The PaO2/FiO2 ratio on admission is independently associated with prolonged hospitalization in COVID-19 patients.

Angelo Zinellu,Giordano Madeddu,Valentina Scano,Vito Fiore,Panagiotis Paliogiannis,Elisabetta Zinellu,Arduino Aleksander Mangoni,Luigi Benedetto Arru,Ciriaco Carru,Alessandro Giuseppe Fois,Ivana Maida,Sergio Babudieri,Pietro Pirina,Andrea De Vito

doi:10.3855/jidc.13288

Abstract

The early identification of factors that predict the length of hospital stay (HS) in patients affected by coronavirus desease (COVID-19) might assist therapeutic decisions and patient flow management. We collected, at the time of admission, routine clinical, laboratory, and imaging parameters of hypoxia, lung damage, inflammation, and organ dysfunction in a consecutive series of 50 COVID-19 patients admitted to the Respiratory Disease and Infectious Disease Units of the University Hospital of Sassari (North-Sardinia, Italy) and alive on discharge. Prolonged HS (PHS, >21 days) patients had significantly lower PaO2/FiO2 ratio and lymphocytes, and significantly higher Chest CT severity score, C-reactive protein (CRP) and lactic dehydrogenase (LDH) when compared to non-PHS patients. In univariate logistic regression, Chest CT severity score (OR = 1.1891, p = 0.007), intensity of care (OR = 2.1350, p = 0.022), PaO2/FiO2 ratio (OR = 0.9802, p = 0.007), CRP (OR = 1.0952, p = 0.042) and platelet to lymphocyte ratio (OR = 1.0039, p = 0.036) were significantly associated with PHS. However, in multivariate logistic regression, only the PaO2/FiO2 ratio remained significantly correlated with PHS (OR = 0.9164; 95% CI 0.8479-0.9904, p = 0.0275). In ROC curve analysis, using a threshold of 248, the PaO2/FiO2 ratio predicted PHS with sensitivity and specificity of 60% and 91%, respectively (AUC = 0.780, 95% CI 0.637-0.886 p = 0.002). The PaO2/FiO2 ratio on admission is independently associated with PHS in COVID-19 patients. Larger prospective studies are needed to confirm this finding.

Highlights

The early identification of factors that predict the length of hospital stay (HS) in patients affected by coronavirus desease (COVID19) might assist therapeutic decisions and patient flow management
In order to address this pressing issue, we retrospectively studied a consecutive series of 50 COVID-19 patients admitted to a dedicated referral centre in north Sardinia (Italy), with clinical and demographic characteristics comparable to those recently described in other COVID-19 cohorts [5-7,1416]
The outbreak of the new pandemic caused by the betacoronavirus SARS-CoV-2 got in trouble several countries all over the world, even those whose health system was believed to be cutting-edge

Summary

Introduction

The early identification of factors that predict the length of hospital stay (HS) in patients affected by coronavirus desease (COVID19) might assist therapeutic decisions and patient flow management. Since the major route of transmission of SARS-CoV2 is via droplet spread, which requires close contact [10], many countries have adopted extraordinary physical distancing strategies to reduce diffusion and mitigate the impact of the pandemic on health care systems, in terms of hospital staff and bed availability. In this context, the implementation of effective patient flow management strategies would benefit from a better understanding of the clinical progress of the disease and the factors that are associated with the length of hospital stay (HS).

Methods

Results

Discussion

Conclusion