No participants were enrolled, no funding obtained to perform study

Healthy biological systems are characterized by a normal range of "variability" in organ function. For example, many studies of heart rate clearly document that loss of the normal level of intrinsic, beat-to-beat variability in heart rate is associated with poor prognosis and early death.

Unlike the heart, little is known about patterns of respiratory variability in illness. What is known is that, like the heart, healthy subjects have a specific range of variability in breath- to-breath depth and timing. Additionally, in animal models, ventilator strategies that re-introduce normal variability to the breathing pattern significantly reduce ventilator-associated lung injury.

Critically ill patients requiring mechanical ventilation offer an opportunity to observe and analyze respiratory patterns in a completely non-invasive manner. Current mechanical ventilators produce real-time output of respiratory tracings that can analyzed for variability.

The investigators propose to non-invasively record these tracings from patients ventilated in the intensive care units for mathematical variability analysis. The purpose of these pilot analyses are to: (1) demonstrate the range of respiratory variability present in the mechanically ve ventilated critically ill and (2) demonstrate the ventilator modality that delivers or permits the closest approximation to previously described beneficial or normal levels of variability. Future studies will use this pilot data in order to determine if the observed patterns of respiratory variability in mechanically ventilated critically ill subjects have prognostic or therapeutic implications.

Different modes of mechanical ventilation allow different levels of patient control of the respiratory pattern. For example, the most common mode of ventilation, called volume control, gives very little control to the patient in the amount of air taken for each breath. Other modes, such as pressure control, pressure regulated volume control and pressure support, allow more patient control of the volume of air delivered by the ventilator. Newer modes, such as airway pressure release ventilation (APRV), allow completely spontaneous patient respirations. All of these modes allow at least some patient control of respiratory rates.

Studies of natural breathing by healthy subjects have shown normal levels of variability in respiratory rate and tidal volume.1 Variability in physiological processes has been associated with health and the loss of variability can presage the onset of illness. For example, normal humans exposed to LPS (lipopolysaccharide-the potent immune-stimulating cell wall component of bacteria) lose their normal respiratory variability. Thus, physiological variability may represent a "hidden vital sign," the monitoring of which may herald important clinical events. Additionally, re-establishing normal levels of variability has therapeutic benefits in animal models. 2

The variability in respiratory patterns in ill patients has not been well studied. For example, it is currently unclear if critical illness results in deviations from normal variability patterns, if ventilator modes allowing increasing patient control of respiration allow patients to attain greater normalcy of respiratory variability, or if deviations from normal respiratory patterns while on ventilator modes which allow for increased levels of spontaneous breathing have prognostic implications.

The purpose of this pilot study is to record respiratory patterns from the ventilators of patients receiving various modes of mechanical ventilation in order to quantify and compare levels of respiratory variability associated with each mode. Our hypothesis is that APRV, a mode that allows spontaneous respiration, will be associated with respiratory variability patterns that most closely approximate that of normal subjects.

We hope that data derived from this study will inform future observational studies correlating respiratory variability during mechanical ventilation with severity of illness and prognosis.

Inclusion Criteria:

• Age > or equal to 18.
• Requires mechanical ventilation.
• Admitted to surgical, medical, or coronary critical care unit
• Triggering ventilator above set rate
• Meets American European consensus definition of acute lung injury: PaO2/FiO2 ratio <300 or SaO2/FiO2 ratio<315; bilateral infiltrates on chest x-ray (CXR); no congestive heart failure (CHF).

Exclusion Criteria:

• None

Patients admitted to adult medical, surgical, and coronary critical care units at Boston Medical Center.