Tinnitus is one of the most common otological symptoms [1]. It is known as the sensation of sound in the ear without an outer audible stimulus and is usually classified according to the duration of the disease. For example, tinnitus is classified as acute when the condition duration is less than three months, and chronic when it is more than three months [2]. The mechanisms of cochlear and stress have been found to be involved in the etiopathogenesis of tinnitus [3]. The association between stress and tinnitus has been reported in a previous study, and stress is considered a risk factor for tinnitus [4]. The sympathetic branch of the autonomic nervous system (ANS) is related to stress, and its chronic form can also affect tinnitus [5]. A previous study suggested a positive correlation between sympathetic tone and distress related to tinnitus. In the same study, a negative correlation between parasympathetic tone and distress related to tinnitus was found [6]. Chronic tinnitus may be associated with chronic stress due to dysfunction of autonomic regulation [7].

ANS regulates the heart rate and heart rhythm. Dysfunction in the ANS might provoke alterations in ventricular repolarization and increase the risk of malignant ventricular arrhythmias. [8]. Many electrocardiogram (ECG) parameters have been used to predict ventricular arrhythmias. For instance, T wave peak-to-end T(p-e) is used as an index of the transmural dispersion of repolarization (TDR), while T(p-e)/QT ratio includes values of the transmural and spatial dispersion of repolarization. TDR is a potential indicator for ventricular arrhythmias [9‐11]. Recently, the index of cardiac-electrophysiological balance (iCEB), which is calculated as QT interval divided by QRS duration (QT/QRS), was identified as a potential marker for anticipating drug-induced ventricular arrhythmias in an animal model [12]. Multiple studies suggested that iCEB can be used as a surrogate of cardiac wavelength, which in turn is an essential factor for ventricular arrhythmias susceptibility but can only be measured via invasive electrophysiology. It has been reported that iCEB might be a noninvasive and quickly assessable marker to identify the raised risk of arrhythmic in patients [13]. The goal of the present study was to evaluate ventricular arrhythmia indexes in patients with tinnitus.

The present study consisted of 240 patients with tinnitus (144 males; mean age = 47.82 ± 8.53 years) and 240 healthy controls (156 males; mean age = 46.54 ± 5.32 years). The clinical characteristics, echocardiographic and laboratory data of the participants are presented in Table 1. There were no significant differences in terms of gender, age, smoking status, BMI, diastolic and systolic blood pressure, LVEF and creatinine, potassium, calcium, magnesium, fasting glucose and thyroid-stimulating hormone levels between the two groups (p > 0.05 for all).

The electrocardiographic findings of the two groups are presented in Table 2. The mean heart rate and QRS duration were similar between the patients and the control group (76.24 ± 10.02 vs 77.32 ± 10.26 beats/min, p = 0.624; 96.04 ± 18.72 vs 97.24 ± 12.32 ms, p = 0.565, respectively). Compared to the control group, QT (376.46 ± 36.54 vs 346.52 ± 24.51 ms), QTc (426.68 ± 24.68 vs 390.42 ± 24.04 ms), T(p-e) (75.86 ± 14.68 vs 62.42 ± 8.64 ms), T(p-e)/QT (0.201 ± 0.06 vs 0.180 ± 0.01) and T(p-e)/QTc (0.177 ± 0.06 vs 0.159 ± 0.02) were higher in patients with tinnitus (p < 0.001 for all). QT/QRS (3.92 ± 0.68 vs 3.56 ± 0.32) and QTc/QRS (4.44 ± 1.03 vs 4.01 ± 0.64) were significantly higher in patients with tinnitus (p = 0.018 and p = 0.008, respectively). A significant positive correlation was found between T(p-e) and disease duration (r = 0.792, p < 0.001) (Fig. 1). In addition, a significant positive correlation was found between T(p-e)/QTc ratio and disease duration (r = 0.500, p < 0.001) (Fig. 2).

In the present study, QT, QTc, T(p-e), T(p-e)/QT, T(p-e)/QTc, iCEB and iCEBc parameters were significantly higher in patients with tinnitus when compared to the control group. In addition, significant positive correlations were found between T(p-e), T(p-e)/QTc ratio and the disease duration. These results point out a potential increased risk of malignant ventricular arrhythmias in patients with tinnitus. To the best of our knowledge, no study has examined these parameters to evaluate the risk of malignant ventricular arrhythmias in patients with tinnitus in the literature.

