DISCUSSION
Secondary pulmonary is the basic unit of pulmonary anatomic function, which has been playing an important role in HRCT(High-resolution computed tomography). The imaging features of secondary pulmonary lobular structure run through all stages of the image performance of COVID-19. According to the anatomical differences of pulmonary lobules, Fleischner and Heitzman proposed that the lung can be divided into peripheral pulmonary cortex and central pulmonary medulla(4,5). Webb put forward the concept of cortical lobule, highlighting the anatomical characteristics of pulmonary cortical lobule. The characteristics of the pulmonary cortical lobule also lay the anatomical basis for the distribution of COVID-19 under the pleura (6,7).Therefore, it is critical to discern the characteristics of lobules and their distribution for accurate diagnosis of COVID-19.
The CT features of viral pneumonia, such as lobular size, fusion lobule, thickening of intralobular interstitium and interlobular septum, that have been respectively reported in the related literatures of influenza and SARS(Severe acute respiratory syndrome)(8-12). In particular, COVID-19 has a virus homology with SARS. The image of COVID-19 is highly similar to the lobular image feature reported by SARS(13-16). Therefore, the evaluation method and scoring standard of five categories of secondary pulmonary lobular structure image based on the anatomical structure of secondary pulmonary lobule were proposed for the first time in this study. The lobular size sign in COVID-19 often shows GGO or consolidation of 1-3centimeters, which is consistent with the results of the lobular size of SARS by Wong. (13). The fusion lobular sign is defined as the shape of wedge-shaped, fan-shaped, irregular or bat wing with the size between 3 centimeters and lung lobe, which does not follow the lung segments or lobe. In the early stage, the interlobular septal thickening sign is generally not obvious, only to make the edge of the lesion clearer, and transit to grid shape in the advanced or the recovery stage. The interlobular interstitial thickening sign likes “paving stone sign” which is very characteristic, and is very specific in the diagnosis of COVID-19. The interlobular vascular thickening sign is more common in the peripheral pulmonary cortex lobular blood vessels, particularly in the early stage of COVID-19.
From the perspective of image thinking, CT images of different types COVID-19 mainly include typical and atypical CT features, and the evaluation of dynamic stages. In this study, combined with the typical imaging diagnostic standards of COVID-19 that were reported by domestic imaging diagnosis guidelines and some literatures (16-18), as well as a summary of 103 confirmed cases, 82(79.61%) single-stage and 12(11.65%) multi-stage cases of the typical COVID-19 have three characteristics. These three typical performances of COVID-19 are consistent with some reports (19-21), but the ideas and concerns of authors are based on the image signs of secondary pulmonary lobular structure, and the analysis from density, size and distribution is more in line with the traditional image thinking mode. As a new infectious disease, COVID-19 has not been fully recognized so far. In our study, we found that the image manifestations of atypical cases are mostly GGO or consolidation, including single or mixed GGO, thickening of bronchovascular bundle with peripheral GGO, mixed GGO in lobes or segments, and halo sign or anti halo sign in a few cases. These findings are consistent with the relevant literature (20-22).
In this study, there are two types of differential diagnosis of COVID-19: infectious disease(other viral pneumonia, bacterial pneumonia, mycoplasma infection, cryptogenic organizing pneumonia, eosinophilic pneumonia) and non-infectious disease(pulmonary hemorrhage, pulmonary edema and traumatic wet lung). To some extent, the imaging manifestations of these diseases may overlap with COVID-19, so it is great significance to combine the imaging signs of secondary pulmonary lobular structure for differential diagnosis between COVID-19 and non-COVID-19. However, there are some limitations in this study:Firstly, there are few cases of non-COVID -19, resulting in partial statistical results bias.Secondly, there is no differential diagnosis between COVID-19 and other relatively certain diseases.Thirdly, in COVID-19 cases, the correlation between secondary pulmonary lobular structural signs and related laboratory indicators was not studied, such as C-reactive protein. This is also the research direction of the next step after increasing the number of samples.