The aim of this study was to determine if central GABA mechanisms are involved in the cardiac vagal withdrawal at the beginning of exercise in man. We tested whether GABA-enhancing effects of a benzodiazepine could be observed in the HR change (R-R interval) immediately following the onset of a brief (10s) isometric contraction (60 % maximum) of the biceps muscle. The difference between the change in R-R interval occurring during the same phase of respiration was compared for placebo (Pla) and 10 mg oral diazepam (Dz) treatment in a double blind, crossover trial. ECG, blood pressure, respiration and biceps muscle tension were recorded. The subjects breathed to a metronome and R-R interval measurements were plotted for early and late inspiration and early and late expiration. The mean values of the first late expiration R-R interval immediately following the start of contraction in early expiration were compared to the same measurements without contraction. Contractions initiated following diazepam treatment resulted in a significantly greater reduction in R-R interval (P < 0.05) implying that GABAergic suppression of cardiac vagal outflow may be responsible for contraction-induced tachycardia in man.

1. Animal studies show that cardiac vagal tone can be modified by gamma-aminobutyric acid neurons acting at several sites in the central nervous system. The present study has attempted to determine whether similar control exists in humans by using midazolam, a benzodiazepine. Benzodiazepines exert their main actions on the central nervous system by interacting co-operatively at the gamma-aminobutyric acid receptor. 2. Twenty patients took part in the study before undergoing cardiac catheterization. After resting for 20 min in a semi-supine position on a couch, ECG, blood pressure and respiration were recorded for 5-min periods with either controlled (fixed) or free respiration. During this time a baroreceptor sensitivity test was conducted. 3. Doses of 1 mg and 5 mg of midazolam were administered intravenously. 4. Five-minute segments of data, before and after midazolam, were subjected to power spectral and time-domain analysis. 5. Midazolam caused a decrease in the high-frequency and an increase in the low-frequency components of the power spectral density plot, and in addition reduced the mean R-R interval and R-R variability expressed as the interquartile difference, and pNN50. There were no significant changes in the sensitivity of the baroreflex or in the systolic, diastolic and average blood pressures. 6. This decrease in variability of heart period, particularly at a controlled respiratory frequency, strongly suggests that cardiac vagal tone in man can be regulated by gamma-aminobutyric acid neurons.

No abstract