A novel and the most notable finding of this study is that a considerable proportion of the patients with severe COVID-19 pneumonia had subclinical interstitial lung lesions, and the frequency was significantly higher than that in the control cases. This finding strongly supports our hypothesis that severe COVID-19 pneumonia could be an exacerbation of subclinical ILD triggered by SARS-CoV-2 infection.
A previous study indicated that small interstitial lung lesions with the UIP pattern (suggestive of early IPF) were often detected in the lung bases of patients with idiopathic acute interstitial pneumonia, and the authors suggested that idiopathic acute interstitial pneumonia could be only an exacerbation of preexisting subclinical ILDs (IPF) [21–24]. Additionally, ILDs, particularly those with the UIP pattern (IPF), have been associated with exacerbation, even if they are only early and focal diseases [22, 24, 29]. In that case, infectious diseases are a common trigger, and viruses such as influenza, human herpes, and cytomegalovirus have been reported to be involved [30–33]. Further, regarding pathogenetic mechanisms, there is an association between severe COVID-19 pneumonia and ILD exacerbation. Initially, COVID-19 pneumonia was considered to be mainly caused by direct cellular damage of type 2 pneumocytes by SARS-CoV-2 infection [34, 35]. However, immunohistochemical studies have shown that virus-infected cells were detected only focally, although DAD affected the entire lungs [36, 37]. Therefore, there should be another mechanism, such as excessive immunological responses. Notably, elevations in the levels of several cytokines have been reported [38–40], and it is currently widely accepted that some cytokines (e.g.; IL-6, IL-8 and others) is related to the development of severe COVID-19 pneumonia [38–40]. Moreover, it is considered that the same cytokines (IL-6 and IL-8) are also essential to trigger ILD exacerbation [41–43]. Thus, these observations seem to support our hypothesis that SARS-CoV-2 could trigger the exacerbation of hidden ILD and the development of DAD through the hyperactivation of cytokines, that is, severe COVID-19 pneumonia.
In addition to the pathological examination, we performed radiological examination using CT images from another series of patients with COVID-19. The results of that analysis—patients with interstitial lesions on CT images had more severe disease than those who did not—further support our hypothesis. Recently, a radiological term “interstitial lung abnormalities (ILA)” was proposed to describe subclinical hidden ILDs on CT images [44, 45]. ILA is defined as an interstitial abnormality detected in patients without a clinical history of ILDs. ILA is seen in at least 5% of the field in any slice of a whole lung CT image [44]. Studies have reported that some cases (up to 70%) of ILA progressed to clinical disease (i.e., equivalent to overt ILDs) [46–49]. Studies on the radiological-pathological associations in interstitial lung lesions suggested that a considerable fraction of ILAs could include pathological UIP lesions [47, 50]. We defined “s/rILD” as an area (≥ 10 mm) of reticular opacities (occasionally with cystic changes = traction bronchial ectasia) in the bilateral lung bases. Our definition of “s/rILD” may be the same as that of “ILA” in a broad sense. In any case, subclinical ILD (conceptually equivalent to ILA) may be a risk factor for severe COVID-19 pneumonia. This is a novel finding of the present study, and we believe it is important to understand the potential pathological bases of DAD in various situations.
Limitations
The limitations of this study are 1) we could not examine a large number of autopsy cases because of infection control purposes; 2) in autopsy cases associated with long disease periods, it was difficult to distinguish preexisting hidden interstitial lesions from COVID-19-related scarring lesions; 3) we could not histologically examine lung tissues from patients with COVID-19 without severe pneumonia as controls; and 4) in the radiological study, hidden interstitial lesions may not have been detectable in the cases with extensive DAD and not all cases could be evaluated by high-resolution computed tomography. We understand the influence of these limitations on the results.