Exploring the role of respiratory virus infections in aspiration pneumonia: a comprehensive analysis of cases with lower respiratory tract infections

Abstract

Background

While respiratory virus infection has been implicated in the onset of bacterial pneumonia, no research has investigated the association of respiratory viruses with the onset of aspiration pneumonia (AP). This study aimed to investigate the role of respiratory virus infections in AP.

Methods

Patients presenting with acute respiratory symptoms and undergoing influenza antigen testing at the emergency department of Okinawa Chubu Hospital from February 2020 to January 2021, and diagnosed with lower respiratory tract infections, were included. Cases were categorized into AP, pneumonia other than AP (non-AP), and acute bronchitis (AB) based on physician diagnoses recorded in medical records. The residual nasal swab specimens were further tested with multiplex PCR tests for respiratory viruses.

Results

A total of 209 subjects were included in the study: 59 in the AP group, 118 in the non-AP group, and 32 in the AB group. The AP group was characterized by older age, higher rates of nursing home residency, a greater prevalence of comorbidities such as cerebrovascular disease and dementia, a lower sputum culture positivity rate, and a different spectrum of causative pathogens compared to the other groups. The virus positivity rate in the AP group was 47%, compared to 50% in the non-AP group and 53% in the AB group, with no significant difference observed. The AP group exhibited the highest rate of only respiratory viruses detected and the lowest rate of both respiratory viruses and bacteria detected among the groups. There was no significant difference in the types of viruses detected between the AP group and the other groups, with rhinovirus being the most frequently detected virus across all groups. In the AP group, virus-negative cases were significantly older on average. No other significant differences in background, symptoms, or clinical data were observed between virus-positive and virus-negative cases within the AP group.

Conclusion

In the AP group, the rate of respiratory virus detections was comparable to that of the non-AP and AB groups. This suggests a potential link between respiratory virus infections and the development of AP, emphasizing the need for novel preventive strategies. While distinguishing between AP patients with and without respiratory virus detections based on clinical findings was challenging, recognizing the frequent involvement of respiratory virus infections in AP highlights the importance of enhanced infection control and awareness in its management.

Background

Aspiration pneumonia (AP) is a category of pneumonia resulting from dysphagia, and it is a frequent cause of hospitalization and mortality among elderly individuals [1, 2]. AP accounts for 5–15% of community-acquired pneumonia (CAP) cases [1, 3], with the majority occurring in the elderly population [4, 5]. In Japan, a super-aging society, AP constitutes nearly 40% of CAP cases [6]. As the global population ages, the proportion of AP within CAP is expected to increase, highlighting the importance of understanding the pathophysiology of AP.

Epidemiological studies have shown that respiratory viruses are detected in 20–30% of CAP cases [7,8,9,10,11,12]. Respiratory virus infections facilitate bacterial pneumonia by promoting the invasion, colonization, and proliferation of common CAP bacterial pathogens, such as Streptococcus pneumoniae and Haemophilus influenzae, in the lower respiratory tract [13, 14]. This rationalizes the frequent detection of respiratory viruses in CAP cases. However, whether this is also true for AP remains unclear. The pathogenesis of AP is typically related to aspiration, where dysphagia leads to aspiration of oropharyngeal contents, The pathogenesis of AP is typically related to aspiration, where dysphagia leads to the aspiration of oropharyngeal contents, which initially causes aspiration pneumonitis—a condition where bacterial infection has not yet occurred—and subsequently to bacterial infection in the lower respiratory tract, resulting in AP [15, 16]. Patients with AP have distinct medical histories and social backgrounds, such as nursing home residency and a history of stroke or neurological disorders, compared to those with CAP excluding AP [17]. Consequently, it remains unclear whether respiratory virus infections play a similar role in AP as in typical bacterial pneumonia. Reports on the relationship between AP and respiratory virus infections are rare. As the global population ages, the proportion of AP cases within CAP is expected to rise, underscoring the need to investigate the prevalence of respiratory virus infections in AP and their clinical relevance.

