[Quantitative analysis of chest CT in coronavirus infected patients and its correlation with clinical features].

Abstract

Objective: To explore the quantitative analysis results of different patterns of chest computed tomography (CT) in patients with coronavirus infection and its relationship with viral load and pathophysiological status.

Methods: A retrospective clinical cohort study was conducted. Patients with coronavirus infection admitted to Qingdao Municipal Hospital from June 9 to 15, 2023 (all patients underwent chest CT examination within 24 hours after diagnosis) were enrolled. The patients were divided into coronavirus infection non-pneumonia group and coronavirus infection associated pneumonia group according to CT findings. Relevant baseline data, such as demographic characteristics, chest CT characteristics, and laboratory indicators within 12 hours before and after CT examination were collected from each group. Spearman correlation test was used to quantitatively analyze the correlation between CT features and laboratory indicators. The receiver operator characteristic curve (ROC curve) was drawn to evaluate the predictive value of each laboratory index for pneumonia in patients infected with coronavirus. Multiple linear regression analysis was used to explore the relationship between different CT patterns such as ground-glass opacity (GGO) and consolidation and ventilatory oxygenation status.

Results: A total of 171 patients were enrolled, including 44 patients in the coronavirus infection non-pneumonia group and 127 patients in the coronavirus infection associated pneumonia group (the incidence of pneumonia was 74.3%). Compared with patients with coronavirus infection alone, patients with coronavirus infection associated pneumonia had significantly lower lymphocyte count (LYM), oxygenation index (PaO₂/FiO₂), total lung capacity, GGO volume and GGO ratio, and significantly higher C-reactive protein (CRP), neutrophil/lymphocyte ratio (NLR), D-dimer, fraction of inspired oxygen (FiO₂) level, real volume variation and consolidation ratio, the differences were all statistically significant. There were no statistically significant differences in the nucleocapin protein (N) gene cycle threshold (Ct) value and open reading frame (ORF) gene Ct value between the two groups. ROC curve analysis showed that, after adjusting for age, gender, CRP level and other related factors, compared with N gene Ct value, ORF gene Ct value, N gene Ct value+LYM, ORF gene Ct value+LYM, the LYM had the most potential diagnosis power for coronavirus infection associated pneumonia. The area under the ROC curve (AUC) of LYM for predicting coronavirus infection was 0.703. When the cut-off value of LYM was 0.7×10⁹/L, the sensitivity was 55.5%, and the specificity was 79.5%, respectively. Multiple linear regression analysis showed that, when adjusted for consolidation ratio, age, gender, Hb and D-dimer levels, the GGO ratio in patients with coronavirus infection associated pneumonia was correlated with PaO₂/FiO₂ (β = -2.18, P < 0.001). When adjusted for GGO ratio, age, sex, Hb and D-dimer levels, the proportion of consolidation in patients with coronavirus infection associated pneumonia was correlated with PaCO₂ (β = 0.36, P = 0.004). When adjusted for GGO ratio, the proportion of consolidation in patients with coronavirus infection associated pneumonia was also associated with NLR (β = 0.79, P = 0.006).

Conclusions: LYM could be a potential marker for predicting coronavirus associated pneumonia, and the correlation seems to be independent of viral load. In addition, in the analysis of imaging features, GGO is associated with hypoxia, while consolidation is associated with PaCO₂ level and inflammation. The increased proportion of consolidation in the whole lung may be detrimental to lung ventilation.