COVID-19 Prognosis via Visual Classification of Computed Tomography Lung Patterns

In patients with COVID-19, visual assessment of computed tomography (CT) images can predict the severity of illness, a resulting decrease in respiratory function, and the need for supplemental respiratory ventilation, according to a study published in BMC Pulmonary Medicine.

Among imaging markers, CT provides the most sensitive radiological technique for the diagnosis of COVID-19. While research using quantitative methods to evaluate lung CT images has suggested the possibility of predicting the severity of the disease, there are limited opportunities in clinical practice to use these quantitative image analysis approaches. Therefore, researchers conducted a retrospective study at a hospital in Tokyo to evaluate the utility of a novel visually-based classification of pulmonary findings from CT images of COVID-19 patients with the following 3 defined patterns: peripheral, multifocal, and diffuse findings of pneumonia. The researchers also evaluated the prognostic value of this classification to predict the severity of COVID-19.

The investigators found that patients with a diffuse pattern of pneumonia were intubated more frequently and for a longer duration than patients with a peripheral or multifocal pattern of pneumonia. They also found that body temperature (P=.04), lymphocyte count (P =.01), neutrophil count (P =.02), c-reactive protein (P <.01), lactate dehydrogenase (P <.01), Krebs von den Lungen-6 antigen (P <.01), D-dimer (P <.01), and steroid (P =.01) and favipiravir (P =.03) administration were significantly different between the diffuse pattern and peripheral and multifocal groups. Thus, these results suggest that the more extensive the abnormal findings in the lungs, the more severe the systemic over-inflammatory response in COVID-19.

“Our simple visual assessment of CT images can predict the severity of illness, a decrease in respiratory function, and the need for supplemental respiratory ventilation among patients with COVID-19,” concluded the authors.

Reference

Yamada D, Ohde S, Imai R, Ikejima K, Matsusako M, Kurihara Y. Visual classification of three computed tomography lung patterns to predict prognosis of COVID-19: a retrospective study. BMC Pulmonary Medicine. 2022;22(1):1. doi:10.1186/s12890-021-01813-y