Before the coronavirus disease 2019 (COVID-19) pandemic, influenza virus infections occurred year-round in Okinawa, Japan. Consequently, in community hospital emergency departments in Okinawa, influenza antigen tests were routinely performed for patients presenting with acute respiratory symptoms, regardless of the season, even in cases of AP. In this study, we analyzed residual samples from these influenza tests for other respiratory viruses to investigate the extent to which respiratory virus infections are involved in cases of AP.

Methods

Patients and data collection

This retrospective study reviewed medical records to identify cases of AP. Simultaneously, cases of non-AP lower respiratory tract infections were extracted for comparison. The study population consisted of patients aged 20 years or older who visited the emergency department of Okinawa Chubu Hospital between February 2020 and January 2021 and underwent a rapid influenza antigen test. Among these patients, those with a physician diagnosis recorded in the medical chart at the time of the emergency department visit as AP, pneumonia other than aspiration pneumonia (non-AP), or acute bronchitis (AB) were included. The criteria for AP and non-AP in this study required the presence of respiratory symptoms and/or fever, new infiltrates on chest X-ray, and a documented diagnosis of either 'aspiration pneumonia' or 'bacterial pneumonia' in the medical records by the attending physician. The criteria for AB included the presence of respiratory symptoms or fever, no new infiltrates on chest X-ray, and a documented diagnosis of 'acute bronchitis' in the medical records. AP cases were defined as those diagnosed by physicians based on the patient's medical history, interview, and physical examination findings [18], in line with prior studies [19]. These initial diagnoses were made by emergency physicians and subsequently reviewed and confirmed by 2 to 4 respiratory specialists. Cases where the initial diagnosis in the medical chart was later revised to conditions other than AP, non-AP, or AB were excluded from the study. After excluding cases, the final included cases were categorized into the AP group, non-AP group, and AB group, respectively. During this period, the hospital had a policy of only performing COVID-19 testing, rather than influenza antigen testing, on patients suspected of having COVID-19. Consequently, samples from these patients were not collected for this study.

Background data, symptoms, physical examination findings, and sputum bacterial culture results at the visit were extracted from the medical records. Routine sputum bacterial culture tests were conducted in the hospital laboratory. Specimens with a saliva-like or serous appearance were routinely discarded by the laboratory without undergoing culture tests. The culture media used included 5% sheep blood agar (Becton Dickinson Japan, Japan), chocolate agar (Becton Dickinson, Japan), and Drigalski-modified agar (Eiken Chemical, Japan). The cultures were incubated at 37 °C in a 5% CO₂ atmosphere for 48 h. Bacterial identification was conducted using the MALDI Biotyper®sirius (Bruker Daltonics Japan, Japan). Antimicrobial susceptibility testing was performed using dry plates (Eiken Chemical, Japan). Oral streptococci were classified as normal flora and were not considered significant pathogens. Anaerobic cultures were not performed routinely. For AP and non-AP cases, the presence of consolidations on chest X-rays at the visit was also extracted from the medical records. As part of standard practice in the hospital, chest X-ray findings were reviewed by multiple staff physicians, including emergency doctors and 2 to 4 respiratory specialists.

PCR testing

The ImunoAce® Flu test kit (Tauns Laboratories, Japan) was utilized. The swab specimens were individually packaged immediately after collection from patients and transported to the hospital laboratory for processing. The residual nasal swab from the rapid influenza antigen testing was then used for PCR testing. These specimens were repackaged and transported in containers to the research laboratory, where they temporarily stored at 4 °C. RNA extraction was performed using the QIAamp® MinElute® Virus Spin Kit (QIAGEN, Netherlands) at the laboratory of the University of the Ryukyus, and the samples were subsequently stored at −80 °C. For polymerase chain reaction (PCR) testing, the samples were thawed and analyzed using the Anyplex™ II RV16 Detection kit (Seegene, Korea). This kit is can detect adenovirus, seasonal coronaviruses, parainfluenza viruses, human rhinovirus, respiratory syncytial virus, metapneumovirus, enterovirus, influenza viruses, and bocavirus. The overall sensitivities and specificities were reported to be 95% and > 98%, respectively [20]. Additionally, to account for the potential inclusion of COVID-19 cases, PCR testing for severe acute respiratory syndrome coronavirus 2 was also conducted using the Allplex™ 2019-nCoV Assay (Seegene, Korea). The sensitivity and negative predictive value were reported to be 98% and 97% [21]. Both the specimen collectors and the testers were required to wear masks and gloves.

