Which one is a better predictor of prognosis in COVID-19: analytical biomarkers or PaO2/FiO2?

Background: The study aimed to describe patient characteristics and outcomes by PaO2/FiO2 (PAFI) and degree of inflammation.

Methods: Retrospective cohort study with data on patients collected from March 1st, 2020 to March 1st, 2023, from the Spanish SEMI-COVID-19 Registry. Non-nosocomial patients with data on PAFI (<100 vs. 100-200 vs. 200-300 vs. >300) who received corticosteroids (CS) for COVID-19 in the first 48 h of admission were included in the study. 5314 patients met the inclusion criteria for the present study. The primary outcome was in-hospital mortality.

Results: Higher in-hospital mortality was found in the groups with PAFI < 100 (51.5% vs. 41.2% vs. 25.8% vs. 12.3%, P < .001). They also required more NIMV, IMV, and ICU admission, and had longer hospital stays. Those patients with PAFI > 300 and 4-5 high-risk criteria presented higher mortality than the patients with PAFI 200-300 and only 1-2 criteria of analytical inflammation. Risk factors associated with higher in-hospital mortality were age [OR = 1.06 (1.05-1.06)], moderate [OR = 1.87 (1.49-2.33)] and severe [OR = 2.64 (1.96-3.55)] degree of dependency, dyslipidemia [OR = 1.20 (1.03-1.39)], higher Charlson index [OR = 1.19 (1.14-1.24)], tachypnea on admission [2.23 (1.91-2.61)], the higher number of high-risk criteria on admission, and lower PAFI on admission. Female gender [OR = 0.77 (0.65-0.90)] and the use of RDSV [OR = 0.72 (0.56-0.93)] were found to be protective factors.

Conclusions: The lower the PAFI and the higher the degree of inflammation in COVID-19, the higher the in-hospital mortality. Inflammatory escalation precedes respiratory deterioration and should serve as an early predictor of severity to deciding the use of anti-inflammatory/immunosuppressive therapy.