Dear Editor,

Reverse triggering (RT) is an under-recognized form of patient–ventilator asynchrony detectable at the bedside [1, 2]. RT can promote ventilator-induced lung injury (VILI) by producing double triggering and stacked breaths resulting in an increase of tidal volume (Vt) and transpulmonary pressure (PL) [2]. Pendelluft contributes by inducing a redistribution of gas during RT leading to a local increase of PL. The aim of this study was the screening of RT phenomenon in deeply sedated, mechanically ventilated ARDS patients either receiving or not neuromuscular blocking agents (NMBA).

This is an ancillary physiologic study derived from a prospective randomized study [3]. We retrospectively analysed the continuous recordings of airway pressure, flow, volume and oesophageal pressure (Pes) over time in a subgroup of 21 ARDS patients ventilated with a PaO2/FiO2 ratio below 150 mmHg. Other inclusion criteria are detailed in the electronic supplementary material (ESM).

Oesophageal pressure measurements were performed via a specific nasogastric feeding tube (SmartCathG®, VIASYS Healthcare, Palm Springs, CA, USA) equipped with an oesophageal balloon. Analysis was manually performed from the recordings using Acknowledge™ (Biopac) software by two intensivists (MM and MG). When RT was identified, the characteristics were analysed over the following 1-h period in order to describe its rate and duration.

For every patient and every recording in which we found the RT phenomenon, characteristics (frequency, entrainment ratio (ER), phase difference) were described or computed according to the methodology described by Akoumianaki et al. [2].

Continuous recordings of airway pressure, flow, volume and Pes were obtained in 21 moderate-to-severe ARDS patients (NMBA group, n = 11; control group not receiving NMBA, n = 10). Recording time totalled 10,920 min (182 h) in the NMBA group and 16,440 min (274 h) in the control group. Sedation consumption is reported in the ESM.

No RT phenomenon was observed in the NMBA group. Three out of 10 patients (33%) in the control group had evidence of RT (p = 0.09). In this group, RT was observed during 0.33–17.88% of the 60 min of the analysis. Characteristics of RT for each patient are detailed in Table 1. Consequences on Vt and PL depend on the time of respiratory cycle in which the RT occurs. The dominant pattern of entrainment was not stable (1/2, 1/3 or minus).

Table 1 Analysis of reverse triggering asynchrony for the three patients in the control group
Full size table

Akoumianaki et al. [2] reported that RT observed in eight deeply sedated ARDS patients was characterized by a stable ER described as 1:1, 1:2 and 1:3. In our study, the ER was not stable. The major difference between the two studies is the duration of analysis i.e. 60 min vs 6–27 min.

Inspiratory efforts generated by the RT caused ample variations of the level of lung pressures (delta Paw, plateau pressure, Pes and PL). We observed evidence of double triggering with stacked breaths leading to potential consequences: increase of the PL and, consequently, the increase of stress and strain on the lungs leading to VILI [4, 5].

We confirmed the existence of RT in deeply sedated ARDS patients not receiving NMBA. Our study supports the experimental data of deleterious effects of strong spontaneous efforts in the early phase of severe ARDS.