An IBM PC-based system for the assessment of cardio-respiratory function using oscillating inert gas forcing signals

Objective. An IBM PC-based real-time data acquisition, monitoring and analysis system was developed for the assessment of cardio-respiratory function, i.e. airway dead space, alveolar volume and pulmonary blood flow, using oscillating inert inspired gas forcing signals. Methods. The forcing gas mixture was generated by an in-house sinusoid gas delivery unit. The system interfaced with a mass spectrometer and an airway flow transducer, and performed real-time tracking of the breath-by-breath end-inspired, end-expired and mixed-expired concentrations. It calculated the cardio-respiratory parameters using two, i.e. continuous and tidal, in-house mathematical models of the lungs. The system's performance was evaluated using a mechanical bench lung, laboratory subjects and awake adults breathing spontaneously. Its predictive accuracy was compared with the measured volumes of the bench lung; single breath CO2 test for airway dead space and N2 washout for alveolar volume in laboratory subjects and awake adults; and thermal dilution technique for pulmonary blood flow in laboratory subjects. Results. Close agreements were found between the true and predicted airway dead space, i.e. mean differences of -2.39%, 14.47% and -17.49%, respectively, and that of alveolar volume, i.e. -8.03%, -3.62% and 7.22%, respectively, in the bench lung, laboratory subject and awake adult studies; and that of pulmonary blood flow (-23.81%) in the laboratory subjects using the continuous lung model. Even closer agreements were observed for airway dead space (-5.8%) and alveolar volume (-4.01%) of the bench lung and for pulmonary blood flow (-8.47%) in the laboratory subjects using the tidal lung model. Conclusions. A system was developed to deliver, monitor and analyse on-line, and in real-time, output data from the sinusoid forcing technique. The technique was administered using the system in various subjects, and produced favourable predictions.