Determining the effects of respiratory load on neural respiratory drive (NRD) is important to understanding load compensation in respiratory disease. The oesophageal diaphragm electromyogram (EMGdi) is both sensitive and reproducible in quantifying NRD in healthy subjects and patients. EMGdi provides a measure of load on the respiratory muscles, accurately reflecting NRD in lung disease when transduction of NRD to intrathoracic pressure, respiratory flow and ventilation is impaired. The parasternal intercostal muscles are obligate inspiratory muscles, contracting in synchrony with the diaphragm. Diaphragm and parasternal intercostal muscle recruitment patterns have been studied in human subjects, but there are few data available when ventilation is increased above resting. Surface EMG of the parasternal intercostal muscles (EMGpara) potentially provides an alternative, non-invasive method to quantify NRD in health and disease. The aim of the study was to compare recruitment patterns of the diaphragm and parasternal intercostal muscles during acute hypercapnia. Eight healthy subjects (median (range) age 31 (27-38) years) underwent CO2 rebreathing to the limit of tolerance. The circuit consisted of a rebreathing bag containing 100% O2 attached to a two way valve and pneumotachograph. Carbon dioxide was continuously monitored using a capnograph sampling from the mouthpiece. EMGdi was measured using a multipair oesophageal catheter and EMGpara recorded from bipolar surface electrodes positioned in the second intercostal space. The root mean square (RMS) of EMGdi and EMGpara was calculated and peak RMS per breath measured. Subjects performed 2 min of voluntary hyperventilation on air prior to rebreathing. EMGdi, EMGpara and ventilation were analysed breath-by-breath. EMGdi and EMGpara were expressed as a percentage of baseline measured while breathing air at rest (EMGdi%BL, EMGpara%BL). Relationships between variables were examined when ventilation was established on the steep linear portion of the CO2 response curve. EMGdi%BL, EMGpara%BL and ventilation increased significantly reaching median (range) values of 588% (231-1220) p<0.05, 586% (182-730) p<0.05, 109l/min (64-358) p<0.05 respectively for the final five breaths of rebreathing. EMGdi%BL and EMGpara%BL increased in parallel in all subjects (median (range) r=0.82, (0.68-0.88) all p< 0.01). There was no difference between the median slopes of the EMGdi 3.24 (0.88-9.01) and EMGpara 4.13 (0.77-6.04) vs ventilation relationships during rebreathing. EMGdi%BL was numerically greater in 4/8 subjects. Bland Altman analysis indicated a mean (SD) bias of 0.086 (2.49) %EMG/L/min between measurements of NRD. EMGpara provides a non-invasive technique to quantify NRD, however measures using EMGpara and EMGdi may not be interchangeable in individual subjects.