Neural respiratory drive (NRD), measured using the parasternal intercostal muscle electromyogram (EMGpara), relates to lung disease severity as quantified by conventional methods in a range of diseases. Reference data from healthy populations are required for the technique to be used as an independent measure of lung disease severity. EMGpara has previously been expressed as a percentage of that obtained during a maximal inspiratory effort (EMGpara%max), restricting the use of the technique to subjects able to reliably perform such manoeuvres. The aim of this study was to investigate variability of both raw EMGpara (rEMGpara) and EMGpara%max in healthy adults.

EMGpara was measured during tidal breathing in 43 healthy adult non-smokers (25 females, median (range) age 32 (19–79)years, mean (SD) BMI 23.4 (3.5) kg/m²), using surface electrodes positioned bilaterally over the second interchondral space. Measurements were made with and without a mouthpiece/pneumotachograph in situ in 20 participants. Repeated measures were obtained within the same testing session in 27 subjects, and at least seven days later in 13 individuals. Spirometry, height, weight, BMI, fat free mass (FFM) via bioelectrical impedance and measures of regional fat distribution (waist/hip ratio and neck circumference) were also recorded.

Mean (SD) EMGpara%max and rEMGpara were 5.88 (3.63)% and 5.06 (2.26)µV respectively. Significant relationships were observed between anthropometric measures and rEMGpara and EMGpara%max (Table 1). rEMGpara and EMGpara%max were unrelated to spirometry variables. Median (range) rEMGpara and EMGpara%max increased significantly with the pneumotachograph in place (4.86 (2.11–8.19)µV versus 5.62 (2.47–10.98) µV and 4.77(1.68–17.00)% versus 6.78 (2.35–20.94)%, both p < 0.0001).

Analysis of variance by subject was used to assess within-subject variability. Measurement error was higher for EMGpara%max than rEMGpara (upper 95% confidence limit of difference between repeat measures of EMGpara%max 3.14%, versus 2.35 µV for rEMGpara; within-subject coefficient of variation EMGpara%max 30.8% versus rEMGpara 24.5%).

rEMGpara appears to be a reproducible marker of NRD. Both rEMGpara and EMGpara%max are influenced by subjects’ anthropometry. Further investigation is required to determine whether these influences are technical or physiological and must be considered when the technique is applied clinically or for research, or when developing reference values.