In total children had complicated

In total, 33 children had complicated pneumonia, which included 24 children 5 years or younger and nine children older than 5 years. Children 5 years or younger had a significantly higher rate of complicated pneumonia (39.3% vs. 13.6%, p < 0.05). Most children with complicated pneumonia presented with pleural effusion (30/33, 90.9%); 21.2% of children had respiratory failure, and 6.1% of children had pneumatoceles. Only 14 (42.4%) of them received macrolide. Compared to children without complications, these children had a significantly lower rate of macrolide treatment (p < 0.05).
Discussion
Community-acquired pneumonia is a common pediatric disease worldwide. The disease severity varies by etiology, age and host immunity. Thus, pediatric CAP shows a aa-dutp in epidemiology and clinical manifestations in different countries. Among the hospitalized children with CAP, 79–85.6% had at least one identified pathogen. The incidence of M. pneumoniae infection ranged 7–37%. Prospective, single-hospital based studies demonstrate that at least one pathogen was detected in 79% of hospitalized children with CAP, which includes 26% of M. pneumoniae, 11% of C. pneumoniae, and 19% of viruses. An analysis study in Singapore identified pathogens in 38.4% of children, which include typical bacteria in 10.3% of infections, M. pneumoniae in 20.3% of infections, and viruses in 5.5% of infections. A local surveillance study can provide more useful information for clinicians. In northern Taiwan, a prospective study demonstrated that S. pneumoniae, M. pneumoniae, and viruses accounted for 42%, 37%, and 41%, respectively, of hospitalized children with CAP. Codetection was common in the current study. The prevalence of CAMP was 26.0% in hospitalized children, which was similar to the findings in previous reports. However, children 5 years or younger accounted for nearly one-half (48%) of children with CAMP, and they had a more complicated clinical course and outcome. Some studies report that the prevalence of children 5 years or younger with CAMP was approximately 46.5–61.5%. Korppi et al also observed that children younger than 4 years had a higher hospitalization rate with CAMP in comparison to children older than 5 years (67% vs. 4%, respectively). However, the studies had some inconsistencies because of the heterogeneity in case definition. Most studies only enrolled hospitalized children, and thus other children having CAMP with mild symptoms may be missed. Another limitation of meta-analysis is the variability of diagnostic tools in each study. Methods that were used to detect M. pneumoniae infection included culture, complement fixation, serologic testing, and molecular-based detection assays. There is still no single “gold standard” for diagnosis. The most convenient diagnostic method are commercial serologic testing kits such as enzyme immunoassay and indirect immunofluorescence. However, the serologic testing depends on the change between two consecutive serum IgG titers, which can only provide evidence of recent infection retrospectively. Single serum IgM positivity can be used to prove acute M. pneumoniae infection, but IgM can only be detected 1–2 weeks after infection. The development of PCR may overcome the limitation of serologic tests and provide sensitive evidence in acute infection. During the community outbreaks of CAMP, Thurman et al found decreased sensitivity of PCR with increased interval between symptom onset and specimen collection; however, the serologic assays have an opposite trend. Many studies have been conducted to compare serologic testing and PCR for the diagnosis of M. pneumoniae infection, and a great discrepancy between these two methods has been observed. A large retrospective study, which investigated more than 10,000 hospitalized children with M. pneumoniae infection, found a higher discrepancy rate in the results between PCR and IgM tests in children younger than 3 years. They suggest that PCR is the preferred method for M. pneumoniae diagnosis in younger children because of their immature immune response to M. pneumoniae infection. Although the advantage of PCR is its high sensitivity, the result of PCR cannot be used to differentiate the asymptomatic carriage of M. pneumoniae in upper respiratory tract from true infection. A combination of M. pneumoniae molecular and serologic methods may be the most suitable way to diagnose M. pneumoniae infection. A single positive result may not be a strong indicator for diagnosis.