ABSTRACT
Objective: the primary purpose of the study was to determine the incidence of abnormal electrocardiograms (ECGs) which affected the changes of exercise programs. The secondary purpose was to study types of ECG abnormality and the time of detecting the abnormality. Methods: The study was retrospective design. The studied population was selected from the followed up patients at our outpatient cardiac rehabilitation clinic from October 1, 2008 to May 31, 2009. The inclusion criteria were patients who had to follow up at our outpatient cardiac rehabilitation clinic and were monitored for ECG telemetry. The medical and ECG telemetry records of the patients were reviewed. The incidence of abnormal ECGs was reported as the percentage of patients with abnormal ECGs. The patients with abnormal ECGs were classified as the ECG changing group and those without abnormal ECGs were classified as the non-ECG changing group. The comparison of both groups was performed by Chi-square test for categorical data and the independent samples t-test for the quantitative data. The median survival time was carefully estimated by Kaplan-Meier method of Survival Analysis. The factors associated with time of detecting abnormal ECGs were analyzed by using Cox proportional hazards model. The data analysis was performed with statistical significant difference of the p-value < 0.05. Results: Five hundred and forty patients, 378 males and 162 females, were enrolled. There were 151 from 540 patients (28%) in the ECG changing group and 389 patients (72%) in the non-ECG changing group. The comparison between two groups indicated that there were no significant differences regarding gender, age, body mass index, diseases and co-morbidities (such as coronary artery disease, cerebrovascular disease, diabetes mellitus, hypertension, dyslipidemia and chronic obstructive lung disease), left ventricular ejection fraction and status of cardiac surgery. The two most common types of ECG abnormality in the ECG changing group were tachyarrhythmia and bradyarrhythmia, respectively. The Cox proportional hazards model revealed that there was no factor associated with time of detecting abnormal ECGs. From the Kaplan-Meier method of Survival Analysis, the median survival time of detecting abnormal ECGs was 61 months (95%CI: 47.6, 74.9). Conclusion: The incidence of abnormal ECGs which affected the changes of exercise programs in our outpatient cardiac rehabilitation clinic was 28%. The most common type of abnormality was tachyarrhythmia. The median survival time for detecting abnormal ECGs was 61 months after cardiac hospital discharge. There was no associated factor with the time of detecting abnormal ECGs.