Airway pressure alters wavelet fractal dynamics and short-range dependence of respiratory variability

Abstract

We sought to quantify the autocorrelation properties and fractal dynamics of interbreath interval (IBI) variability to see if short-term airway pressure (Paw) increases affected scaling behaviour. Data were collected in anaesthetised children at basal (Ba), medium (Me) and high (Hi) Paw. Consistent with short-range dependence, the autocorrelation function was significant at five (Ba) to three (Hi) lags but lost its significance at Me. Wavelet variance assessed in IBI suggests biphasic power law correlations with scale. In the low scale window for the crossover point, the positive exponents differed from the uncorrelated components in surrogate series at Ba (p<0.007) and Hi (p<0.002), supporting the existence of deterministic, persistent properties. On the contrary, white noise output was captured at Me and an antipersistent, random pattern characterised the high scale window. Intermediate Paw seems to reset the system to a stochastic, less complex behaviour while basal and high Paw would amplify the gain of non-random, correlated processes. This Paw-dependent modulation of IBI fractality and short-term memory indicates that integrative breathing control has an inherent multi-scaling structure.