Electrocardiographic Profiles and Associated Factors in Ghanaians with Type 2 Diabetes

Abstract

Background: Globally, type 2 diabetes (T2D) remains highly prevalent among adults and is closely linked to cardiovascular diseases, including coronary artery disease (CAD). Individuals with T2D face a twofold increased risk of developing CAD compared to the general population. CAD significantly raises the likelihood of fatal complications, such as heart failure, myocardial infarction, and sudden cardiac death. Electrocardiography (ECG) serves as a valuable tool for identifying individuals with CAD. For individuals with type 2 diabetes, routine resting electrocardiograms are advised by the American Diabetes Association and the European Society of Cardiology/European Association for the Study of Diabetes. However, the benefits of these routine screenings have not been well established in many populations, including Ghanaians with T2D, due to a lack of data on the prevalence and patterns of ECG abnormalities in these groups. General Aim: To determine the prevalence of resting ECG abnormalities and their associations with cardiovascular risk factors and current/prior adverse cardiac events in Ghanaian adults with T2D. Methodology: In a cross-sectional design, resting standard 12-lead supine digital resting electrocardiography (ECG) testing was conducted on 430 Ghanaian adults with T2D managed at the National Diabetes Management and Research Centre at the Korle-Bu Teaching Hospital in Accra. A standard 12-lead supine digital resting electrocardiography (ECG) was recorded using the Vitalograph BT12 ECG™ and its Spirotrac® Software (Vitalograph, United Kingdom). The ECG tracings were processed with the device’s inbuilt University of Glasgow 12-Lead ECG Analysis Algorithm1. Two additional methods were used to evaluate each ECG: Minnesota coding and a trained physician's assessment. The ECG abnormalities were classified into seven main groups namely 1) QS pattern abnormalities, 2) QRS-axis deviation abnormalities; 3) R-waves abnormalities; 4) ST-segment abnormalities; 5) atrioventricular conduction defects; 6) ventricular conduction defects; and 7) arrhythmias including atrial flutter, atrial fibrillation and ventricular contractions or premature atrial. The identified ECG abnormalities were further categorized as either minor or major abnormalities. The prevalence of each abnormality was determined. Univariate logistic regression analyses were performed to assess the associations of potential predictor variables with the various pooled abnormality categories. Following this, a multivariate logistic regression model was built to assess predictors of each abnormal ECG category. The odds (OR) and 95% confidence interval (95% CI) were determined. A statistical test of significance was set at p-value<0.05. Results: In total, 61.6% of the study participants had abnormal ECG (23.9% had major abnormalities and 37.7% had minor abnormalities). Overall, 73.9% of the study participants were women. The mean (±standard deviation) age, BMI, systolic BP, diastolic BP, duration of diabetes, HbA1c, and total cholesterol concentrations were 61.87 (±10.09) years, 29.46 (±5.81) kg/m2, 135.33 (±18.50) mmHg, 75.56(±11.20). 18.03 (±10.01) years, 8.74 (± 2.25) % and 4.96 (± 1.45) mmoll/L. The proportion of smokers, individuals on statin therapy, and individuals who used insulin were 4.9%), 39.8%), and 36.3%, respectively. Compared to the normal ECG, individuals in the abnormal ECG group were significantly older (62.85 vs 60.30, p=0.011). Aside age, the two groups did not differ in any of the baseline characteristics. In the study population, the mean heart rate was 73 beats per minute. The mean RR interval, PR interval, PQ interval, ST segment duration, QT interval, and corrected QT interval were respectively 839.24 ms, 167.13 ms, 166.11 ms, 114.99, 399.86 ms, and 422.36 ms respectively. The mean P wave and QRS complex widths were 114.83 ms and 94.97 ms respectively. The mean pooled p-axis, QRS-axis, and T-axis were 114.83°, 32.46°, and 32.60°, respectively. While the mean heart rate did not significantly differ between the two groups, the mean RR interval was 4% longer in the abnormal ECG group compared to the normal ECG group (853.94 ms vs. 818.54 ms, p = 0.020). Similarly, the mean PQ interval was 4.5% longer in individuals with abnormal ECG compared to those with normal ECG (169.21 ms vs. 161.83 ms, p = 0.018). The mean QT interval was 2% longer in participants with abnormal ECG compared to those with normal ECG (402.63 ms vs. 394.40 ms, p = 0.015). The mean QRS axis also differed significantly between the two groups (29.56° in the abnormal ECG group vs. 36.26° in the normal ECG group, p = 0.037). Additionally, the mean T-axis showed a significant difference between the groups (29.00° in the abnormal ECG group vs. 36.20° in the normal ECG group, p = 0.045). Among participants who reported symptoms of CAD, the majority (16.3%) reported angina symptoms; 3.0% of the study population reported symptoms of myocardial infarction. The proportion of individuals with CAD symptoms was higher in the abnormal ECG group compared with the normal ECG group (22.3% vs 9.7%, p=0.001). This was driven largely by differences in angina symptoms (20.8% vs 9.1%, p=0.001). In the multivariable regression model, hypertension was independently associated with QS pattern anomalies [odds ratio 1.65, 95% confidence interval 1.21 – 1.88, p= 0.049], very tall amplitude R-waves [1.31 (1.07 – 1.82), 0.041], ST-segment abnormalities [1.64 (1.31 – 1.93), 0.033], and ventricular conduction disorders [1.68 (1.32 – 1.99), 0.042]. HbA1c concentration > 7% was associated with abnormal Q-waves [1.27 (1.11 – 1.37), 0.032] while obesity was associated with atrioventricular conduction abnormalities [1.77 (1.36 – 2.17), 0.022] Conclusion: This study shows a high burden of ECG abnormalities in Ghanaians with T2D. Which correlated with angina symptoms. Hypertension was the CVD risk factor most frequently independently associated with specific ECG abnormalities in our study population, including QS pattern anomalies, very tall amplitude R-waves, ST-segment abnormalities, and ventricular conduction disorders. Poor glycemic control and obesity were also independently associated with abnormal Q-waves and atrioventricular conduction abnormalities, respectively. These findings could inform the institution of frequent ECG testing in select groups of Ghanaians with T2D, to aid in the prevention of adverse CVD events.