Prevalência, fatores de risco e desfechos associados a uma ventilação mecânica não protetora em pacientes sob risco para Síndrome do Desconforto Respiratório Agudo: um estudo de coorte

Abstract

Background: Despite evidence that protective mechanical ventilation (MV) may be beneficial to patients without ARDS, observational studies have shown that adhesion to them are still variable. Objective: This study aimed to evaluate the proportion of patients without ARDS who received non-protective MV, and the risk factors for adopting this strategy. We also assessed the association of non-protective MV with mortality. Methods: This prospective cohort study was carried out in one ICU of a University Hospital in Brazil, and included 116 patients in MV, with at least one risk for developing ARDS. Ventilatory parameters were collected twice daily for seven days, and patients were separated into two groups (protective or non-protective MV), based on the driving pressure (DP) or the tidal volume (VT) Results: Non-protective MV occurred in 32.8% of the patients based on the DP (95% CI 24.7-41.9%), and in 42.2%, based on the VT (95% CI 33.5-51.5%). Pneumonia, compliance of the respiratory system (Crs), ratio of partial pressure arterial oxygen and fraction of inspired oxygen (PaO2/FIO2) and pH were independent factors associated with non-protective MV, based on the DP; while Simplified Acute Physiologic Score-3 (SAPS 3), predicted body weight (PBW), and PaO2/FIO2, were those associated with non-protective MV based on the VT. Non-protective MV based on the DP, but not on the VT, was independently associated with hospital mortality. Conclusions: Potentially injurious ventilatory settings are still frequent in patients at risk for ARDS, and notably higher DPs associate with hospital mortality. We should be aware in adjusting these parameters, mainly in situations in which there is a higher risk for setting non-protective ones, such as the presence of pneumonia; lower levels of PaO2/FIO2, Crs, and pH; higher SAPS 3 scores, and lower PBW.