Tinnitus is known as the sensation of sound in the ear without an outer audible stimulus [17]. The etiopathogenesis of tinnitus is not only associated with the cochlear mechanisms but also with stress. Current pathophysiological models link the initiation and maintenance of tinnitus with stress [18, 19]. In addition, studies have shown that stress is an effective factor that can turn mild tinnitus into a severe one [20]. The sympathetic branch of the ANS plays a major role in stress [21]. ANS also plays a major role in the development of ventricular arrhythmias, as it is a major modulator of ventricular repolarization. Several mechanisms linking ANS dysfunction and ventricular arrhythmias are very well established, albeit some details are still to be unravelled [22]. Methods such as hormonal mediator tests, pharmacological tests and various psychological stress tests could be used to evaluate ANS functions. One of these methods is heart rate variability (HRV), which is controlled by sympathetic and parasympathetic cardiac nerves. While sympathetic activity increases heart rate, parasympathetic activity decreases heart rate. HRV studies have also improved our knowledge of the effects of physiological events on tinnitus. It has been found that parasympathetic activity and sympathetic activity can be assessed using HRV parameters. The association between ANS dysfunction and cardiac autonomous dysfunction has been investigated in many studies through using HRV. A previous study showed an increase in sympathetic activity in patients with rheumatoid arthritis using HRV parameters and reported that this might cause malignant ventricular arrhythmias [23]. Patients with systemic lupus erythematosus have been found to have abnormal HRV, which reflected increased sympathetic activity and cardiac autonomic dysfunction [24]. Choi et al. evaluated HRV parameters in patients with tinnitus and demonstrated that sympathetic activity increased in these patients [25]. They also observed a positive correlation of the tinnitus period with sympathetic activity [25]. Another study that investigated the relationship between neural activity and the severity of tinnitus showed a positive correlation between sympathetic markers and tinnitus distress, and a negative correlation between tinnitus distress and parasympathetic markers [26]. Betz et al. observed that HRV parameters were similar between patients with tinnitus and the healthy control group. However, subjectively, they reported an increased level of tinnitus after exposure to stress in the same study [27]. HRV measurements are complex and quite different, as different researchers measure different time units, evaluation and comparison using the obtained data may give inaccurate outcomes [28].

T wave represents ventricular repolarization at ECG. T(p-e) interval has been used as an index of the TDR. HRV assesses the sympathetic-parasympathetic balance at the sinus node level, while measures of ventricular repolarization are influenced, among other factors, by the sympathetic-parasympathetic balance at the ventricular level. Parameters such as T(p-e)/QT and T(p-e)/QTc ratios, which are sensitive, have recently been used to predict ventricular repolarization dispersion [29, 30]. An increase in T(p-e)/QT ratio and T(p-e) interval has been reported in various cardiac diseases such as Brugada, long QT and short QT syndromes, and noncardiac diseases such as autoimmune hepatitis, psoriasis and obstructive sleep apnoea [30‐33]. Lecca et al. reported that QTc was prolonged among workers operating at a site where stress level was found to be elevated and suggested that work-related stress had subclinical effects on the cardiac function autonomic regulation [34]. Afsin et al. indicated that the T(p-e)/QT ratio and T(p-e) interval increased among patients with panic disorder and suggested that this might be due to the dysfunction of ANS through elevated sympathetic activity [35]. Similarly, it has been reported that T(p-e) and T(p-e)/QT values increased in patients with gastroesophageal reflux disease, which might be due to increased sympathetic activity [36]. A previous study reported that tinnitus is associated with sleep deprivation, emotional difficulties and social interaction problems [37]. A positive correlation has been reported between sleep deprivation and T(p-e)/QT and T(p-e) values [38]. Our results showed a significant increase in T(p-e) and T(p-e)/QT values in patients with tinnitus and this might be due to the dysfunction of ANS through elevated sympathetic activity, which is consistent with previous studies.

iCEB is a noninvasive parameter used to estimate the risk of ventricular proarrhythmia. iCEB may offer knowledge for both repolarization and depolarisation of the cardiac action potential. Such a feature may estimate the cardiac proarrhythmia risk better than T wave with repolarization only, TDR and QT interval variability. There are few studies on iCEB in the literature. Nafakhi et al. demonstrated that an increase in iCEB was related to the raised pericardial fat volume in patients with suspected coronary artery disease [39]. Yumurtaci et al. demonstrated that patients with acute myocarditis who had arrhythmia had higher levels of iCEB, T(p-e), T(p-e)/QTc and T(p-e)/QT than those without arrhythmia [40]. Our study showed that iCEB and iCEBc values were significantly higher in patients with tinnitus compared to the control group. In our study, the larger values of iCEB and iCEBc observed in the patients versus the controls mostly reflect the longer QTc value (no differences were found in QRS duration).

Our result revealed that QT, QTc, T(p-e), T(p-e)/QT, T(p-e)/QTc iCEB and iCEBc values were significantly higher in patients with tinnitus compared to the healthy group. In addition, significant positive correlations were found between T(p-e), T(p-e)/QTc ratio and disease duration. According to our results, patients with tinnitus may have an increased risk of malignant ventricular arrhythmias. Further studies with follow-up are needed to validate our findings.