Titration of continuous positive airway pressure by the pattern of breathing: analysis of flow-volume-time relationships by a noninvasive computerized system.

Infants can defend or even dynamically elevate their functional residual capacity with additional respiratory muscle work by retarding early expiratory airflow (V) with postinspiration inspiratory muscle activity and/or laryngeal narrowing, or by starting inspiration before expiration to the relaxation volume has been completed. In order to study the effect of continuous positive airway pressure (CPAP) on both phenomena in 23 infants (birthweight 1,746 +/- 417 g), we elevated the airway pressure stepwise in 0.2 kPa increments. A computerized bedside flow-volume (V/V) analysis was used for evaluation. In 16 "responders" early expiration braking decreased and "premature inspiratory interruption" was postponed at an "appropriate CPAP level." The linear segment (relaxation line) of the V/V-loop was lengthened until expiratory time reached a maximum. Elevation of CPAP beyond this level again produced a rapid, shallow pattern, often combined with flow acceleration late in expiration (recruitment of expiratory muscles). In the remaining seven infants (non-responders) these latter signs of excessive airway pressure already occurred at the lowest CPAP levels applied during the "titration trials." Respiratory rate without CPAP was different between responders (84 +/- 17/min) and non-responders (46 +/- 17/min). This approach for determining the appropriate CPAP level might reduce the risk of respiratory muscle fatigue.

An infant ventilator technique for resistive unloading during spontaneous breathing. Results in a rabbit model of airway obstruction.

The combined system of ventilator circuit, endotracheal tube, and lung commonly imposes a resistive load on spontaneous breathing efforts. It is possible to compensate for this positive resistance by a device generating a "negative ventilator resistance" (NVR), i.e. delivering a positive pressure during inspiration and a negative pressure during expiration in constant proportion to the instantaneous flow of the spontaneous breathing. The concept of NVR implies that there must not be any phase lag between flow and pressure signals. In eight anesthetized, intubated, spontaneously breathing rabbits (mean body wt 3570 g, range 2900-4600 g), challenged either by aerosolized histamine or an extrapulmonary resistive load, lung mechanical data were calculated from esophageal pressure and flow signals. Each animal served as its own control with and without NVR. In a total of 39 experiments, NVR was applied in amounts between 1 and 15 kPa.s/L. During both types of additional resistive load, NVR immediately reduced the resistive work of breathing. There was a strong linear correlation between the amount of NVR applied and the decrease in total resistance, where the total resistance equals the resistive load on the animal's respiratory muscles (sum of the resistances of all components of the combined respirator-lung system): r = 0.93, p less than 0.001. The relationship between NVR and the drop in resistive work per mL of tidal volume was similar: r = 0.85, p less than 0.001. Throughout the experiments, NVR operated in perfect synchronization with the animal's spontaneous breathing activity.

High frequency oscillatory and conventional mechanical ventilation in experimental surfactant deficiency: a study using a new infant ventilator technique.

The performance of a new infant ventilator system had to be evaluated. Technically it is characterized by flow (V)- and pressure (P)-transducers mounted immediately near the endotracheal tube. A microcomputer works as a function generator and governs servo-controllers for V and P thus offering a multiplicity of different modes both of the conventional (CMV) and high frequency oscillatory (HFO) type. The additional dead space imposed by the system is identical with its internal compressible volume of 2 ml. Serial pulmonary lavages were performed in 17 adult rabbits while on CMV. PaO2 per unit of mean airway pressure (MAP) decreased thereby from 95.9 +/- 29.3 to 9.0 +/- 6.7 (kPa/kPa). The animals were then alternately ventilated by HFO (5, 10, or 20 Hz) and CMV, at matched MAP's. No significant difference in PaO2 between the two methods was revealed in intra-animal comparisons except a slight superiority of CMV at MAP's above 1.7 kPa (P less than 0.05). There was no clear linear relationship between PaO2 and MAP both at CMV and HFO. A strong increase in PaO2 often occurred beyond a MAP threshold. In 37 postlavage HFO runs at 5 Hz in 13 animals volume amplitudes of 3.19 +/- 0.5 ml/kg of bodyweight resulted in PaCO2 levels of 6.29 +/- 1.87 kPa. Except in one experiment (10 Hz) volume amplitudes below the natural dead space produced arterial hypercapnia.