Correlation between short-term heart rate variability and QT-Interval with left ventricular systolic function in patients with stable Coronary Artery Disease (CAD): a cross-sectional study

ABSTRACT


INTRODUCTION
Coronary Artery Disease (CAD) is a pathological process caused by plaque buildup in the arterial walls that supply blood to the heart and other parts of the body. 1 CAD remains the leading cause of death worldwide, with an estimated annual count of 17.9 million globally. 2atients with stable CAD have plaque destabilization, which leads to Acute Coronary Syndrome (ACS). 2 Stable CAD patients with chronic myocardial ischemia and hypoxia may suffer dysfunction and imbalance of autonomic nerve regulation.Heart rate is controlled by a balance between the sympathetic nervous system and the parasympathetic nervous system. 3Heart Rate Variability (HRV) is the physiological phenomenon of variation in the time interval between heartbeats and can be used to assess the autonomic function of the heart. 3It is measured by the variation of the R-R electrocardiography wave interval.
Various cardiovascular diseases can be detected using the widely used Electrocardiogram (ECG).QT interval is defined as the duration from the beginning of the QRS complex to the end of the T wave. 4 QT interval can become a determinant of mortality in the general population. 5In one study, mortality rates associated with QT interval values had a "U-shape" curve. 5Medium (420 to 440 milliseconds) and extensive (> 440 milliseconds) interval lengthening increased mortality by 1.5 and 1.7-fold.Another study showed that healthy middleaged men with QT intervals greater than 420 ms had 4.3 times the risk of death

ORIGINAL ARTICLE
or LV peak GLS) as the outcome was carried out through Pearson or Spearman's correlation and linear regression analysis.The data analysis was conducted with the statistical software application IBM SPSS version 21 for Windows.

RESULTS
A total of thirty-three stable coronary artery disease patients were included in this study.Most of the subjects were men (70% vs. 30%) (Table 1).The mean age was 62.3 ± 8.9 years.Almost all but three people (91%) had hypertension.One-third of the patients had diabetes.The mean routine blood laboratory, renal function, and lipid profile were within normal limits.
The distributions of the short-term heart rate variability variables were skewed to the right.However, the QT interval (included QTc by Bazett and Fridericia) and left ventricular systolic function are normally distributed.The median time of SDNN, SDANN, and rMMSD was 34.8 ms, 16.2 ms, and 23.4 ms, respectively.Meanwhile, pNN50 and triangular index medians were 1% and 18.0, respectively.The VLF, LF, and HF median were 1135.8,289.8, and 172.9 ms 2 , respectively.The mean of QT interval, QTc Bazett, and QTc Fridericia were 412.2, 421.0, and 417.8 ms.The mean peak of GLS was -20.9%, with a variation of about 3%.The mean ejection fraction was 68.3%, with a variation of about 4% (Table 1).
The correlation between peak GLS and ejection fraction was weak (Pearson's r = 0.016, p=0.929).Non-parametric correlation analysis between sHRV variables and left ventricular systolic function showed no significant correlation.Table 2 presents the correlation between the two variables of left ventricular systolic function, peak GLS and ejection fraction (EF), with QT interval.There was a significant correlation between QTc interval and LV peak GLS.The relationship between LV peak GLS and QTc Bazett interval was weak (Pearson's r = 0.35; p = 0.046).However, we found a significant moderate relationship between LV peak GLS and QTc Fridericia (Pearson's r = 0.426; p = 0.013) (Table 2).
Table 3 provides the univariate linear regression analysis result between QTc interval and LV peak GLS.QTc interval correlated negatively with the absolute value of LV peak GLS.The regression estimated that each 10 ms QTc Bazett prolongation reduces 0.36 points of LV peak GLS percentage (B=0.036;β=0.35;R 2 =0.122; p=0.046).Furthermore, the regression also estimated that each 10 ms QTc Fridericia prolongation reduces 0.52 point of the absolute value of LV peak GLS percentage (B=0.052;β=0.426;R 2 =0.181; p=0.013).Scatter diagrams with the regression line are provided in Figure 1.

