The airway occlusion pressure (P_0.1) to monitor respiratory drive during mechanical ventilation: increasing awareness of a not-so-new problem

Importance of monitoring respiratory drive during mechanical ventilation

An inadequate respiratory drive under mechanical ventilation, either too high or too low, has recently been incriminated as a risk factor for both lung [1] and diaphragmatic injury [2]. Monitoring and controlling the drive to breathe might, therefore, be important for clinical practice. However, respiratory drive assessment has mostly been limited to research purposes, with few techniques available at the bedside [3]. A simple non-invasive measure, the airway occlusion pressure (P_0.1), i.e. the pressure developed in the occluded airway 100 ms after the onset of inspiration (Fig. 1), was first described 40 years ago. Currently, nearly all modern ventilators provide a means of measuring P_0.1. Despite having a better understanding of the importance of the respiratory drive during mechanical ventilation, no recommendations exist about its use.

Fig. 1

Measurement of airway occlusion pressure (P_0.1). Airway pressure (P_aw in red) and flow (in blue) during an un-occluded breath and a breath during an end-expiratory occlusion of two different patients under assisted mechanical ventilation with two very different levels of drive. P_0.1 is measured from the P_aw tracing as the drop in airway pressure during the first 100 ms of the breath against an occluded airway. Patient corresponding to tracing (a) has a lower ventilatory support and higher respiratory drive than patient corresponding to tracing (b). PS pressure support level, PEEP positive end-expiratory pressure level

Original description and rationale

In healthy subjects, Whitelaw et al. [4] performed random, short end-expiratory occlusions through a special circuit during both resting and CO₂ rebreathing. They found that the decrease in airway pressure (P_aw) during the first 100 ms (i.e. 0.1 s) of an occluded breath was relatively constant, consistent for each patient in each condition, and correlated better with end-tidal CO₂ than minute ventilation. They named this new parameter airway occlusion pressure or P_0.1 (Fig. 1).

Several characteristics make P_0.1 a good measure of respiratory center output. There is no conscious or unconscious reaction to the mechanical load during the first milliseconds of an unexpected occlusion. Since it starts from end-expiratory lung volume, any drop in P_aw is independent of the recoil pressure of the lung or thorax. Because the flow is interrupted, P_0.1 is independent of resistance, and there is no change in lung volume that could induce inhibitory reflexes or modify the force–velocity relationship. Finally, there is a good correlation between P_0.1 and inspiratory effort measured either by the work of breathing (WOB) or the pressure–time product [5, 6]. Importantly, P_0.1 is still reliable during respiratory muscle weakness if spontaneous breathing is preserved [7].

Range of values

In healthy subjects, P_0.1 varies between 0.5 and 1.5 cmH₂O [3]. In stable, non-intubated patients with COPD, P_0.1 varies between 2.5 and 5.0 cmH₂O [3]. Ranges of P_0.1 from 3.0 to 6.0 cmH₂O have been reported in patients with ARDS under mechanical ventilation, and from 1.0 to as high as 13 cmH₂O during weaning.

Sources of errors and potential pitfalls

There is a significant breath-to-breath variability of P_0.1, and an average of 3–4 values of P_0.1 in one patient in one clinical condition should be obtained to represent a reliable index of respiratory drive [8]. In patients with intrinsic positive end-expiratory pressure (PEEPi), there is a delay between the onset of inspiratory effort and the drop in airway pressure during an end-expiratory occlusion. P_0.1 measured at the mouth in non-intubated patients can underestimate respiratory drive [9]. However, Conti et al. [10] proved that measurement of P_0.1 from the drop in P_aw, since flow reaches zero during the triggering phase of the ventilator, is a reliable surrogate of the decay in esophageal pressure during the first 100 ms of the effort. The difference between the two measurements is small and clinically acceptable (− 0.3 ± 0.5 cmH₂O).

