Find in Library

Abstract Background Recent studies have shown an association between baseline arterial carbon dioxide pressure (PaCO2) and outcomes in patients with acute respiratory distress syndrome (ARDS). However, PaCO2 probably varies throughout the disease, and few studies have assessed the effect of longitudinal PaCO2 on prognosis. We thus aimed to investigate the association between time-varying PaCO2 and 28-day mortality in mechanically ventilated ARDS patients. Methods In this retrospective study, we included all adult (≥ 18 years) patients diagnosed with ARDS who received mechanical ventilation for at least 24 h at a tertiary teaching hospital between January 2014 and March 2021. Patients were excluded if they received extracorporeal membrane oxygenation (ECMO). Demographic data, respiratory variables, and daily PaCO2 were extracted. The primary outcome was 28-day mortality. Time-varying Cox models were used to estimate the association between longitudinal PaCO2 measurements and 28-day mortality. Results A total of 709 patients were eligible for inclusion in the final cohort, with an average age of 65 years, of whom 70.7% were male, and the overall 28-day mortality was 35.5%. After adjustment for baseline confounders, including age and severity of disease, a significant increase in the hazard of death was found to be associated with both time-varying PaCO2 (HR 1.07, 95% CI 1.03–1.11, p<0.001) and the time-varying coefficient of variation for PaCO2 (HR 1.24 per 10% increase, 95% CI 1.10–1.40, p<0.001) during the first five days of invasive mechanical ventilation. The cumulative proportion of exposure to normal PaCO2 (HR 0.72 per 10% increase, 95% CI 0.58–0.89, p = 0.002) was associated with 28-day mortality. Conclusion PaCO2 should be closely monitored in mechanically ventilated ARDS patients. The association between PaCO2 and 28-day mortality persisted over time. Increased cumulative exposure to normal PaCO2 was associated with a decreased risk of death.