Reliability of airway occlusion pressure (P0.1) in predicting extubation failure in critically ill patients

The airway occlusion pressure (P0.1) reflects the central respiratory drive and correlates with patients' inspiratory effort. This study aimed to assess whether changes in P0.1 (ΔP0.1) during the spontaneous breathing trial (SBT), as well as SBT-induced changes in the products of P0.1 and the rapid shallow breathing index (RSBI) (Δ(P0.1 × RSBI)) and P0.1 and respiratory rate ((Δ(P0.1 × RR)), can predict extubation failure (EF) in critically ill patients.

In this bicentric prospective study, we enrolled 114 intensive care unit (ICU) patients who had been mechanically ventilated for more than 48 h and successfully tolerated a 30-min SBT on pressure support ventilation. Respiratory variables, including P0.1 and RSBI, as well as hemodynamic parameters, were recorded within 5 min of SBT initiation and at its completion. EF was defined as the need for reintubation and resumption of mechanical ventilation within 7 days. Areas under the receiver operating characteristic curves (AUCs) were calculated for both relative and absolute ΔP0.1, Δ(P0.1×RSBI), and Δ(P0.1×RR).

Among the 114 patients, 23 (20.2%) experienced EF. At baseline, P0.1, P0.1×RR, and P0.1×RSBI values did not differ significantly between the success and failure groups. During the SBT, P0.1, P0.1×RR, and P0.1 × RSBI increased significantly in the failure group, while remaining unchanged in the success group. Absolute ΔP0.1 demonstrated poor predictive performance for EF, with AUC of 0.69 (P=0.005). In contrast, absolute Δ(P0.1×RSBI), and absolute Δ(P0.1 × RR) showed moderate predictive ability, with AUCs of 0.79 (P<0.001) and 0.76 (P<0.001), respectively. In the multivariable logistic regression analyses, the absolute, but not the relative, ΔP0.1, Δ(P0.1×RR), and Δ(P0.1×RSBI) remained independently associated with EF after adjustment for potential confounders.

Variations in P0.1 and in the combined parameters P0.1×RSBI and P0.1×RR during the SBT provided limited value for predicting EF in critically ill patients.

ClinicalTrials.gov, NCT05802745.