Nonlinear measures of heart beat intervals differ in female patients with chest pain

Abstract Objective: In many female patients with chest pain, false positive exercise electrocardiography (ECG) test results have been observed in clinical setting. We hypothesized that linear and nonlinear measures of heart beat interval might differ between female patients with chest pain that was related to coronary artery disease and those with chest pain that was not related to it. Methods and results: A prospective, cohort study at the setting of primary care and cardiology referral practice in Japan was performed. The subjects were 89 out of 126 consecutive female patients who had chest pain with positive exercise ECG test. Clinical data (including age, post-menopausal state and the number of risk factors for cardiovascular disease), linear heart beat analysis, and nonlinear heart beat dynamics by 24-h electrocardiogram were compared between female patients showing chest pain in significant coronary artery lesions and those without. With respect to clinical data, post-menopausal state was associated with the presence of coronary artery disease. A ratio between low- and high-frequency power of less than 3.0 in linear heart beat analysis and approximate entropy of less than 1.4 in nonlinear measurement were associated with the presence of coronary artery disease. In ROC curve, approximate entropy increased the ability of clinical data to predict coronary artery disease. Conclusions: We concluded that nonlinear measure of approximate entropy as a predictor of coronary angiography findings might be important in the management of female patients having chest pain with positive exercise ECG test results.