Introduction Severe asthma affects airway calibre and can be monitored using IOS. Sample entropy (SampEn) is a measure of complexity and is defined as the probability that sequences of patterns (template size) in time series which are initially closely related, that is, within a fraction of the standard deviation (tolerance level) of the time-series remain so within subsequent time frames. Fractal scaling is a measure of self similarity and scale invariance measured in a time-series and quantifies the memory found within as a consequence. We hypothesised that fractal scaling and SampEn will be useful in characterising severe asthma.
Methods 66 GINAstage 4–5 severe asthmatics (Mean(Sem) age; 54.1(1.4), Sex M:F; 31:35, post-bronchodilator FEV1% predicted; 81.02 (2.7)%) and 27 Controls (Mean(Sem) age; 48.4(2.2), Sex M:F;9:18, post-bronchodilator FEV1% predicted; 108.2 (2.8)%) were recruited. Impulse oscillometry was performed at 5–35 Hz, with impulses triggered every 0.2 s for 150 s, at baseline and 15 min after 400 mcg inhaled salbutamol. Detrended fluctuation analysis was used to derive the fractal scaling exponent α1. SampEn was derived using a custom program. SampEn and α1 were both obtained from airway resistance at 10 Hz over the 150 s time-series. Triplicate measurements of 150 s were repeated in 18 randomly selected asthmatics from our cohort after 6 months.
Results SampEn was significantly increased compared to controls (Abstract P122 Table 1) and correlated significantly with exacerbation frequency from the previous 12 months (Asthma Baseline- p=0.007, rs=0.3; Post-Bronchodilator- p=0.009, rs=0.3). Fractal scaling was also found to be present in airway resistance in severe asthma (α1=0.94 (0.03)) and showed an inverse relationship with SampEn (p=0.0352, r=−0.4). Increased SampEn was associated with worse ACQ scores (p=0.027, rs=0.3) and lower AQLQ scores (p=0.023, rs=−0.2). SampEn measurements were repeatable (an Intra-class correlation of 0.74) in the triplicate series. In keeping with other studies, airway resistance was significantly increased in severe asthma.
Conclusions SampEn a measure of complexity is (1) increased in severe asthma (2) a repeatable measure (3) associated with a lower quality of life and exacerbation frequency. The ability of this technique to monitor asthma stability and to predict future exacerbations by stochastic modelling needs to be explored.