Reduction of oscillatory pressure along the endotracheal tube is indicative for maximal respiratory compliance during high-frequency oscillatory ventilation: a mathematical model study

We hypothesized that during high-frequency oscillatory ventilation (HFOV), a reduction of peak-to-peak oscillatory pressure along the endotracheal tub e is maximal when respiratory system compliance is maximal. We made a mathe matical model of the endotracheal tube and the respiratory system of a neon ate suffering from idiopathic respiratory distress syndrome (IRDS). The mod el consisted of linear viscous and inertive elements, a non-linear endotrac heal tube resistance, and a non-linear compliance allowing for alveolar rec ruitment and overdistention. Respiratory compliance was maximal at the tran sition between maximal recruitment and minimal overdistention. A new variab le, the oscillatory pressure ratio (OPR), was defined as the ratio between peak-to-peak oscillatory pressures at the distal end and the proximal openi ng of the endotracheal tube, respectively. The respiratory variables of fou r patients were fed into the model, and the relationship between respirator y system compliance and OPR was determined. OPR decreased as compliance increased, except for very low compliances belo w where 0.08 mL(.)cmH(2)O(-1), and OPR increased with increasing compliance . The relationship between mean airway pressure (P-aw) and OPR revealed tha t the minimal OPR (range, 0.37-0.78) and maximal respiratory compliance coi ncided at the same P,,. However, the relationship did depend on oscillation frequency, applied oscillatory pressure, and endotracheal tube resistance, parameters that may change during clinical application of HFOV. When 81 pe rmutations of nominal and extreme respiratory variables were used in the mo del, the minimum OPR (0.60 +/- 0.23) and maximum compliance coincided in al l cases. These model experiments support our hypothesis. The results indicate that t he OPR may be a useful index to optimize lung expansion, where lung recruit ment is maximal and overdistention minimal. In vivo tests will be needed to reveal the feasibility and reliability of such an index for biomedical and clinical application. (C) 2001 Wiley-Liss, Inc.