Quantitative Chest CT analysis in COVID-19 to predict the need for oxygenation support and intubation

Abstract

OBJECTIVE: Lombardy (Italy) was the epicentre of the COVID-19 pandemic in March 2020. The healthcare system suffered from a shortage of ICU beds and oxygenation support devices. In our Institution, most patients received chest CT at admission, only interpreted visually. Given the proven value of Quantitative CT analysis (QCT) in the setting of ARDS, we tested QCT as an outcome predictor for COVID-19.METHODS: We performed a single centre retrospective study on COVID-19 patients hospitalized from January 25th, 2020 to April 28th 2020, who received CT at admission prompted by respiratory symptoms such as dyspnea or desaturation. QCT was performed using a semi-automated method (3D-Slicer). Lungs were divided by Hounsfield Unit intervals. Compromised lung (%CL) volume was the sum of poorly and non-aerated volumes (-500,100HU). We collected patient’s clinical data including oxygenation support throughout hospitalization.RESULTS: Two hundred twenty-two patients (163 males, median age 66, IQR 54-6) were included; 75% received oxygenation support (20% intubation rate). Compromised lung volume was the most accurate outcome predictor (logistic regression, p<0.001). %CL values in the 6-23% range increased risk of oxygenation support; values above 23% were at risk for intubation. %CL showed a negative correlation with PaO2/FiO2 ratio (p<.001) and was a risk factor for in-hospital mortality (p<.001)CONCLUSIONS: QCT provides new metrics of COVID-19. The compromised lung volume is accurate in predicting the need for oxygenation support and intubation and is a significant risk factor for in-hospital death. QCT may serve as a tool for the triaging process of COVID-19.

Quantification of COVID-19 Opacities on Chest CT – Evaluation of a Fully Automatic AI-approach to Noninvasively Differentiate Critical Versus Noncritical Patients

“…Our findings may indicate that the level of lung manifestation could also be associated with a higher probability of concomitant diseases. Furthermore, this study supports previous findings in terms of association between quantification features and clinical parameters [ 8 , 27 ]. ICU and non-ICU patients significantly differed in Opacityscore.…”

“…Clinical workload and organizational effort have continuously increased during the COVID-19 pandemic [17] . Non-invasive and rapid screening for potential ICU candidates by applying fully automatically calculated CT-scoring systems may result in a more time-efficient clinical workflow, which may be necessary in the scenario where COVID-19 admission rates rapidly increase [8] . In this context, our experience with the prototype's average evaluation time of approximately 2 minutes for a chest 1.0 HR-CT scan emphasizes time-efficient application in clinical routine.…”

“…Certain studies have shown that quantitative analysis of COVID-19 like lung findings on chest CT is correlated with clinical and laboratory data and partially might be superior to previous clinical biomarkers in predicting COVID-19 progression to severe illness [ 8 , 18 , 19 ]. Nevertheless, manual or semi-automatic quantification of lung lesions on chest CT scans is still prone to inter-observer variance and remains time-consuming [9] .…”

“…Xiong et al showed that laboratory data like CRP, erythrocyte sedimentation rate and LDH correlated with the severity of abnormalities manually quantified on chest CT scans [7] . Lanza et al found that semi-automatic quantification of compromised lung volume in COVID-19 patients on chest CT scans correlated with the need for oxygenation support and intubation and might therefore serve as a triaging and prediction tool in COVID-19 positive patients [8] . However, manual and semi-automatic quantification and feature extraction of pulmonary opacifications is time-consuming and prone to inter-observer variance complicating implementation in clinical routine [9] .…”

Pathophysiology of COVID-19-associated acute respiratory distress syndrome: a multicentre prospective observational study

“… 15 Total lung weight was estimated from standard non-contrast chest CT scans (done at clinical levels of PEEP) with a dedicated medical imaging software equipped with a semiautomated segmentation algorithm (3D Slicer). 16 Presence of pulmonary intravascular clots was assessed by analysing CT-pulmonary angiograms using software installed on the IntelliSpace Portal release 11. The application uses an advanced automatic computer-aided design algorithm for detecting filling defects.…”