We demonstrated that severe pneumonia children with ADV infection had more abnormal laboratory findings and more severe clinical outcomes than cases without ADV infection, including suffering more respiratory failure, longer LOS and higher mortality.
In our study, the median age of severe pneumonia children was 10 months; and compared with cases without ADV infection, cases with ADV infection were older. The study in Singapore showed most pediatric patients infected with ADV were below 2 years old (17), and another study showed the median age of children with adenovirus pneumonia in Malaysia was 1.08 years (18), which was similar to ours. It supposed that the newborns who got immunity from their mothers could defense the ADV infection, but as times goes on, the immune defense from mothers was fading, and their own immune system were not yet fully mature, children around age of one were vulnerable of ADV infection. It has been confirmed that ADV mainly infects immunocompromised people, so the children aged around one year may be at higher risk of ADV infection. However, fatal cases had also been reported among newborns (19, 20), therefore, more evidence is needed to support this hypothesis.
25.3% of severe pneumonia children with ADV infection were combined infected by other microorganism, like RSV, influenza virus or parainfluenza viruses. A study in Turkish also showed 12.1% of respiratory infection children were detected with at least two virus infection, and the most common viral agent was HRV followed by ADV (21). Only 2.1% of cases without ADV infection were detected with more than one virus. According to this finding, we supposed that children infected with ADV had a weaker immune system than children infected without ADV.
More children infected without ADV had cardiovascular disease, while there was not statistical significance for other diseases. However, a Taiwan study suggested prematurity and congenital heart diseases do not show statistical significance for ADV pneumonia, but they are associated with disease severity (22). Underlying neurological disease and respiratory disease were more in severe ADV infection and pneumonia. Tsou showed patients with underlying condition, especially neurologic diseases were more likely suffering ADV infections (23). One study about risk factors associated with mortality of pneumonia children reported malnutrition was the most common factor related with fatality (24). The study of Zampoli et al. in South Arica reported that ADV associated pneumonia children reported 34.0% were malnourished (25). But in our study, children with ADV infection had a lower risk of combining others disease than cases without ADV infection.
In our study, severe pneumonia children with ADV infection had more abnormal laboratory results than cases without ADV infection. Higher levels of LDH indicated the more severe injury and reflect the possibility of hepatitis (26). Severe pneumonia cases had a high serum level of LDH, which was consisted with Erez study (27), while the LDH level of cases with ADV infection was twice higher than that of cases without ADV infection. Lai et al. had similar results in both the serum and pleural fluid in severe ADV respiratory infection (28). Wu et al. suggested that a high serum level of LDH and a low lymphocyte count could be used as predictors for the severity of adenovirus respiratory infection in children (29). On the contrary, both severe pneumonia cases with and without ADV infection in our results showed high lymphocyte count, but only cases with ADV infection had low percentage of lymphocyte. It means the lymphocyte might not be an appropriate predictor for severity ADV infection in children.
In addition, our study demonstrated that severe pneumonia children showed bad status with low level of albumin and long coagulation time. The study of Miao et al. showed severe adenovirus pneumonia children with low serum albumin may have poor prognosis (30). Cases with ADV infection had higher level of CRP, which was consists with Chen’s study, revealing elevated CRP levels were common in ADV infection, even without superimposed bacterial infection (31). Specially, cases with ADV infection showed high level of serum glucose. We supposed ADV might damage children’s langerhans β cell, but more evidence needed to support this hypothesis.
Only a few severe pneumonia children received the antiviral therapy, but majority of them received the antibiotic therapy, especially the cases with ADV infection. The benefits of treatment with antiviral therapy for severe ADV pneumonia were still not well-confirmed (32). Due to the limitation of test methods, detecting all of the concomitant bacterial infection was difficult (14); and the impact of bacterial co-infection on disease severity and mortality had been reported in patients with viral infection (33, 34). So, most clinicians use antibiotic drugs based on clinical experience. In some randomized controlled trials and observational studies, rapid recognition of viruses was not associated with reducing the antibiotic use (35, 36). Additionally, we found more cases with ADV infection were treated by corticosteroid like adrenaline as the first aid medicine, oxygen therapy and invasive mechanical ventilation, which revealed children with ADV infection might be more serious. It’s worth noting that a randomized clinical trial found among patients with severe pneumonia, the acute use of corticosteroids can reduce treatment failure compared with placebo. Therefore, they suggested the use of corticosteroids as adjunctive treatment for severe pneumonia patients (37).
64.0% of severe pneumonia children with ADV infection suffered respiratory failure on admission, which was more than double the figure of cases without ADV infection. They also had longer duration at hospital. In our result, the 30-day mortality in hospital among cases with ADV infection was 9.3%, consistent with Wu’s study (29), higher than 2.5% of cases without ADV infection. But it’s lower than another study in 415 hospitalized pediatrics under 6 years of age with ALRI caused by ADV in Argentina from 1988 to 2005 with 15% mortality (9). The high risk in mortality with ADV infection suggested that ADV surveillance programs should be in place to monitor peaks in infection rates (8).
Like most retrospective epidemiological studies, data were often incomplete and analyses may be biased. We were unable to detect all kind common virus to compare ADV infection with other viral infection. Therefore, the finding in our study should be interpreted with caution.