Outcome patterns in COVID-ARDS compared to ARDS caused by other respiratory infections: an individual patient data analysis of six observational studies

TL;DR

This study compares mortality rates in patients with ARDS caused by COVID-19 versus other respiratory infections, finding higher 60-day mortality in the former after adjusting for risk factors.

Contribution

The study provides individual patient data analysis comparing mortality patterns in COVID-ARDS and non-COVID-ARDS after adjusting for risk factors and geographic region.

Findings

Unmatched analysis showed higher 28-day and 60-day mortality in COVID-ARDS patients.

After adjusting for risk factors, only 60-day mortality remained higher in the COVID-ARDS group.

Mortality rates became similar after adjusting for geographic region.

Abstract

Mortality rates among patients with acute respiratory distress syndrome (ARDS) due to COVID-19 vary widely, and there is a lack of studies comparing mortality patterns in COVID-ARDS to ARDS caused by other respiratory infections (non-COVID-ARDS).

A post hoc analysis was performed on a database comprising individual patient data from four observational studies on COVID-19 and two pre-pandemic studies; the analysis was limited to patients with ARDS receiving mechanical ventilation. Propensity score matching was used to control for factors known to be associated with mortality and for geographic region.

The analysis included 8,117 patients: 6,702 with COVID-ARDS and 1,415 with non-COVID-ARDS. Lung-protective ventilation, including low tidal volume and high positive end-expiratory pressure, was more often applied in the COVID-ARDS group. In unmatched analysis, mortality at day 28 and day 60 was 37.0 versus 33.5% (p = 0.012) and 44.2 versus 36.4% (p = 0.002) in COVID-ARDS versus non-COVID-ARDS, respectively. In a matched analysis, including 4,580 patients with COVID-ARDS and 1,274 patients with non-COVID-ARDS, mortality at days 28 and 60 was 35.7% versus 33.5% (p = 0.149) and 42.5% versus 36.5% (p = 0.002), respectively. In a matched analysis controlling for geographic region, mortality at days 28 and 60 did not differ between COVID-ARDS and non-COVID-ARDS.

In an unmatched analysis, patients with COVID-ARDS had higher mortality at day 28 and day 60 than patients with non-COVID-ARDS; after adjusting for risk factors, only 60-day mortality remained higher, and after adjusting for geographic region, mortality rates were similar.

Figures