Several studies have been conducted on eosinophilic airway inflammation caused by respiratory viral infections. In fundamental medical research, respiratory syncytial virus infection is often reported to be involved in eosinophilic airway inflammation [4–6]. Clinical studies have demonstrated eosinophilic inflammation due to respiratory virus infections using upper respiratory tract specimens, such as nasal secretions [7, 8]; however, clinical studies using lower respiratory tract specimens, such as BAL, are rare. There have been small studies on rhinovirus and respiratory syncytial virus, showing slight elevation of the BAL eosinophil fraction due to infection [9, 10]. However, except for a few case reports of influenza (H1N1) and COVID-19 [11–13], there are no reports of cases of eosinophilic pneumonia diagnosed with significant BAL eosinophil fraction elevation due to respiratory virus infection.
Eosinophilic airway inflammation caused by respiratory viral infections is considered a protective response to respiratory viruses, and basic medical and clinical studies have shown that eosinophilic inflammation is protective against respiratory viral infections [14–17]. Eosinophils produce and contain molecules with antiviral activity, such as RNase and active nitrogen species [16]. In addition, eosinophils can capture and inactivate viruses [17]. Therefore, the increase in the eosinophil fraction in BAL fluid due to respiratory virus infection is a rational response to protect against respiratory viruses. Note that eosinophilic airway inflammation caused by respiratory virus infection is considered relatively mild. For example, eosinophilic inflammation caused by parasitic infections is more severe than that caused by respiratory viral infections.
In the present study, the elevation of the BAL eosinophil fraction in the PCR-positive group was slight, whereas two patients with positive results for parainfluenza virus had a highly elevated eosinophil fraction in BAL and were diagnosed with eosinophilic pneumonia. In these two patients, eosinophilic pneumonia was probably caused by parainfluenza infection since there was no other obvious cause. To the best of our knowledge, these patients are the first reported cases of eosinophilic pneumonia caused by parainfluenza virus. Basic studies have shown that eosinophils in guinea pig BAL fluid are also increased by infection with parainfluenza virus [18], and in two case reports of parainfluenza virus pneumonia, the eosinophil fraction in BAL fluid was slightly elevated [19]. Thus, reports of eosinophilic airway inflammation due to parainfluenza infection are not uncommon and are consistent with the causes of eosinophilic pneumonia, as in the cases in this study. Recently, it was reported that respiratory viruses are often detected by PCR testing in cases of interstitial lung disease, including eosinophilic pneumonia [20]. The present study suggests that parainfluenza virus infection may be a cause of eosinophilic pneumonia. It is likely that there are many more cases of eosinophilic pneumonia caused by respiratory viral infections in clinical practice. Highly sensitive multiplex testing will further clarify the relationship between respiratory viral infections and allergic respiratory diseases.
There were obvious limitations to this study. First, this was a retrospective single-center study. PCR testing was performed as a routine test on BAL specimens during the study period. However, it was not mandatory and may not have been performed on cases deemed unnecessary by the attending physician. The number of such cases is unknown. It is possible that there was a bias in the backgrounds of the cases included in this study or in the respiratory viruses prevalent in our region. Second, the number of positive cases was too small to properly perform multivariate analysis of all possible confounders. Third, the degree of eosinophilic airway inflammation is different for different respiratory viruses. A previous report showed that respiratory syncytial virus infection caused higher eosinophil counts than metapneumovirus infection in peripheral blood [21]. When comparing rhinovirus and influenza virus, infection of the former was higher in peripheral blood [22]. Additionally, some respiratory viruses, adenovirus and bocavirus, probably do not cause eosinophilic inflammation [23, 24]. However, further research on respiratory viruses is needed.