Factors associated with progression to severe condition and treatment prolongation in respiratory syncytial virus infection: a retrospective study in Japan

Overview Background: In recent years, the epidemic of respiratory syncytial virus (RSV) has been seen in the summer. In daily clinical practice, we noted that RSV infection tends to become more severe and its treatment tends to be longer, particularly in patients hospitalized in the summer. Thus, we investigated factors associated with the progression to severe condition and the prolongation of treatment in RSV infection. Methods: Subjects were pediatric patients diagnosed with RSV infection and hospitalized for treatment during the 5 year s from April 2014 to March 2019. Information on age, sex, season of hospitalization, gestational age, and bronchial asthma of each subject was collected from inpatient medical records. Duration of oxygen administration was used as an index for treatment duration and the use of devices for severe cases as an index for severity. Multiple linear regression analysis and logistic regression analysis were performed to examine the variables associated with duration of oxygen administration and use of a device for severe cases. Results: Data from 298 patients were analyzed. Duration of oxygen administration was significantly associated with bronchial asthma (partial regression coefficient: 0.804, p = 0.010). Hospitalization in summer was significantly associated with use of a device for severe condition (adjusted odds ratio: 5.89, 95% confidence interval: 1.72-20.18). Conclusion: The present study showed that bronchial asthma is a risk factor for prolongation of treatment and infection in summer is a risk factor for progression to severe condition in cases of RSV infection. These findings suggested that children with bronchial asthma and infection in summer need to be treated carefully regardless of their age, sex, and gestational age.

hospitalization for treatment.

Methods
The present study subjects were children who were diagnosed with RSV infection and hospitalized for treatment during the 5 years from April 2015 to March 2019 at Showa University Koto Toyosu Hospital (Tokyo, opened at the end of March 2015, a secondary medical care facility) and they were retrospectively examined. RSV infection was diagnosed using a rapid antigen kit, in principle.
Summer was defined as April through September and winter as October through March of the following year, by diving a year into halves using 14.5˚C as the threshold temperature, referring to the mean temperature by month in the past 5 years in Tokyo [9].
Based on the inpatient medical record, sex, gestational age in weeks, age in months, duration of oxygen administration, presence or absence of bronchial asthma, continuous inhalation of isoproterenol (ISO), and use or no use of high flow nasal cannula (HFNC) were surveyed. Devices for severe cases were defined as continuous inhalation of ISO and use of HFNC. Gestational age in weeks was categorized into the following three groups according to the cutoff ages for administration of palivizumab: <29 weeks, 29 to <35 weeks, and ≥35 weeks. Duration of oxygen administration was used as an index for treatment duration and use or no use of devices for severe cases as an index for severity.
At our hospital, oxygen administration is started when oxygen saturation as measured using pulse oximetry (SpO2) is decreased to 93% or lower and the physicians in charge determine the discontinuation timing, considering changes in SpO2, auscultation findings, systemic condition, and other factors. Presence of bronchial asthma was defined as ongoing treatment in Steps 1 to 4 based on the diagnosis by physicians in charge. Devices for severe cases were introduced when the modified pulmonary index (mPI) [10] was 10 points or higher and the physicians in charge determined that they were necessary. Selection of a device was left to the physicians in charge depending on the circumstances.
In our study, multiple linear regression analysis was conducted to investigate the variables associated with duration of oxygen administration. Logistic regression analysis was performed to evaluate the variables related to use of a device for severe cases. In these models, gestational age, age of the time of hospitalization (months), sex, hospitalization season (summer or winter), and bronchial asthma were included as explanatory variables. P < 0.05 was considered statistically significant. JMP® ver 14.0 (SAS) was used for statistical analysis. Ages of our study subjects ranged from 0 months to 4 years and 6 months. One hundred and four subjects (34.9%) were younger than 6 months old, 82 (27.5%) were aged 6 months or older and younger than 1 year old, 91 (30.5%) were 1 year old, and 21 (7.0%) were 2 years or older. Twenty-three subjects (7.7%) had bronchial asthma and 27 (9.1%) were treated using devices for severe cases. Palivizumab was indicated for a total of 6 subjects at the time of admission and 3 of them were actually treated with it. Subject characteristics are shown by hospitalized season (summer or winter) in Table 1.
Number of hospitalized patients in summer was greater than in winter. Subjects hospitalized in summer tended to receive the administration of oxygen for a greater number of days and to be treated by the devices for severe cases than those hospitalized in winter.
The factors related to the duration of oxygen administration in patients with RSV infection were examined ( Table 2). The presence of bronchial asthma was significantly associated with the duration of oxygen administration (partial regression coefficient; 0.804, p = 0.010). Age at the time of hospitalization, sex, hospitalization season and gestational age were not significantly associated with the duration of oxygen administration.
Variables associated with use of device for severe conditions were evaluated (Table 3).
Hospitalization season (summer) significantly increased the OR for use of device for severe conditions (OR: 5.98, 95%CI: 1.76-20.36). These results remained, even after adjustment for age at the time of hospitalization, sex, gestational age and bronchial asthma.

