Evaluation of the Effect on Heart Rate Variability of Some Agents Acting at the β-Adrenoceptor using Nonlinear Scatterplot and Sequence Methods

Abstract

There is evidence that the processes regulating heart rate variability (HRV) reflect nonlinear complexity and show “chaotic” determinism. Data analyses using nonlinear methods may therefore reveal patterns not apparent with the standard methods for HRV analysis. We have consequently used two nonlinear methods, the Poincaré plot (scatterplot) and cardiac sequence (quadrant) analysis, in addition to the standard time-domain summary statistics, during a normal volunteer investigation of the effects on HRV of some agents acting at the cardiac beta-adrenoceptor. Under double-blind and randomized conditions (Latin square design), 25 normal volunteers received placebo, salbutamol 8 mg (β2-adrenoceptor partial agonist), pindolol 10 mg (β2-adrenoceptor partial agonist), or atenolol 50 mg (β1-adrenoceptor antagonist). Single oral doses of medication (at weekly intervals) were administered at 22:30 hours, with sleeping heart rates recorded overnight. The long-term (SDNN, SDANN) and short-term (rMSSD) time-domain summary statistics were reduced by salbutamol 8 mg and increased by atenolol 50 mg compared with placebo. The reductions in both SDNN and SDANN were greater after salbutamol 8 mg compared with pindolol 10 mg. The reduced HRV after pindolol 10 mg differed from the increased HRV following atenolol 50 mg. The Poincaré plot, constructed by plotting each RR interval against the preceding RR interval, was measured using a reproducible computerized method. Scatterplot length and area were reduced by salbutamol 8 mg and increased by atenolol 50 mg compared with placebo; scatterplot length and area were lower after pindolol 10 mg compared with atenolol 50 mg. Geometric analysis of the scatterplots allowed width assessment (i.e., dispersion) at fixed RR intervals. At the higher percentiles (i.e., 90% of scatterplot length: low HR), salbutamol 8 mg reduced and atenolol 50 mg increased dispersion; at lower percentiles (i.e., 10%, 25%, and 50% length), atenolol 50 mg and pindolol 10 mg increased dispersion compared with placebo and salbutamol 8 mg. Cardiac sequence analysis (differences between three adjacent beats; ΔRR vs. ΔRRn+1) was used to assess the short-term patterns of cardiac acceleration and deceleration. Four patterns were identified: +/+ (a lengthening sequencing), +/− or −/+ (balanced sequences), and finally −/− (a shortening sequence). Cardiac acceleration episodes (i.e., number of times ΔRR and ΔRRn+1 were both changed) were increased in quadrants −/− and +/+ following pindolol 10 mg and salbutamol 8 mg; the beat-to-beat difference (ΔRRn+1) was reduced after salbutamol 8 mg compared with the three other groups. These results demonstrated a shift towards sympathetic dominance (β-adrenoceptor partial agonist salbutamol 8 mg) or parasympathetic dominance (β1-adrenoceptor antagonist atenolol 50 mg); pindolol 10 mg exhibited HR-dependent effects, reducing HRV at low but increasing variability at high prevailing heart rates. These nonlinear methods appear to be valuable tools to investigate HRV in health and to study the implications of perturbation of HRV with drug therapy in disease states.