A26-24 Noninvasive Quantification of Energy Transfer During Mechanical Ventilation

Rationale

Insufflation energy transfer (ET) is viewed increasingly as a key driver of ventilator-induced lung injury. Current formulations, however, omit patient-generated respiratory effort (Pmus), a potentially important contributor to ET. The objective was to develop and validate clinically mathematical equations to quantify individual insufflation ET in the presence of Pmus using digitally acquired airway pressure (Paw) and flow (Faw) signals.

Methods

A one-compartment model was used to derive equations relating the pressure-time product of Pmus (PmusPTP) to ET during volume-controlled (VCV) and pressure-controlled (PCV) ventilation. The equations were validated using high-fidelity Paw and Faw signals previously acquired at 31.25 Hz and stored as 131-s epochs in a database. Numerically calculated respiratory system elastance (Ers) and resistance (Rrs) were validated against data from passively ventilated (Pmus ≈ 0) VCV epochs displaying ventilator-applied end inspiratory-holds. Predicted ET during VCV and PCV were compared to values determined by trapezoidal integration of pressure-volume diagrams.

Results

Forty selected epochs contained 2,812 breaths for analysis. Predicted Ers and Rrs matched measured values with minimal bias and narrow limits of agreement (Ers: R² = 0.95; bias -0.4 ± 1.4 cmH₂O·L⁻¹; Rrs: R² = 0.98; bias 0.7 ± 1.1 cmH₂O·s·L⁻¹). There was a strong correlation between calculated and measured ET per insufflation for all recorded VCV breaths (R2 =0.97; n = 1258). Bland-Altman analysis showed bias 0.02 with ± 0.11 J limits of agreement (LOA).Similar results were found for PCV: n = 1554; R² = 0.92; bias -0.01 ± 0.20 J.

Conclusions

This study presents and validates a noninvasive method that uses only airway pressure and flow signals to dynamically quantify energy transfer to mechanically ventilated patients while accounting for respiratory effort. Equations derived from a classic one-compartment model accurately predicted individual breath pressure-time product, tidal volume, and energy transfer during volume- and pressure-controlled ventilation. The method provides a physiologic framework to explore, in real-time and non-invasively, the dynamics and clinical significance of ventilator-to-patient energy transfer and provide a physiological basis for ventilator software capable of real-time monitoring of insufflation energy transfer.

This abstract is funded by: None