The association between nonalcoholic fatty liver disease and corrected QT interval prolongation among generally healthy Iranian population: Fasa Cohort Study (FACS)

Abstract Background There are limited studies about the association between nonalcoholic fatty liver disease (NAFLD) and corrected QT interval (QTc) prolongation worldwide. Hypothesis Therefore, we designed the current study to determine this association in a large cohort of a generally healthy population. Methods We analyzed the data of 4603 individuals aged 35–70 who participated in the Fasa Cohort Study (FACS). Based on 12‐lead electrocardiograms, QT intervals were calculated and corrected by Bazzet's formula. A QTc interval of more than 430 ms in men and 450 ms in women was considered prolonged. The Fatty Liver Index was used to identify the participants with NAFLD. Results Of all participants, 1550 (33.6%) met the NAFLD criteria. In subjects of both genders with NAFLD, the mean values of the QTc interval were considerably higher than in those without NAFLD (p < .001). After adjusting for a wide range of confounders, including age, gender, smoking status, physical activity, total cholesterol, high‐density lipoprotein‐cholesterol levels, diabetes, and hypertension status, in linear regression analysis, the standardized β coefficient of QTc interval among participants with NAFLD was 2.56 ms (95% confidence interval [CI]: 0.49–4.64). After controlling the same confounders, the odds ratio of NAFLD for a prolonged QTc interval in men was 1.47 (95% CI: 1.18–1.84; p < .001) and in women was 1.39 (95% CI: 1.15–1.68; p < .001) using logistic regression analysis. Conclusions NAFLD was a risk factor for QTc interval prolongation. Awareness about the risk of NAFLD in increasing the potential cardiac arrhythmias should be raised to lower cardiac mortality.


| INTRODUCTION
Nonalcoholic fatty liver disease (NAFLD) was initially described as "a less known and until now an anonymous hepatic disease." 1 Our understanding of this condition has quickly grown during the last few decades. Without excessive alcohol consumption or viral hepatitis, NAFLD is now defined as the accumulation of macrovesicular steatosis in more than 5% of hepatic cells. 2 This disease includes a spectrum ranging from steatosis to nonalcoholic steatohepatitis, which can lead to liver fibrosis and cirrhosis. 3 NAFLD is the most common cause of chronic liver dysfunction. NAFLD affects 25.24% of the population overall and has links to metabolic risk factors such as diabetes, hypertension, dyslipidemia, and obesity, according to a meta-analysis study including 22 countries. 4,5 The prevalence of NAFLD has gradually increased in Asia due to lifestyle changes. 6 Furthermore, NAFLD is an emerging health problem in Iran, with a total prevalence of 33.9%. 7 The time between ventricular depolarization and repolarization in the ECG is known as the QT interval. It is measured as the time from the start of the QRS complex to the end of the T wave. Due to the effect of heart rate (HR) on QT interval, a corrected QT interval (QTc) adjusted for the HR is mainly used. 8 Prolonged QTc interval in ECG is a well-known risk factor for ventricular arrhythmia, tachycardia, and sudden cardiac death (SCD) in people with cardiovascular risk factors and otherwise healthy populations. [9][10][11] Notably, the duration of the QT interval, even within a range of reference, predicts cardiovascular death in the general population. 10,12 The association between NAFLD and an increased risk of cardiovascular mortality has been reported in previous studies. 13 The QTc interval prolongation, which can result in ventricular arrhythmia and SCD, could partly explain this increased risk in patients with NAFLD. 9,10 Even though previous studies have reported the association between NAFLD and prolonged QTc interval, most have been conducted in Western or far East specific populations such as those with chronic medical conditions, including type 2 diabetes mellitus, women, or steelworkers.
For diagnosis of NAFLD, all similar studies used abdominal ultrasonography, which is operator dependent, qualitative, and may be less accessible than Fatty Liver Index (FLI) in health centers. 14,15 Moreover, factors such as underlying diseases, ethnicity, lifestyle, and diet might influence the results, and FLI is a quantitative, easy-to-employ 14 parameter for diagnosing NAFLD. Therefore, using the FLI score, we evaluated the association between NAFLD and prolonged QTc interval among the generally healthy Iranian population.

