In the present study, we compared the aetiology of CAP among patients with HM before and during the COVID-19 pandemic. During the pandemic, numerous studies have focused on the incidence and treatment of COVID-19 in immunocompromised patients; however, limited research has been conducted on aetiologies of CAP other than COVID-19, particularly on changes in them, in such patients. Nonetheless, considering the susceptibility of immunocompromised patients to various opportunistic infections, it is crucial to examine the changes in pathogens that can cause CAP in the COVID-19 era.
In contrast to the findings of previous studies that reported a decrease in the incidence of CAP in the COVID-19 era [14–17], in this study, the number of patients with HM hospitalised with CAP in the COVID-19 era did not decrease compared with that during the pre-COVID-19 era. Notably, the incidence of viral pneumonia demonstrated a significant decrease; however, the incidence of bacterial and fungal pneumonia remained unaffected, in contrast to other research findings. Although previous studies have reported a decline in the incidence of bacterial pneumonia [14], this study has not exhibited any decrease in its incidence during the COVID-19 pandemic. The unchanged incidence of bacterial pneumonia despite the rigorous infection control measures adopted during the COVID-19 pandemic indicates that the most significant risk factor for bacterial CAP in patients with HM is their immune status [18].
The incidence of pneumonia caused by respiratory viruses, including influenza A, parainfluenza, adenovirus, and rhinovirus, decreased significantly during the COVID-19 pandemic. Bilateral pulmonary infiltration on chest radiographs also showed a significant decrease during the COVID-19 pandemic, which was hypothesised to be a result of a decreased incidence of viral pneumonia [19]. Respiratory viruses usually cause mild upper respiratory tract infections; however, previous studies have reported that respiratory viruses are important pathogens causing CAP in immunocompromised patients and older adults [20, 21]. As reported in other studies, the implementation of various preventive measures following COVID-19 outbreaks is likely to not only curtail the transmission of COVID-19 but also reduce the incidence of viral pneumonia caused by other respiratory viruses [2–4]. In this study, we observed an increase in the incidence of RSV during the winter of 2021, which is likely attributable to the RSV outbreak in South Korea due to the relaxation of domestic infection control measures in November 2021 [22].
CAP of unknown aetiology is relatively common among immunocompromised patients [23, 24]. In this study, a substantial number of patients (39.3%) presented with CAP of unknown aetiology, despite our only including patients who underwent BRS to minimise the number of patients with an undetermined aetiology. The incidence of CAP of unknown aetiology did not decrease during the COVID-19 pandemic; however, this may not necessarily indicate that these cases were not caused by pathogens such as viruses or bacteria. The lack of testing for atypical infections and diagnostic limitations of laboratory techniques prevents us from drawing a definitive conclusion. It is plausible that these cases were associated with the patients' underlying diseases or immunocompromised status, but further investigation is needed to establish a causal relationship.
We investigated the risk factors for in-hospital mortality in patients with HM who had CAP. Consistent with previous studies [25, 26], high SOFA scores and low platelet counts were found to be risk factors for in-hospital mortality in patients with HM who had CAP. In general, while the CURB-65 (confusion, uraemia, respiratory rate, blood pressure, age ≥ 65 years) criteria or pneumonia severity index (PSI) are commonly used to predict the prognosis of patients with pneumonia, previous studies have shown that they are not effective in immunocompromised patients with cancer [27]. Thus, we opted to use the SOFA score instead. Thrombocytopenia is also associated with the severity of and poor prognosis in patients with infections [28, 29]. Platelets play a role in inflammation and host defence mechanisms against microbial agents. Additionally, thrombocytopenia reflects bone marrow failure in patients with HM [30]. Therefore, thrombocytopenia could be a significant predictor of poor prognosis in patients with HM and CAP.
This study has several limitations. First, its retrospective design and single-centre implementation may introduce a selection bias, potentially affecting the significance of our findings. Nevertheless, we meticulously examined all admitted patients with CAP who underwent BRS in this single, large, 4-year cohort with consistent treatment protocols. The objective of our study was to assess the impact of the COVID-19 pandemic on the changing pattern of aetiology of CAP in patients with HM; thus, the setting of this study did not significantly deviate from that of a prospective observational study. Second, conducting a single-centre study may have had limitations in determining the overall aetiology of CAP in all patients with HM. However, this institution serves as the largest Asian centre for haematological disorders, performing over 500 bone marrow transplantations annually. Therefore, these data can be considered reasonably representative. Third, patients’ COVID-19 histories were not investigated. Previous studies reported that COVID-19 is a risk factor for bacterial or fungal co-infections and can play a significant role in patient prognosis [31, 32]. However, due to the retrospective nature of the study, it was not possible to conduct an investigation. Additional studies are needed to evaluate the impact of a COVID-19 history on the prognosis of patients with CAP and HM.