical echocardiographic measurements in all cases, left ventricular end-diastolic diameter, left ventricular end systolic diameter, end-diastolic interventricular septum thickness, end-diastolic left ventricular posterior wall thickness, ejection fraction, and fractional shortening measurements were made.

Proportions in independent groups were compared using Chi-square analysis. When numerical variables in more than two independent groups met the normal distribution condition in the groups, the one-way ANOVA test was performed with the Kruskal Wallis test when the normal distribution condition was not met. Student’s t test was used to compare groups. Statistical significance level was accepted as p < 0.05. SPSS 15.0 for Windows program was used for statistical analysis.

Dıscussıon

Although coeliac disease is known as small bowel disease, it has become a disease of every system with its extra-gastrointestinal system findings that have been revealed and evident in recent years. ^{Reference Maki, Lohi, Walker, Goulet, Kleinman and tal5} Most patients have an atypical or silent clinical course. ^{Reference Maki, Lohi, Walker, Goulet, Kleinman and tal5,Reference Fasano and Catassi6} The disease is more common in women. ^{Reference Ivarsson, Persson and Nystrom7,Reference Fasano, Berti and Gerarduzzi8} Cardiovascular complications in coeliac disease are known to be one of the leading causes of morbidity and mortality. ^{Reference Ludvigsson, Montgomery and Ekbom9} Autoimmune myocarditis, idiopathic dilated cardiomyopathy, ischaemic heart disease, and aortic dysfunction can be seen. ^{Reference Ludvigsson, James and Askling1,Reference Sari, Ertem and Sari10} Left ventricular dysfunction leads to arrhythmia development. ^{Reference Bigger11} In a study of 52 patients, the incidence of coeliac disease was reported to be 5.7% in patients with dilated cardiomyopathy. ^{Reference Curione, Barbato and De Biase12} In a large population-based cohort study conducted in Sweden, a positive relationship was found between coeliac disease and dilated cardiomyopathy and ischaemic heart disease. ^{Reference Ludvigsson, James and Askling1}

There are many mechanisms explaining cardiovascular involvement in coeliac disease. Chronic malabsorption seen in coeliac disease leads to cardiomyopathy due to nutritional deficiencies. Plasma homocysteine levels increase due to malabsorption of folic acid and vitamin B12. Increasing plasma homocysteine level causes structural and functional abnormalities of cardiomyocytes, causing cardiomyopathy. ^{Reference Lim, Tzemos and Farquharson13} Intestinal permeability abnormalities seen in coeliac disease cause increased absorption of antigens and infectious agents from the intestine. This causes the activation of immune mechanisms and eventually myocardial damage. ^{Reference van Elburg, Uil and Mulder14} Villous atrophy seen in coeliac disease causes impaired absorption of thiamine, riboflavin, magnesium, calcium, selenium, and carnitine, which play a role in myocardial metabolism. ^{Reference Curione, Barbato and Viola15} Autoimmune mechanism triggered by gliadin in coeliac disease causes cardiac damage. Deamidated gliadin binds to HLA-DQ2 or HLA-DQ8 with high affinity and causes activation of T lymphocytes. Activated T lymphocytes secrete high amounts of pro-inflammatory cytokines such as Tumor Necrosis Factor. These secreted cytokines are thought to trigger atherosclerosis by causing endothelial damage and cause ischaemic heart diseases and heart failure. There are studies showing that the absorption of oligoelements is better with a gluten-free diet, which has an advantage on myocardial contractility and electrical stability. ^{Reference Peracchi, Trovato and Longhi16}

Many of the electrocardiogram parameters provide preliminary information in terms of current or possible clinical situations. QT interval and T wave are important for showing cardiac repolarisation. In our study, we compared the Tp-e interval and QT dispersion values of coeliac patients with healthy children. QT interval reflects the sum of the depolarisation and repolarisation times of the ventricular myocardium. It has been shown that the increase in QT dispersion, which is considered to indicate regional heterogeneity in myocardial repolarisation, causes serious ventricular arrhythmias and sudden cardiac death through the reentry mechanism. ^{Reference Pye and Cobbe17} Tp-e interval corresponds to ventricular repolarisation dispersion. Increased ventricular repolarisation dispersion is seen as an important risk factor for ventricular arrhythmias. The prolonged Tp-e interval reflects the abnormal distribution of ventricular repolarisation and is associated with an increased risk of ventricular arrhythmia. ^{Reference Castro-Torres, Carmona-Puerta and Katholi2} Therefore, the Tp-e interval can be used as a non-invasive screening method for arrhythmogenesis.

Cardiovascular diseases characterised by left ventricular dysfunction such as idiopathic congestive heart failure and myocarditis are seen in coeliac disease. ^{Reference Frustaci, Cuoco and Chimenti18} In a study conducted with 187 patients with autoimmune myocarditis, 9 patients had coeliac disease. Heart failure was detected in 5 of 9 patients, and ventricular arrhythmia was detected in 4 of them. When these patients were given a gluten-free diet, it was observed that cardiac functions improved and arrhythmias decreased. ^{Reference Frustaci, Cuoco and Chimenti18} Polat et al. detected subclinical left ventricular dysfunction in paediatric coeliac patients with positive endomysium antibody IgA levels. ^{Reference Polat, Urganci and Yalcin19} Sarı et al demonstrated the subclinical effect of coeliac disease on left ventricular systolic function using strain echocardiography. ^{Reference Sari, A.D. and Akin20} In our study, ejection fraction and fractional shortening values were found to be significantly lower in coeliac patients compared to the control group (p = 0.001 and p = 0.002, respectively). Ventricular arrhythmia is more common in patients with left ventricular dysfunction than normal left ventricular function. Increased ventricular repolarisation dispersion is associated with malignant arrhythmias. Prolonged QT interval is one of the markers of delayed cardiac repolarisation in electrocardiography. Corozza et al. found prolongation of the QT interval in 33% of coeliac patients in their adult age study. ^{Reference Corazza, Frisoni and Filipponi21}

