Temporal Radiographic Trajectory and Clinical Outcomes in COVID-19 Pneumonia: A Longitudinal Study

Area of Science:

• Radiology
• Pulmonology
• Critical Care Medicine

Background:

• Limited understanding of COVID-19 pneumonia's radiographic trajectory and its impact on clinical outcomes.
• Need to correlate temporal changes in pneumonia extent with patient prognosis.

Purpose of the Study:

• To identify distinct radiographic trajectories of COVID-19 pneumonia.
• To analyze differences in laboratory biomarkers, critical care utilization, and clinical outcomes based on these trajectories.

Main Methods:

• Utilized deep learning to quantify pneumonia extent from serial chest radiographs in 2,385 COVID-19 patients.
• Employed latent class growth modeling to define radiographic trajectories.
• Applied Cox regression to assess critical care needs and outcomes.

Main Results:

• Three pneumonia trajectories identified: steady (80.7%), downhill (5.7%), and uphill (13.6%).
• The uphill and downhill groups showed higher risks for critical care support and ICU admission.
• The uphill group had a significantly higher risk of in-hospital mortality (HR, 8.2).

Conclusions:

• Radiographic trajectories of COVID-19 pneumonia correlate with temporal changes in BUN and CRP.
• These trajectories predict critical care needs and clinical outcomes.
• Tailoring therapeutic strategies based on identified disease progression patterns is recommended.