Outcomes

The primary outcome was the rate of viral infections in AP cases compared to the non-AP and AB groups. Secondary outcomes included the comparison of the detection rates of viral species and bacterial pathogens among the groups, and the comparison of clinical features between virus-positive and virus-negative cases within each group.

Statistical analysis

In the two-group comparisons, categorical variables were evaluated using the Chi-square test or Fisher's exact test, and continuous variables were assessed using the t-test or Mann–Whitney U test. For the three-group comparisons, categorical variables were evaluated using the Chi-square test or likelihood ratio test, and continuous variables were analyzed using the Kruskal–Wallis test. All statistical analyses were performed using IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk, NY, USA).

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki (2013 revision) and the Ethical Guidelines for Medical and Biological Research Involving Human Subjects (Japanese Ministry of Education, Culture, Sports, Science and Technology & Ministry of Health, Labour and Welfare, 2014; revised 2017). The study protocol was approved by the ethics committees of Okinawa Chubu Hospital (2019–63) and the University of the Ryukyus (R2-1628). The study utilized residual specimens from routine medical care, which would have otherwise been discarded, imposing no additional burden on participants. Given its retrospective design and use of anonymized data, the requirement for individual informed consent was waived by the ethics committees. Instead, an opt-out approach was implemented by posting study-related information on the hospital’s website, allowing eligible patients—those who underwent influenza antigen testing in the emergency department during the study period—to request exclusion. To ensure privacy protection, only anonymized data were used for analysis, and all collected data were securely stored in a locked facility at Okinawa Chubu Hospital. If a participant opted out, their data were excluded and appropriately discarded.

Results

A total of 209 subjects were included in the study, with 59 in the AP group, 118 in the non-AP group, and 32 in the AB group (Fig. 1). The AP group had several characteristics that differed significantly from the other two groups: a higher median age, a higher rate of residence in nursing homes or use of day-care facilities, a higher prevalence of past central nervous system diseases and/or mental disorders such as stroke or dementia, and a lower sputum culture positivity rate. Although the hospitalization rate was not significantly different from that of the non-AP group, it was close to 100%. The proportion of cases with consolidation on chest X-ray was significantly lower in the AP group compared to the non-AP group (Table 1). These characteristics of the AP group are consistent with the commonly recognized features of AP [17, 22].

Fig. 1

Flowchart of registration. The medical records of patients who underwent influenza antigen testing during the study period were reviewed, and cases diagnosed with pneumonia or acute bronchitis in the emergency department were extracted. The classification of aspiration pneumonia and non-aspiration pneumonia was also based on the medical record entries. Cases in which the diagnosis recorded in the medical records changed during the subsequent clinical course were excluded. *Exclusion of cases in which the diagnosis recorded in the medical records changed during the subsequent clinical course. Abbreviations: AP, aspiration pneumonia; non-AP, pneumonia except for aspiration pneumonia; AB, acute bronchitis

Table 1 Characteristics of each group

The differences in detected pathogens between the AP group and the other groups were clear in terms of bacterial species. The bacterial species detected in the AP group, such as Klebsiella pneumoniae and Pseudomonas aeruginosa, were consistent with those typically found in HAP, while those in the non-AP and AB groups, including Haemophilus influenzae, Moraxella catarrhalis, and Streptococcus pneumoniae, were consistent with pathogens typically found in CAP [23]. However, there was no noticeable difference in the types of infected viruses between the AP group and the other groups. In all groups, rhinovirus was the most frequently detected virus, with other viral species being detected in smaller numbers (Table 1).

The virus positivity rate in the AP group was 47%, compared to 50% in the non-AP group and 53% in the AB group, with no significant difference (p = 0.873, Cramér's V = 0.036) (Fig. 2). The distinctive feature of the distribution of causative pathogens in the AP group, compared to other groups, is that it had the highest rate of ‘Only respiratory viruses detected’, the lowest rate of ‘Both respiratory viruses and bacteria detected’ (Fig. 3). Additionally, the AP group had the highest rate of ‘No pathogens identified’. These proportions were not statistically significant (chi-square test, p > 0.05), with Cramér's V values ranging from 0.1 to 0.3. In cases of ‘Both respiratory viruses and bacteria detected’, the combinations predominantly involved rhinovirus and the bacteria most frequently detected in each group. No specific or distinctive patterns were observed (Supplementary data 1).

