Cîrjaliu Roxana-Elena, Gurrala Sri Vidhya, Nallapati Balaji, Krishna Vamsi, Oancea Cristian, Tudorache Emanuela, Marc Monica, Bratosin Felix, Bogdan Iulia, Rosca Ovidiu, Barata Paula Irina, Hangan Laurentiu Tony, Chirilă Sergiu Ioachim, Fildan Ariadna-Petronela
Faculty of Medicine, "Ovidius" University of Constanta, 900470 Constanta, Romania.
NRI Academy of Medical Sciences, NTR Health University, Mangalagiri Road, Chinakakani, Guntur, Andhra Pradesh 522503, India.
Diseases. 2024 Nov 8;12(11):285. doi: 10.3390/diseases12110285.
Pulmonary fibrosis detected during the acute phase of SARS-CoV-2 infection may significantly influence patient prognosis. This study aimed to evaluate the prognostic value of initial high-resolution computed tomography (HRCT) findings of pulmonary fibrosis in hospitalized COVID-19 patients and to examine how these findings relate to disease severity and clinical outcomes, with a particular focus on the development and validation of predictive scoring systems. In this multicentric prospective cohort study from January 2023 to January 2024, 120 adult patients with confirmed SARS-CoV-2 infection requiring hospitalization were enrolled from two Romanian university hospitals. Patients were categorized based on the presence (n = 60) or absence (n = 60) of pulmonary fibrosis signs on admission HRCT scans, identified by reticular opacities, traction bronchiectasis, honeycombing, and architectural distortion. Biochemical analyses, severity scores (SOFA, APACHE II, NEWS 2), and novel compound scores combining clinical and radiological data were assessed. Patients with HRCT evidence of pulmonary fibrosis had significantly higher severity scores and worse clinical outcomes. The HRCT score alone was a strong predictor of severe COVID-19 (area under the ROC curve [AUC] = 0.885), with a best cutoff value of 9.72, yielding 85.7% sensitivity and 79.8% specificity. Compound Score 1, integrating SOFA, APACHE II, and HRCT scores, demonstrated excellent predictive performance with an AUC of 0.947, sensitivity of 92.5%, and specificity of 88.9%. Compound Score 2, combining systemic inflammation markers (SIRI, SII) and NEWS 2, also showed a strong predictive capability (AUC = 0.913), with 89.2% sensitivity and 85.7% specificity at the optimal cutoff. Regression analysis revealed that Compound Score 1 had the highest hazard ratio for severe COVID-19 outcomes (HR = 4.89; 95% CI: 3.40-7.05), indicating its superior prognostic value over individual markers and traditional severity scores. Initial HRCT findings of pulmonary fibrosis are significantly associated with increased disease severity in hospitalized COVID-19 patients. The HRCT score is a valuable prognostic tool, and, when combined with clinical severity scores into Compound Score 1, it enhances the prediction of severe COVID-19 outcomes with high sensitivity and specificity. These compound scores facilitate the early identification of high-risk patients, guiding clinical decision-making and optimizing patient management to improve outcomes.
在新型冠状病毒2型（SARS-CoV-2）感染急性期检测到的肺纤维化可能会显著影响患者预后。本研究旨在评估住院的2019冠状病毒病（COVID-19）患者初始高分辨率计算机断层扫描（HRCT）肺纤维化表现的预后价值，并研究这些表现与疾病严重程度及临床结局的关系，特别关注预测评分系统的开发与验证。在这项于2023年1月至2024年1月开展的多中心前瞻性队列研究中，从罗马尼亚的两家大学医院招募了120例确诊SARS-CoV-2感染且需要住院治疗的成年患者。根据入院时HRCT扫描上是否存在肺纤维化征象（n = 60）将患者分类，肺纤维化征象通过网状影、牵拉性支气管扩张、蜂窝状改变和结构扭曲来确定。评估了生化分析、严重程度评分（序贯器官衰竭评估[SOFA]、急性生理与慢性健康状况评分系统II[APACHE II]、国家早期预警评分2[NEWS 2]）以及结合临床和放射学数据的新型复合评分。有HRCT证据显示存在肺纤维化的患者严重程度评分显著更高，临床结局更差。单独的HRCT评分是重症COVID-19的有力预测指标（ROC曲线下面积[AUC]=0.885），最佳截断值为9.72，灵敏度为85.7%，特异度为79.8%。整合了SOFA、APACHE II和HRCT评分的复合评分1显示出优异的预测性能，AUC为0.947，灵敏度为92.5%，特异度为88.9%。结合全身炎症标志物（全身炎症反应指数[SIRI]、全身免疫炎症指数[SII]）和NEWS 2的复合评分2也显示出较强的预测能力（AUC = 0.913），在最佳截断值时灵敏度为89.2%，特异度为85.7%。回归分析显示，复合评分1对重症COVID-19结局的风险比最高（风险比[HR]=4.89；95%置信区间[CI]：3.40 - 7.05），表明其在预后价值上优于个体标志物和传统严重程度评分。住院COVID-19患者初始HRCT肺纤维化表现与疾病严重程度增加显著相关。HRCT评分是一种有价值的预后工具，当与临床严重程度评分结合形成复合评分1时，可提高对重症COVID-19结局的预测，具有高灵敏度和特异度。这些复合评分有助于早期识别高危患者，指导临床决策并优化患者管理以改善结局。
