Partially and totally unloading respiratory muscles based on real-time measurements of work of breathing. A clinical approach

Abstract

Objective: To evaluate the clinical feasibility of using real-time measurements of work of breathing obtained at the bedside with a portable, commercially available respiratory monitor as an objective and quantifiable guideline for appropriately setting pressure support ventilation (PSV) to partially and totally unload the respiratory muscles in patients with respiratory failure.

Design: In vivo measurements of work of breathing were performed on a consecutive series of patients after applying incremental levels of PSV.

Setting: University teaching hospital in a surgical ICU.

Patients: Thirty adults (18 men and 12 women, ages 20 to 77 years) who had acute respiratory failure were studied. All patients had an endotracheal or a tracheostomy tube in place and were breathing spontaneously, receiving continuous positive airway pressure and PSV.

Interventions: Intraesophageal pressure (indirect measurement of intrapleural pressure) was measured with an esophageal balloon catheter positioned in the mid- to lower-third of the esophagus. Tidal volume was obtained by positioning a flow sensor between the "Y" piece of the breathing circuit and the endotracheal or tracheostomy tube. Airway pressure was measured from a catheter attached to the flow sensor. Data from these measurements were directed to the respiratory monitor (CP-100, Bicore Monitoring Systems) which calculates work of breathing performed by the patient using the Campbell diagram. Work of breathing performed by the ventilator to inflate the respiratory system was calculated by the monitor by integrating the change in airway pressure and tidal volume. Initially, the level of PSV was set to 0 cm H2O and work measurements were obtained. Pressure support ventilation was then increased until the work performed by the patient decreased to a range of 0.3 to 0.6 J/L, which corresponds to a normal range for physiologic work of breathing (ie, partial respiratory muscle unloading), and then until the work decreased to 0 J/L (ie, total respiratory muscle unloading).

Results: Work performed by the patient varied inversely (r = -0.83; p < 0.001) and work performed by the ventilator varied directly with the level of PSV (r = 0.94; p < 0.001). Work performed by the patient was 1.5 +/- 0.3 J/L at zero pressure support ventilation and decreased significantly to 0.50 +/- 0.1 J/L (p < 0.05) as the level of PSV was increased to 18 +/- 7 cm H2O. The respiratory muscles were partially unloaded under these conditions. Patient work decreased to 0 J/L and ventilator work increased when the muscles were totally unloaded at a PSV level of 31 +/- 8 cm H2O.

Conclusion: We propose an objective and goal-oriented clinical approach for using PSV by directly measuring the work of breathing performed by the patient with an easy to operate, bedside respiratory monitor and then applying pressure support ventilation to decrease the work to appropriate levels. Partially or totally shifting the workload from the respiratory muscles to the ventilator is appropriate under specific clinical conditions.