DISCUSSION
This study is the first study to investigate the correlation between short-term HRV parameters and QT interval with left ventricular systolic function in the stable CAD population in Indonesia, as several studies focused more on populations with acute coronary syndromes.This study involved 33 patients with stable coronary artery disease.These study subjects' most commonly found risk factors were obesity and hypertension.
The correlation analysis between ejection fraction and GLS peak showed a weak correlation, indicating that changes in the LV GLS peak occurred in patients with normal ejection fraction.This is consistent with the findings of Radwan H et al., demonstrating that changes in peak GLS might have occurred in patients with normal left ventricular ejection fraction. 8he results of HRV parameters in this study, especially SDNN, SDANN, pNN50, triangular index of the time domain, did show a lower value when compared to their normal values. 9In particular, SDNN should be the focus because it has a predictive role in morbidity and mortality. 14The median SDNN in this study was 34.8 ms, which Kleiger RE et al., classified that SDNN below 50 ms had a worse prognosis in postmyocardial infarction patients. 15However, it indicated that the stable coronary artery disease patients in this study also had poor SDNN values.
Non-parametric correlation analysis between sHRV variables and left from coronary heart disease than those with QT intervals < 385 ms. 6The corrected QT interval formula such as Fridericia and Bazett are widely used in clinical practice.Bazett's corrected procedure provided an adequate correction for heart rates ranging from 60-100 bpm.The Fridericia formula should be used for heart rates of less than 60 or more than 100 bpm as it is proven to be more accurate. 7o our knowledge, no studies have assessed the correlation between HRV and QT intervals to left ventricular systolic function in a stable CAD population.Based on those mentioned above, this study investigated the correlation between short-term HRV (sHRV) parameters and QT interval with left ventricular systolic function in stable coronary artery disease patients.

METHODS
This cross-sectional study was conducted in Prof. Dr. R. D. Kandou Hospital, Manado, from March to September 2021 by consecutive technique sampling.The study subjects were patients with stable angina pectoris who visited the Cardiovascular Clinic Department.The inclusion criteria were patients with sinus rhythm with no previous history of coronary revascularization, not on antiarrhythmic drugs (other than betablocker), potential drugs which cause QTprolongation, and pacemaker.The subjects underwent clinical history, physical, and further diagnostic examinations (clinical history, cardiac biomarkers, and ECG) to define significant CAD.Patients with acute heart failure, acute coronary syndrome requiring revascularization, ventricular or supraventricular arrhythmias on the ECG recording at rest, left bundle branch block or ST-segment deviation > 1 mm on the ECG, atrial fibrillation, a history of myocardial infarction, abnormal regional cardiac wall motion, aortic stenosis, hypertrophic cardiomyopathy, and uncooperative patients were excluded.Setting the type I and II errors to be 5 and 20%, respectively, a minimum of 32 patients was obtained from the sample size calculation.
The relationship between sHRV and QT interval with left ventricular systolic function variables (LV ejection fraction ventricular systolic function showed no significant correlation.This is in accordance with a study by Shehab et al., which found no relationship between HRV and LV ejection fraction. 16Tran N et al. found changes in the frequencydomain HRV and an increase in left ventricular strain in patients undergoing transcutaneous vagal nerve stimulation. 17arson's correlation test between the corrected QT interval (both Bazett and Fridericia) and LV peak GLS showed a weak to moderate positive correlation, despite no significant correlation with LV EF.QT correction with Fridericia's formula showed a stronger relationship compared to Bazett's.In univariate regression analysis, for each 10 ms QTc Bazett or Fridericia prolongation reduce 0.36 or 0.52 point of the absolute value of LV peak GLS percentage.
QT prolongation in patients with left ventricular dysfunction has been widely reported, but the underlying mechanism remains unclear.A study using data from the Euro Heart Failure survey showed that QT interval prolongation was associated with the presence of left ventricular systolic dysfunction, but this association was due to a prolongation of the QRS duration. 18In a study conducted by Davey P in the heart failure population, QTc interval was correlated with left ventricular systolic function measured using fractional shortening. 19QT interval prolongation and QT dispersion in the heart failure population also worse prognoses. 20In the stable coronary artery disease population as in this study, the QTc interval can also be used to predict early systolic dysfunction characterized by abnormal LV peak GLS, despite having normal LVEF and normal global cardiac wall motion.A possible underlying mechanism for QT prolongation in stable angina pectoris patients is the presence of relative ischemia in certain areas of the myocardium that activate the Na + /H + pump to prevent intracellular acidosis. 21,22

CONCLUSION
There was no statistically significant correlation between short-term HRV with peak GLS and ejection fraction, indicating that the parameters were not recommended to illustrate the left ventricle systolic function in patients with stable coronary artery disease.There was