Fig. 2

Respiratory virus PCR-positive rates in groups. The respiratory viral infection rate in the AP group was 47%, showing no significant difference compared to the non-AP (50%) and AB (53%) groups (Chi-square test, p = 0.873, Cramér's V = 0.036). The chi-square test did not show a statistically significant association between the variables (p = 0.873). Furthermore, the effect size as measured by Cramér's V was 0.036, indicating that there is a negligible relationship between the groups. Based on this, it is unlikely that increasing the sample size would meaningfully alter the results, as the observed association appears to be minimal. Abbreviations: AP, aspiration pneumonia; non-AP, pneumonia except for aspiration pneumonia; AB, acute bronchitis

Fig. 3

Proportion of viral and bacterial detections. The AP group had higher proportions of cases with ‘Only respiratory viruses detected’ and ‘No pathogens identified’ compared to the other groups. Notably, these proportions were not statistically significant (chi-square test, p > 0.05), with Cramér's V values ranging from 0.1 to 0.3. *Patients with sputum culture not performed were excluded. Abbreviations: AP, aspiration pneumonia; non-AP, pneumonia except for aspiration pneumonia; AB, acute bronchitis

Regarding the differences in clinical characteristics between virus-positive and virus-negative cases, in the AP group, virus-negative cases had a significantly higher mean age (Table 2). Other than this, no significant differences in background, symptoms, or clinical data were observed between virus-positive and virus-negative cases in both the AP group and the other groups.

Table 2 Comparison of characteristics between virus-positive and virus-negative groups

Discussion

The high rate of respiratory virus infections in the non-AP and AB groups was anticipated [7,8,9,10,11,12]. However, it is surprising that the AP group also exhibited a similar rate of respiratory virus infections. While the statistical results suggest that a larger sample size might have revealed significant differences in infection rates, the high infection rate observed in the AP group itself remains striking. As demonstrated by the characteristics of the AP group in this study, many patients with AP have underlying conditions that reduce their activities of daily living and spend the majority of their time in care facilities [17]. It is well known that respiratory virus infections prevalent in the community also spread within nursing care facilities [24,25,26]. Therefore, it can be inferred that cases in the AP group were exposed to respiratory viruses at a similar frequency and were infected with similar types of viruses as those in the general population.

The high rate of virus detection in the AP group raises the possibility that respiratory virus infections may play a role in the onset of AP. It is well known that respiratory virus infections can promote bacterial growth in the lower respiratory tract, contributing to the development of CAP [13, 14]. However, no reports indicate that respiratory virus infections facilitate lower respiratory tract infections by the primary pathogens of AP, such as non-fermenting gram-negative bacteria and Enterobacteriaceae. The present observation that the proportion of ‘Both respiratory viruses and bacteria detected’ was lower in the AP group compared to the non-AP group may contribute to this discussion. The lower bacterial detection rate in the AP group compared to other groups is likely due to the inclusion of aspiration pneumonitis cases in bacterial culture-negative cases. This is because clinically differentiating between AP and aspiration pneumonitis is difficult [27]. Considering that many bacterial culture-negative cases in the AP group were positive for respiratory viruses, it can be inferred that respiratory virus infections may be involved in cases of aspiration pneumonitis. These findings show a sequential disease process wherein respiratory virus infections initially lead to an increase in aspiration pneumonitis, some of which subsequently progresses to AP [15]. Another important implication is the possibility that the AP group may include cases of pure viral infections. If point-of-care multiplex PCR testing results had been available to physicians, some cases might have been diagnosed as viral infections rather than AP. Future advancements in point-of-care multiplex PCR testing are expected to shed light on this issue. Although differences in the rates of detected combinations of viruses and bacteria were observed, they were not statistically significant. Further large-scale studies are needed to validate these findings.

