Effect of mechanical power on mortality in invasively ventilated ICU patients without the acute respiratory distress syndrome: An analysis of three randomised clinical trials

Abstract

BACKGROUNDThe mechanical power of ventilation (MP) has an association with outcome in invasively ventilated patients with the acute respiratory distress syndrome (ARDS). Whether a similar association exists in invasively ventilated patients without ARDS is less certain.OBJECTIVETo investigate the association of mechanical power with mortality in ICU patients without ARDS.DESIGNThis was an individual patient data analysis that uses the data of three multicentre randomised trials.SETTINGThis study was performed in academic and nonacademic ICUs in the Netherlands.PATIENTSOne thousand nine hundred and sixty-Two invasively ventilated patients without ARDS were included in this analysis. The median [IQR] age was 67 [57 to 75] years, 706 (36%) were women.MAIN OUTCOME MEASURESThe primary outcome was the all-cause mortality at day 28. Secondary outcomes were the all-cause mortality at day 90, and length of stay in ICU and hospital.RESULTSAt day 28, 644 patients (33%) had died. Hazard ratios for mortality at day 28 were higher with an increasing MP, even when stratified for its individual components (driving pressure (P<0.001), tidal volume (P<0.001), respiratory rate (P<0.001) and maximum airway pressure (P=0.001). Similar associations of mechanical power (MP) were found with mortality at day 90, lengths of stay in ICU and hospital. Hazard ratios for mortality at day 28 were not significantly different if patients were stratified for MP, with increasing levels of each individual component.CONCLUSIONIn ICU patients receiving invasive ventilation for reasons other than ARDS, MP had an independent association with mortality. This finding suggests that MP holds an added predictive value over its individual components, making MP an attractive measure to monitor and possibly target in these patients.TRIAL REGISTRATIONClinicalTrials.gov Identifier: NCT02159196, ClinicalTrials.gov Identifier: NCT02153294, ClinicalTrials.gov Identifier: NCT03167580.