Specific aspects are important when interpreting P_0.1 displayed by ventilators: decompression of air in the circuit can result in underestimation of P_0.1, and ventilators use different methods to measure P_0.1. Some ventilators perform a short end-expiratory occlusion when a manoeuver is activated, but others display a breath-to-breath estimation based on the trigger phase. The trigger phase in modern ventilators is often less than 50 ms, which could result in underestimation of P_0.1 especially if the inspiratory effort is high [11].

P _0.1 under mechanical ventilation

P_0.1 can be useful to adjust the level of ventilatory support due to its close correlation with inspiratory effort. Higher values of P_0.1 indicate insufficient levels of support while lower values correspond to excessive assistance [5], both during assist-controlled and spontaneous modes of ventilation. We recently showed that P_0.1 can detect excessive levels of inspiratory effort in patients under pressure control, intermittent mandatory and synchronized intermittent mandatory ventilation [12]. The optimal threshold of P_0.1 was 3.5 cmH₂O with a sensitivity of 92% and a specificity of 89%. Pletsch-Assuncao et al. [13] recently reported an optimal threshold of P_0.1 ≤ 1.6 cmH₂O to diagnose overassistance defined by WOB < 0.3 J/L or > 10% ineffective efforts (with a sensitivity of 62% and a specificity of 87%). This had a lower performance than the presence of a respiratory rate ≤ 17 bpm, possibly related to the technique to measure P_0.1 (manual occlusion) and the definition of overassistance. Interestingly, a closed-loop algorithm to automatically adjust the level of support based on a target P_0.1 has proved feasible [14].

P_0.1 can be used to adjust external positive end-expiratory pressure (PEEP) in patients with hyperinflation. Mancebo et al. [6] proved that a decrease in estimated P_0.1 with the addition of external PEEP indicates a drop in PEEPi and in WOB with reasonable sensitivity and specificity.

More recently, Mauri et al. [15] found that P_0.1 is a sensitive indicator of respiratory drive in patients with severe ARDS undergoing venous–venous extracorporeal membrane oxygenation. They showed that a change in sweep gas flow resulting in a change in PaCO₂ was well reflected by P_0.1, varying on average between 0.9 and 3.0 cmH₂O.

P_0.1 has extensively been studied as a predictor of weaning success or failure. Originally, a high P_0.1 during a spontaneous breathing trial was associated with failure, suggesting that a high respiratory drive could predict weaning failure. As elegantly proved by Bellani et al. [16], patients failing a trial of decrease in support during weaning are unable to increase oxygen consumption in response to an increased drive (i.e. higher P_0.1).

However, overlap in P_0.1 values between success and failure groups was evidenced as more data were published, and no threshold accurately predicts weaning outcome using P_0.1 alone or combined with other parameters [17]. This is explained by the complex pathophysiology of weaning failure and the design of experiments. P_0.1 was often measured during pressure support, known to underestimate inspiratory effort after extubation [18]. Despite having no magic value to predict weaning outcome, clinicians can still get information concerning the respiratory drive (high or low). In this context, very high values (for example, higher than 6 cmH₂O) are associated with failure.

Conclusions and future directions

P_0.1 is a useful and valid measure of respiratory drive in mechanically ventilated patients. Work is needed to build a bridge between research and clinical practice since this parameter is now easily available. The accuracy of P_0.1 displayed by modern ventilators (using flow or pressure trigger), and in different clinical conditions (e.g., presence of PEEPi) needs to be studied.

Additionally, a practical approach to the use of P_0.1 displayed by ventilators needs to be evaluated. P_0.1 can be used to detect excessive or insufficient levels of inspiratory effort and better guide muscle- and lung-protective ventilation strategies. The latter should include adjusting ventilator settings, CO₂ removal and titration of sedative drugs to achieve an acceptable range of inspiratory effort for a given clinical condition. In particular, P_0.1 could be of great value during a sensitive period: transition from fully controlled to assisted modes of ventilation.

P_0.1 can already provide clinicians with information regarding the drive of their patients, it is sensitive to ventilator settings, and may be useful during weaning. Considering the importance of patients’ respiratory drive under mechanical ventilation, it seems that it is time to start using it in the clinical setting.