Repeated Lung Ultrasound versus Chest X-ray-Which One Predicts Better Clinical Outcome in COVID-19?

Evaluation of a Lung Ultrasound Score in Hospitalized Adult Patients with COVID-19 in Barcelona, Spain

Background/Objectives: During the COVID-19 pandemic and the burden on hospital resources, the rapid categorization of high-risk COVID-19 patients became essential, and lung ultrasound (LUS) emerged as an alternative to chest computed tomography, offering speed, non-ionizing, repeatable, and bedside assessments. Various LUS score systems have been used, yet there is no consensus on an optimal severity cut-off. We assessed the performance of a 12-zone LUS score to identify adult COVID-19 patients with severe lung involvement using oxygen saturation (SpO2)/fractional inspired oxygen (FiO2) ratio as a reference standard to define the best cut-off for predicting adverse outcomes. Methods: We conducted a single-centre prospective study (August 2020-April 2021) at Hospital del Mar, Barcelona, Spain. Upon admission to the general ward or intensive care unit (ICU), clinicians performed LUS in adult patients with confirmed COVID-19 pneumonia. Severe lung involvement was defined as a SpO2/FiO2 ratio <315. The LUS score ranged from 0 to 36 based on the aeration patterns. Results: 248 patients were included. The admission LUS score showed moderate performance in identifying a SpO2/FiO2 ratio <315 (area under the ROC curve: 0.71; 95%CI 0.64-0.77). After adjustment for COVID-19 risk factors, an admission LUS score ≥17 was associated with an increased risk of in-hospital death (OR 5.31; 95%CI: 1.38-20.4), ICU admission (OR 3.50; 95%CI: 1.37-8.94) and need for IMV (OR 3.31; 95%CI: 1.19-9.13). Conclusions: Although the admission LUS score had limited performance in identifying severe lung involvement, a cut-off ≥17 score was associated with an increased risk of adverse outcomes. and could play a role in the rapid categorization of COVID-19 pneumonia patients, anticipating the need for advanced care.

Lung Ultrasound Score in COVID-19 Patients Correlates with PO2/FiO2, Intubation Rates, and Mortality

Introduction

The point-of-care lung ultrasound (LUS) score has been used in coronavirus 2019 (COVID-19) patients for diagnosis and risk stratification, due to excellent sensitivity and infection control concerns. We studied the ratio of partial pressure of oxygen in arterial blood to the fraction of inspiratory oxygen concentration (PO2/FiO2), intubation rates, and mortality correlation to the LUS score.

Methods

We conducted a systematic review using PRISMA guidelines. Included were articles published from December 1, 2019-November 30, 2021 using LUS in adult COVID-19 patients in the intensive care unit or the emergency department. Excluded were studies on animals and on pediatric and pregnant patients. We assessed bias using QUADAS-2. Outcomes were LUS score and correlation to PO2/FiO2, intubation, and mortality rates. Random effects model pooled the meta-analysis results.

Results

We reviewed 27 of 5,267 studies identified. Of the 27 studies, seven were included in the intubation outcome, six in the correlation to PO2/FiO2 outcome, and six in the mortality outcome. Heterogeneity was found in ultrasound protocols and outcomes. In the pooled results of 267 patients, LUS score was found to have a strong negative correlation to PO2/FiO2 with a correlation coefficient of -0.69 (95% confidence interval [CI] -0.75, -0.62). In pooled results, 273 intubated patients had a mean LUS score that was 6.95 points higher (95% CI 4.58-9.31) than that of 379 non-intubated patients. In the mortality outcome, 385 survivors had a mean LUS score that was 4.61 points lower (95% CI 3.64-5.58) than that of 181 non-survivors. There was significant heterogeneity between the studies as measured by the I2 and Cochran Q test.

Conclusion

A higher LUS score was strongly correlated with a decreasing PO2/FiO2 in COVID-19 pneumonia patients. The LUS score was significantly higher in intubated vs non-intubated patients with COVID-19. The LUS score was significantly lower in critically ill patients with COVID-19 pneumonia that survive.

MSTAC: A Multi-Stage Automated Classification of COVID-19 Chest X-ray Images Using Stacked CNN Models

This study introduces a Multi-Stage Automated Classification (MSTAC) system for COVID-19 chest X-ray (CXR) images, utilizing stacked Convolutional Neural Network (CNN) models. Suspected COVID-19 patients often undergo CXR imaging, making it valuable for disease classification. The study collected CXR images from public datasets and aimed to differentiate between COVID-19, non-COVID-19, and healthy cases. MSTAC employs two classification stages: the first distinguishes healthy from unhealthy cases, and the second further classifies COVID-19 and non-COVID-19 cases. Compared to a single CNN-Multiclass model, MSTAC demonstrated superior classification performance, achieving 97.30% accuracy and sensitivity. In contrast, the CNN-Multiclass model showed 94.76% accuracy and sensitivity. MSTAC's effectiveness is highlighted in its promising results over the CNN-Multiclass model, suggesting its potential to assist healthcare professionals in efficiently diagnosing COVID-19 cases. The system outperformed similar techniques, emphasizing its accuracy and efficiency in COVID-19 diagnosis. This research underscores MSTAC as a valuable tool in medical image analysis for enhanced disease classification.

Long COVID in Young Patients: Impact on Lung Volume Evaluated Using Multidetector CT

PURPOSE

To evaluate using quantitative analysis on chest CT images a possible lung volume reduction in Long COVID patients who complain mild respiratory symptoms, with chest CT negative for inflammatory findings.

MATERIALS AND METHODS

CT images of patients from 18 to 40 years old who underwent chest CT scan at our institution were analyzed retrospectively, using AwServer Thoracic VCAR software for a quantitative study. Exclusion criteria were inflammatory findings at CT, previous lung surgery, lung cancer, and breath artifacts that invalidate the quality of images. Patients were divided into two groups: in the first one ("post-COVID") were patients who had previous SARS-CoV-2 infection, confirmed by an RT-PCR, who underwent chest CT from 3 to 6 months after their negativization for long COVID symptoms; in the control group ("non-COVID"), were enrolled patients who underwent a chest CT scan from January 2018 to December 2019, before the spread of COVID in Italy.

RESULTS

Our final population included 154 TC, 77 post-COVID patients (mean age 33 ± 6) and 77 non-COVID patients (mean age 33 ± 4.9). Non statistical significative differences were obtained between groups in terms of age, sex, and other characteristics that affect total lung capacity such as obesity, thoracic malformations, and smoking habit. Mean values of the total lung volume (TV), right-lung volume (RV), and left-lung volume (LV) in the post-COVID group compared with non-COVID group were, respectively: 5.25 ± 0.25 L vs. 5.72 ± 0.26 L (p = 0.01); 2.76 ± 0.14 L vs. 3 ± 0.14 L (p = 0.01); 2.48 ± 0.12 L vs. 2.72 ± 0.12 L (p = 0.01).

CONCLUSION

In patients with symptoms suggesting Long COVID and negative chest CT macroscopic findings, quantitative volume analysis demonstrated a mean value of reduction in lung volume of 10% compared to patients of the same age who never had COVID. A chest CT negative for inflammatory findings may induce clinicians to attribute Long COVID mild respiratory symptoms to anxiety, especially in young patients. Our study brings us beyond appearances and beyond the classic radiological signs, introducing a quantitative evaluation of lung volumes in these patients. It is hard to establish to what extent this finding may contribute to Long COVID symptoms, but this is another step to gain a wider knowledge of the potential long-term effects caused by this new virus.