Adjusting ventilator settings to relieve dyspnoea modifies brain activity in critically ill patients: an electroencephalogram pilot study

Abstract

Dyspnoea is frequent and distressing in patients receiving mechanical ventilation, but it is often not properly evaluated by caregivers. Electroencephalographic signatures of dyspnoea have been identified experimentally in healthy subjects. We hypothesized that adjusting ventilator settings to relieve dyspnoea in MV patients would induce EEG changes. This was a first-of-its-kind observational study in a convenience population of 12 dyspnoeic, mechanically ventilated patients for whom a decision to adjust the ventilator settings was taken by the physician in charge (adjustments of pressure support, slope, or trigger). Pre- and post-ventilator adjustment electroencephalogram recordings were processed using covariance matrix statistical classifiers and pre-inspiratory potentials. The pre-ventilator adjustment median dyspnoea visual analogue scale was 3.0 (interquartile range: 2.5–4.0; minimum-maximum: 1–5) and decreased by (median) 3.0 post-ventilator adjustment. Statistical classifiers adequately detected electroencephalographic changes in 8 cases (area under the curve ≥0.7). Previously present pre-inspiratory potentials disappeared in 7 cases post-ventilator adjustment. Dyspnoea improvement was consistent with electroencephalographic changes in 9 cases. Adjusting ventilator settings to relieve dyspnoea produced detectable changes in brain activity. This paves the way for studies aimed at determining whether monitoring respiratory-related electroencephalographic activity can improve outcomes in critically ill patients under mechanical ventilation.

The breathing brain: The potential of neural oscillations for the understanding of respiratory perception in health and disease

“…Most interestingly, the cortical readiness potential was also observed prior to inspiration in COPD patients during quiet breathing (Nguyen et al., 2018), indicating that experimental manipulations in healthy participants can likely evoke naturalistic breathlessness characteristic of respiratory disease, at least in terms of neural processing. Furthermore, the presence of this spontaneous activity was related to the intensity of reported dyspnea (Nguyen et al., 2018), and its disappearance was noted in parallel with reported dyspnea relief following ventilator settings change in dyspneic patients (Raux et al., 2019), providing first evidence of spontaneous EEG activity relating to the subjective assessment of breathlessness.…”

Combined head accelerometry and EEG improves the detection of respiratory‐related cortical activity during inspiratory loading in healthy participants

“…Data analyses were performed using Matlab (Mathworks Inc) version 2017b. The methodology of the EEG covariance‐based classifier has been described previously (Hudson et al, 2016; Raux et al, 2019; Navarro‐Sune et al, 2017). It uses an outlier detection approach to test for “altered” brain activity compared to “reference” activity and here, was used to test between loaded and quiet breathing in both postures.…”

“…In this paper, we introduced two improvements with respect to previous publications (Hudson et al, 2016; Raux et al, 2019; Navarro‐Sune et al, 2017) a new classifier based on accelerometer data. It can be either used alone or associated with the EEG classifier (i.e., ‘Fusion’ of distances) to improve the overall performances. a time buffer to average the distance of a given segment with previous distances, introducing a “smoothing” effect.…”

“…Electroencephalography activity (EEG) can be used to detect respiratory‐related cortical activity, as a pre‐motor potential or ‘ Bereitschaftspotential ’ (Macefield & Gandevia, 1991). Pre‐motor potentials have been detected before inspiration (pre‐inspiratory potentials) or before expiration (pre‐expiratory potentials) in loaded breathing or simulated ‘patient‐ventilator asynchrony’ in healthy control participants (Raux et al, 2007a, b; Hudson et al, 2016; Morawiec et al, 2015), during resting breathing in respiratory or neurological disease (Nguyen et al, 2018; Georges et al, 2016; Launois et al, 2015) and in critically ill patients who are mechanically ventilated (Raux et al, 2019). Respiratory‐related cortical activity is usually accompanied by respiratory discomfort or dyspnoea (Morawiec et al, 2015; Nguyen et al, 2018; Georges et al, 2016; Raux et al, 2019).…”

“…Pre‐motor potentials have been detected before inspiration (pre‐inspiratory potentials) or before expiration (pre‐expiratory potentials) in loaded breathing or simulated ‘patient‐ventilator asynchrony’ in healthy control participants (Raux et al, 2007a, b; Hudson et al, 2016; Morawiec et al, 2015), during resting breathing in respiratory or neurological disease (Nguyen et al, 2018; Georges et al, 2016; Launois et al, 2015) and in critically ill patients who are mechanically ventilated (Raux et al, 2019). Respiratory‐related cortical activity is usually accompanied by respiratory discomfort or dyspnoea (Morawiec et al, 2015; Nguyen et al, 2018; Georges et al, 2016; Raux et al, 2019). Thus, it was proposed that EEG could be used in a brain–ventilator interface (BVI) to detect respiratory discomfort and consequently improve treatment in mechanically ventilated patients (Navarro‐Sune et al, 2017).…”