The mechanism by which respiratory virus infections lead to aspiration pneumonitis and bacterial AP likely involves the impairment of dysphagia. In elderly patients with a history of cerebrovascular disease, degenerative diseases, or dementia, respiratory virus infections can lead to delirium due to fever, which may promote aspiration. Additionally, pharyngitis or increased respiratory rate may further contribute to aspiration. However, reports examining these considerations are only from pediatric clinical studies [28, 29]. Respiratory virus infections can cause central and peripheral nervous system dysfunction [30], which could potentially affect the nerves involved in swallowing. Despite this, there are few clinical reports on this matter [31]. Further research is needed to elucidate the relationship between respiratory virus infections, dysphagia, and AP.

Most cases of AP require hospitalization. Furthermore, due to underlying conditions, many of these patients also require medical and nursing care [3]. If AP patients have concurrent respiratory virus infections, there is concern that such infections could spread within healthcare settings through close contact with caregivers [32, 33]. For common respiratory viruses, specific containment measures are typically unnecessary, as standard precautions such as masks, gloves, and gowns are sufficient [33]. However, as highlighted by the findings of this study, distinguishing respiratory viral infections in AP patients remains clinically challenging. It is, therefore, crucial for healthcare providers to recognize AP patients as high-risk cases for concurrent respiratory viral infections. From a broader perspective, if respiratory virus infections indeed contribute to the onset of AP, raising public awareness about preventing respiratory viral infections could be a more effective strategy to reduce the incidence of AP itself. Vaccination campaigns are one such approach, but the scope of viral species covered by currently available vaccines remains limited. Advancements in vaccine development are thus eagerly anticipated to expand protection against a wider range of respiratory viruses.

This study has several limitations. First, the classification of AP relied on physician diagnosis, which may lack the strict criteria. This approach could have introduced selection bias and potentially affected the classification of AP. It may have led to the inclusion of cases of respiratory virus infections unrelated to aspiration in the AP group, which could partially explain the higher observed rate of viral infections in this group. To address these potential inaccuracies, we compared PCR results after highlighting differences in background characteristics between the AP and other groups. However, this approach does not fully mitigate all potential issues. Secondly, this study is a retrospective analysis conducted at a single regional hospital, which may limit the generalizability of its findings. The characteristics of aspiration pneumonia patients may vary depending on the country, region, and the type of hospital [34]. Notably, the overall rate of respiratory virus infections observed in this study was higher than previously reported. Additionally, the small sample sizes in each group may have amplified the impact of these biases. To validate these findings, multicenter studies involving hospitals with diverse roles and patient populations are necessary. Third, interpreting pathogen testing results requires caution. For PCR testing, its high sensitivity means that positive results in respiratory specimens do not always indicate active infection, and false positives can occur [35]. Conversely, rare viral species undetectable by the PCR kits used in this study may be present among PCR-negative cases. The low detection rate of causative pathogens in bacterial pneumonia may reflect factors such as poor specimen quality or prior antibiotic use, which could also affect the AP group. Additionally, even when pathogens are detected, contamination or colonization cannot be excluded. The absence of oral streptococci or anaerobic bacteria in this study raises the possibility of underestimating bacterial infection rates in the AP group, although recent evidence suggests these pathogens play a limited role in AP [19]. Finally, the sample collection period coincided with the COVID-19 pandemic. In this study, cases suspected of COVID-19 at the time of their emergency department visit were excluded. Most of these excluded patients had negative COVID-19 test results, suggesting that some of them may have had community-acquired respiratory virus infections [36]. While this may have led to a decrease in the number of study participants, it likely did not significantly impact the primary objective of investigating the respiratory virus detection rate in the AP group.

Conclusion

In AP cases, respiratory virus infections were observed at a high rate comparable to those in non-AP and AB cases. This finding suggests a potential association between respiratory virus infections and the development of AP and highlights the possibility of novel preventive strategies, such as focusing on respiratory virus infection prevention, including common cold prevention. Distinguishing between AP cases with or without respiratory virus infections based on clinical findings was challenging, making it difficult to recommend changes in treatment approaches based on this factor. However, recognizing that physician-diagnosed AP often involves viral infections may instead highlight the importance of enhanced infection control measures in managing AP. Given the limitations of this study, further advanced research is needed to validate these findings.