Parasympathetic function during deep breathing in the general population: relation to coronary risk factors and normal range

OBJECTIVES: To examine the association between the parasympathetic function assessed by deep breathing-induced heart rate variability (HRV) and coronary risk factors and to establish a reference range for deep breathing-induced HRV. DESIGN: Cross-sectional population-based study. SETTING: The municipality of Horsens, Denmark. SUBJECTS: One hundred and ninety-four individuals aged 40-67 years without diabetes, atrial fibrillation or a pacemaker randomly selected from the population. MAIN OUTCOME MEASURES: During deep breathing at 6 respiratory cycles min-1, an ECG was taken and in three consecutive cycles the longest R-R interval during expiration (E) and the shortest during inspiration (I) were selected. The mean ratio (E/Iratio) and the mean difference in instantaneous heart rate (E/Idiff) were taken as expressions of the parasympathetic function. RESULTS: In multivariate analysis, E/Idiff was reduced with increasing age (P < 0.0005) and left ventricular mass (P = 0.008), with ECG sign of probable previous myocardial infarction (P = 0.020) and the use of cardiac medication (P = 0.018) and positively correlated to heart rate (P = 0.030). The E/Iratio was diminished with increasing age (P = 0.001), left ventricular mass (P = 0.003), waist-hip ratio (P = 0.044), with ECG sign of probable previous myocardial infarction (P = 0.012) and use of cardiac medication (P = 0.020), but no association was found with heart rate (P = 0.92). In both E/Idiff and E/Iratio, no correlation was found to lipids, blood pressure or alcohol consumption. In the group not on cardiac medication, without left ventricular hypertrophy or ECG sign of probable myocardial infarction, E/Idiff and E/Iratio were still independently correlated to age and left ventricular mass. In this group, equations defining the age-corrected 5th percentile were calculated. CONCLUSIONS: The parasympathetic function as assessed by deep breathing-induced HRV in the general population is reduced in older people, and in individuals on cardiac medication, with left ventricular hypertrophy or ECG signs of myocardial infarction. Even in healthy persons the parasympathetic function is inversely associated with age and left ventricular mass. Values of E/Idiff above (4.39-0.033 x age)2 and readings of E/Iratio above 1 + exp (-1.12-0.0198 x age) can be regarded as normal.