In recent studies, Tp-e interval, Tp-e / QT ratio, and Tp-e / QTc ratio are parameters showing the increase in ventricular repolarisation dispersion. ^{Reference Gupta, Patel and Patel3} The prolonged Tp-e interval is associated with increased mortality in patients with long QT syndrome, Brugada syndrome, and acute myocardial infarction. Since the Tp-e / QT ratio is not affected by heart rate and body weight, it is a more sensitive marker than others. ^{Reference Gupta, Patel and Patel3}

In our study, Tp-e interval, Tp-e / QT ratio, and Tp-e / QTc ratio were found to be significantly higher in coeliac patients compared to the control group. Left ventricular dysfunction seen in coeliac disease is thought to play a role in the development of arrhythmia. ^{Reference Bigger11} Therefore, these markers may be helpful in predicting ventricular arrhythmia and mortality in coeliac patients. Demirtaş et al. in their study conducted with 38 adult age group of coeliac patients found that the Tp-e interval, Tp-e / QT ratio, and Tp-e / QTc ratio of coeliac patients were significantly higher than the control group consisting of 38 patients. ^{Reference Demirtas, Yayla and Yuksel22}

In our study, no statistically significant correlation was found between disease duration and Tp-e interval, Tp-e / QT ratio, Tp-e / QTc ratio. The reason for this is that 63.2% of the patients who participated in our study were able to adapt to a gluten-free diet.

Chronic inflammation seen in Coeliac disease is thought to cause atherosclerosis and vascular damage. In a study conducted with coeliac patients, the relationship of endothelial dysfunction with the increase in carotid intima-media thickness, which is an indicator of macrovascular diseases, was demonstrated. ^{Reference Libby, Ridker and Hansson23} Studies have shown that there is a relationship between carotid intima-media thickness and cardiac repolarisation disorders. ^{Reference Ede, Ibrahim Serin and Ede24} In a study conducted in paediatric patients with ulcerative colitis and crohn’s disease with chronic inflammation, QT dispersion and QTc dispersion were found to be prolonged compared to the normal population. This study predicts that inflammatory bowel diseases are at risk for ventricular arrhythmias. ^{Reference Bornaun, Yilmaz and Kutluk25} In another study, it was found that QT dispersion, QTc dispersion, and Tp-e interval were prolonged in paediatric Kawasaki patients compared to the normal population (26). In our study, no statistically significant difference was found between the patient and control groups in terms of QT dispersion and QTc dispersion values.

In our study, the correlation of tissue transglutaminase IgA positive or negativity with the ventricular repolarisation markers Tp-e interval, Tp-e / QT ratio, Tp-e / QTc ratio, QT dispersion, and QTc dispersion were evaluated. Tp-e interval, Tp-e / QT ratio, and Tp-e / QTc ratio were found to be statistically significantly higher in patients with positive tissue transglutaminase IgA compared to patients with negative tissue transglutaminase IgA. QT dispersion did not differ significantly. In a study involving 288 patients with end-stage heart failure without clinical signs of coeliac disease awaiting heart transplantation, almost half of the patients were found to be positive for tissue transglutaminase IgA and tissue transglutaminase IgG (73). In another study, there is evidence that tissue transglutaminase antibodies have an antiangiogenic effect that can disrupt the functioning of the vascular system, impair myocardial electrical activity and contraction, and trigger atherosclerosis. ^{Reference Sane, Kontos and Greenberg27} We found tissue transglutaminase antibody positivity in our study compatible with the literature in order to predict cardiac involvement.

In our study, the correlation between endomysium antibody IgA positive or negativity and Tp-e interval, Tp-e / QT ratio, Tp-e / QTc ratio, QT dispersion, and QTc dispersion were evaluated in coeliac patients. Only the Tp-e / QT ratio of patients with endomysium antibody IgA positive was found to be statistically significantly higher compared to patients with endomysium antibody IgA negative. The reason for differentiation with tissue transglutaminase can be shown as tissue transglutaminase positive while endomysium antibody IgA negative in 4 patients. In the literature, there are studies showing that left ventricular function and myocardial performance decrease in patients with endomysium antibody IgA positive compared to negative ones. ^{Reference Saylan, Cevik and Kirsaclioglu28} In our study, more patients are needed to use endomysium antibody IgA positivity as a marker for predicting cardiac arrhythmias.

In our study, Marsh stages of coeliac patients Tp-e interval, Tp-e / QT ratio, Tp-e / QTc ratio, QT dispersion, and QTc dispersion were compared. There was no significant relationship between Marsh stages and ventricular repolarisation markers. The marsh stage of the patients shows the stages at the time of diagnosis. We could not find a significant relationship between marsh staging and repolarisation markers due to the patients' compliance with gluten-free diet and histological improvement beginning within 6–24 months.

In conclusion, our study gives important findings in terms of detecting the early signs of future cardiovascular events in paediatric coeliac patients. Repolarisation differences compared to the normal population may reflect early subclinical findings associated with coeliac disease. Prospective wider studies are needed to capture these early changes and their future significance. Since coeliac patients included in our study did not have electrocardiogram in the healthy period, that could be an important data for detecting electrocardiogram changes before and after coeliac disease, limits our study for the inability to evaluate electrocardiogram retrospectively.