Discussion
In this study, we specified, in a dualistic manner, duration of oxygen administration (prolongation of treatment duration) and use or no use of a device for severe cases (severity) as indices for the difficulty of treatment of RSV infection. The results suggested that bronchial asthma is a risk factor for prolongation of duration of oxygen administration (prolongation of treatment duration) and morbidity in summer is a risk factor for progression to a severe condition.
Being a premature infant, a history of perinatal respiratory disorder, malnutrition, congenital heart disease, and chronic neurological disorder are known risk factors for progression of RSV infection to a severe condition [11]. American Academy of Pediatrics proposed that bronchiolitis in general involves the risk for progression to a severe condition if it occurs within 12 weeks after birth [12]. For RSV in particular, young age in months has been reported to be a predictive factor for progression to a severe condition The present study showed that, among pediatric patients with RSV infection, the length of the period during which they required oxygen administration significantly differed between those with bronchial asthma and those without. Considering this finding and the past report suggesting that RSV infection is a precipitating factor for bronchial asthma, we can conclude that the duration of oxygen administration in cases of RSV infection is prolonged when there is underlying bronchial asthma.
The present study also showed that a significantly greater number of patients who were infected with RSV in summer were treated by continuous inhalation of ISO or use of HFNC.
Use of ISO or HFNC corresponds to 10 or higher mPI score and means progression to a severe condition. Patients with bronchial asthma are commonly treated by continuous inhalation of ISO when their condition becomes severe. However, among 23 subjects treated by continuous inhalation of ISO, only 3 had bronchial asthma in our study.
Therefore, we speculated that the progression to a severe condition in subjects who had RSV infection in summer does not reflect the worsening of bronchial asthma, but simply suggests the worsening of lower respiratory tract lesions. Although the subjects in this study included many infants younger than 12 months old, age in months at the time of hospitalization was not a factor influencing the progression to a severe condition.
Among the subjects in our study, palivizumab was indicated for a total of 6 subjects at the time of admission and 3 of them were actually treated with it. Devices for severe cases were not used in these 3 subjects treated with palivizumab. The other 3 subjects who did not receive the administration of palivizumab were admitted in September and were infected with RSV before the scheduled day of the administration of palivizumab. Use of a device for severe cases was confirmed in a subject who was not treated with palivizumab.
Progression to a severe condition might have been prevented by earlier administration of palivizumab in this subject.
In Japan where the climate is temperate, epidemic of RSV used to be commonly believed to occur in winter, mainly in December. In recent years, however, the epidemic period has changed. In Okinawa, where the climate is subtropical and greatly differs from the climate in other prefectures, the epidemic period has been considered to be from February to September [16]. A similar tendency has recently been observed in the Kanto region. A report indicated that RSV was more active at 24˚C or higher temperature and 6˚C or lower temperature and most stable when the humidity was 40% [17]. Another report showed that RSV was more active when the relative humidity was 45% to 65% [18]. Still another report demonstrated that RSV preferred a narrow range of humidity and suggested that its detection rate is decreased by 3.8% when the humidity increases by 1% [19]. From 2014 to 2019, the mean temperature in June to September matched the range of temperature at which RSV is considered to become more active. In contrast, relative humidity was higher than the humidity at which RSV is considered to be more active. This suggests that RSV becomes more active at higher relative humidity. It is necessary to continue to observe the epidemics and manifestations carefully, paying attention to climate conditions.

Limitations
Because this study was a retrospective evaluation and the medical history of the patients was obtained from their medical record, patients with bronchial asthma might not have been completely categorized into the group with this disease. Duration of oxygen administration, continuous inhalation of ISO, and indication of HFNC were not evaluated strictly based on uniform indices such as numbers and scores and therefore might have been influenced by subjective views of primary physician and physicians in charge.
Determination of past bronchial asthma cannot necessarily be considered to be in accordance with certain standard diagnostic criteria, either, because presence of bronchial asthma was defined as ongoing treatment from Step 1 to Step 4 based on the diagnosis by physicians in charge. The possibility that these evaluation methods affected the risk factors for prolongation of duration of oxygen administration and progression to a severe condition cannot be denied. Moreover, genetic search for viruses was not performed and it was not determined whether or not any mixed infection with other viruses occurred because of retrospective method of this study. In the future, it is desirable to collect samples and perform these analyses for patients with severe conditions.

Conclusion
Bronchial asthma was a risk factor for prolongation of treatment duration and infection in summer was a risk factor for progression to a severe condition in RSV infection. The present study suggested that when children with bronchial asthma are infected with RSV, treatment should be provided with consideration of possible long-term administration of oxygen. Moreover, this study indicated that infection in summer needs to be treated carefully. In response to the changes in the epidemic period of RSV in recent years, timing of the administration of palivizumab has been re-examined. However, expansion of the range of age in months for the administration of palivizumab is desired for pediatric patients with bronchial asthma. Tables.pdf