| Study population
The Fasa Cohort Study (FACS) enrolled 10 138 patients aged between 35 and 70 from October 2014 to September 2016. Details of the study's design and methods are described elsewhere. 16 People were invited by healthcare experts in the rural health system of Sheshdeh district and its 24 surrounding villages to participate in this study. 16 At the start of the study, all participants provided written informed consent. Our sample included all subjects with available ECG data (n = 9059). Subjects who met one or more of the following criteria were excluded (n = 4456): missing data, alcohol intake of 30 g/day or more in men and 20 g/day or more in women (safety levels of consumption to prevent Alcoholic Liver Disease), 17 having underlying diseases and reported taking medication with a known or potential risk for QTc prolongation. 18 Finally, the analysis included a total of 4603 subjects. Figure 1 summarizes the study enrollment flowchart with more details.

| Data acquisition
Each participant completed an internet-based questionnaire that collected basic information such as age, gender, smoking status, history of alcohol consumption, and chronic diseases such as diabetes and hypertension.
All medication taken within the past 2 weeks before registration was asked about and recorded. According to CredibleMeds, a database of drugs with a risk of QT prolongation, drugs with a known or potential risk for torsades de pointes have been listed and used to identify the taken medication with influential QTc prolongation factors. 18 A 20-item questionnaire based on the International Physical Activity Questionnaire (IPAQ) 16,19 was used to assess rural Iranians' routine physical activity.
The MET-value of each activity was multiplied by its duration to calculate MET-min. Total physical activity (MET/24 h) was the sum of all activities. The average daily intake of 125 food parameters was measured per a validated quantitative Food Frequency Questionnaire (FFQ) 16,20 for energy intake assessment, and the results were reported in kcal/day.

| Measurements
For anthropometric calculations, height was measured by a stadiometer, and a digital scale measured weight with an accuracy of 0.1 cm and 0.1 kg, respectively. Body mass index (BMI) was calculated as weight in kilograms divided by the square of the body height in meters (kg/m 2 ). Waist circumference (WC) was measured by a tape measure at the midpoint of the inferior border of the lowest rib to the anterior superior iliac spine.
After a 15-min rest, participants' blood pressure was measured while seated, using the individuals' right arm consecutively two times with an interval of 5 min. Then the average systolic and diastolic blood pressure was reported in mmHg.

| Blood chemistry
After fasting for 10-14 h, all participants had the following blood

| Electrocardiogram
An expert technician obtained a 12-lead ECG from each participant after they rested for 15 min. A computer-based device (Cardiax ® ) performed ECGs. Participants were asked to shave the precordium area and refrain from moving or talking for better electrode attachment. The computer software (Cardiax ® , version 3.50.2; International Medical Equipment Developing Co. Ltd.) interpreted and reported the ECGs, then exported them to the central data collection software. We used ECG data that includes HR (bpm), QT interval duration (ms), and QTc interval duration (ms), which were approved by a cardiologist. 16 The QT interval (QTc) was calculated using Bazett's formula (QTc = QT/(RR) 1/2 ). 21 Cut-off values for prolonged QTc were considered >430 ms for men and >450 ms for women. 10 2.6 | Definition of NAFLD FLI, a cost-effective, noninvasive method, was used to determine participants with or without NAFLD as a standard predictor of hepatic steatosis severity. This score is calculated using BMI, serum TG levels, WC, and GGT by the following formula 14 :  22 Therefore, men with a score of more than 46.9 and women with a score of more than 53.8 were assumed to have NAFLD.

| RESULTS
Four thousandsix hundred and three participants were studied, with 2095 (45.5%) males and 2508 (54.5%) females. The mean age was 47.65 years old (SD 9.10 years). Table 1 shows the clinical characteristics of the study population based on the presence of NAFLD. One thousand five hundred and fifty participants (33.6%) met NAFLD criteria based on the FLI. Systolic blood pressure, diastolic blood pressure, WC, BMI, FBS, TG, LDL, GGT, total cholesterol, AST, and ALT levels were significantly higher in NAFLD participants than in non-NAFLD participants (p < .001). Women comprised more than half of the NAFLD participants (56.9%). The prevalence of diabetes and hypertension was higher, while smoking was lower in NAFLD participants than non-NAFLD participants (p < .001). The amount of physical activity was significantly lower in subjects who had NAFLD (p < .001). As shown in Table 1, there is no statistically significant difference between participants with and without NAFLD in terms of age, energy intake (kcal/day), and history of alcohol consumption less than 30 g/day in men and 20 g/day in women (p > .05).

| DISCUSSION
We investigated the association between NAFLD and prolonged QTc interval in a large cohort of a generally healthy population in southern Iran. After controlling for a wide range of potentially confounding variables, our study found a positive relationship between NAFLD and prolonged QTc interval in the aimed population.
Our findings are consistent with the previous studies investigating the association between NAFLD and QTc prolongation. [23][24][25][26] According to a study by Targhar et al. 23