Human bocavirus and human metapneumovirus in hospitalized children with lower respiratory tract illness in Changsha, China

Background Lower respiratory tract illness is a major cause of morbidity and mortality in children worldwide, however, information about the epidemiological and clinical characteristics of LRTIs caused by HMPV and HBoV in China is limited. Objectives Human bocavirus (HBoV) and human metapneumovirus (HMPV) are two important viruses for children with lower respiratory tract infections (LRTI). We aimed to assay the correlation between viral load and clinical characteristics of HBoV and HMPV with LRTI in Changsha, China. Methods Nasopharyngeal aspirates (NPAs) from children with LRTI were collected. Real‐time PCR was used to screen HBoV and HMPV. Analyses were performed using SPSS 16.0 software. Results Pneumonia was the most frequent diagnosis. There was no significant difference between HBoV‐ and HMPV‐positive patients in age (P = .506) or hospitalization duration (P = .280); 24.1% and 18.2% were positive for HBoV and HMPV. HBoV infections peaked in summer (32.2%), and HMPV infections peaked in winter (28.9%). The HBoV‐positive patients had a shorter hospitalization duration than the HBoV‐negative patients (P = .021), and the HMPV‐positive patients had a higher prevalence of fever than the HMPV‐negative patients (P = .002). The HBoV viral load was significantly higher among patients aged <1 year (P = .006). The mean HBoV and HMPV viral loads were not significantly different between patients with single infections and coinfections. Patients infected with HBoV only were older than those coinfected with HBoV and other respiratory viruses (P = .005). No significant difference was found in the clinical characteristics of patients infected with HMPV only and those coinfected with HMPV and other respiratory viruses. Conclusion Pneumonia was the most frequent diagnosis caused by HBoV and HMPV. Neither HBoV nor HMPV viral load was correlated with disease severity.


| INTRODUCTION
HMPV, which was originally isolated in the Netherlands 1 and is closely related to avian metapneumovirus, was the first human disease-causing pathogen identified in the genus metapneumovirus. 2 HMPV causes various clinical symptoms, such as cough, wheezing, and fever. 2 Young children and the elderly are more susceptible to HMPV infection. 2,3 Indeed, HMPV mainly infects children under 5 years of age and causes upper respiratory tract and severe lower respiratory tract infections. [4][5][6][7][8] HBoV was discovered in Sweden and classified in the family Parvoviridae, 9 and causes various clinical symptoms, including cough, wheezing, and fever. 2 HBoV infections are largely confined to infants and young children (<24 months). Globally, the average prevalence of HBoV in respiratory tract samples ranges from 1.0 (CI: 0.0-2.0) to 56.8% (CI: 46.9-66.8). 10 However, information about the epidemiological and clinical char-      (Table 1).

| Real-time PCR for HBoV and HMPV
To standardize quantification of HBoV and HMPV, known numbers of DNA or RNA transcripts containing the primer targets were used in 10-fold serial dilutions (10 0 to 10 7 copies/μL). Any amplification detected before 40 cycles was considered positive. Quantification of >10 0 copies/μL (10 3 copies/mL) was considered a positive result. An internal positive control gene (GAPDH), positive control, and negative (water) control were included in all reactions.

| Statistical analysis
Viral loads were expressed as initial copy numbers per real-time PCR reaction, and quantitative results were transformed as the log number of viral copies/μL. Continuous variables are expressed as means ± standard deviation (SD) and were compared by independent samples t test. Categorical variables are expressed as frequencies or percentages. Comparisons were performed by chi-squared or Fisher's exact test. Analyses were performed using SPSS 16.0 software. Two-sided P-values <.05 were considered to indicate statistical significance.

| Patients' characteristics
In total, we analyzed 1092 samples collected from LRTI patients

| Clinical characteristics of HBoV-and HMPVpositive patients
Patients were divided into two groups: those with and without HBoV or HMPV infections (Table 2) (Table 5).

| Viral load and disease severity
According to the British Thoracic Society Guidelines for the Management of Community-Acquired Pneumonia in Children 11 and the clinical diagnosis, LRTIs were classified as mild to moderate or severe. The mean HBoV and HMPV viral loads did not differ significantly between the two groups (P = .053 and P = .231) (Figure 3).

| DISCUSSION
In this study, we used real-time PCR to explore the etiology and ex- with LRTIs. This is higher than previous study (62.34%) in Changsha that used traditional PCR, 18 and lower than that a study in Lanzhou that used real-time PCR. 6 On the one hand, the high detection rate in our study could be caused by the greater sensitivity of real-time PCR compared with traditional PCR. On the other hand, it is possible that the detection rate differs geographically. RSV was the most frequently virus detected in patients with respiratory infections, followed by PIV3, HBoV, ADV, and HMPV. Our data support previous reports of RSV, PIV3, HBoV, ADV, and HMPV as the major agents associated with LRTIs among children in a hospital setting. 12,[19][20][21][22] The HBoV infection rate (24.1%) in the present study is consistent with earlier reports (1.9%-24.6% 8,9,16,[23][24][25][26], and the incidence of HMPV (18.2%) in patients with respiratory tract infections is similar to that in other regions (1.5%-18% 2,9,17,23,[26][27][28]. Therefore, HMPV and HBoV are major causes of LRTIs worldwide.
The seasonal peaks of HBoV and HMPV infections vary among countries because of differences in climatic and geographic factors.
In this study, HBoV activity peaked in summer, in agreement with the report by Jiang. 29 In contrast, detection of HBoV in Lanzhou peaked in December and April, 30 possibly due to the dry, cold weather in Lanzhou and the warm, humid weather in Changsha. HMPV detection peaked in winter, which is in line with previous studies. 2,7,18,31 In contrast, in Hong Kong, HMPV detection peaks in spring/summer. 32 The seasonality of HBoV differed geographically, possibly due to climatic factors.
The most frequent symptoms of HBoV-and HMPV-positive patients were cough and fever, in accordance with previous reports. 6,18,33 However, Deng reported that in Chongqing, wheezing was the most frequent symptom exhibited by HBoV-positive patients with severe LRTIs. 16 There was no difference between the HBoV-and HMPV-positive patients in the incidence of fever, tachypnea, cyanosis, O 2 therapy, or WBC count. The HBoV-positive patients had a shorter hospitalization duration than HBoV-negative patients (P = .021). In contrast, Deng reported that HBoV-positive patients had a longer hospitalization duration. The longer hospitalization duration of HBoV-negative patients in our study may have been caused by the presence of other viruses (such as RSV and PIV3) in many of them. Also, hospitalization duration was significantly associated with age (≤6 months), maternal smoking during pregnancy, and a family history of asthma. 19 A high HMPV viral load contributes to development of fever. 4,5 Indeed, HMPV-positive patients had a higher incidence of fever than HMPV-negative patients (P = .002) in our study.
We assessed the relationship between viral load and clinical features (age, gender, respiratory rate, temperature, cyanosis, hospitalization duration, and WBC count). The only significant association of HBoV-or HPMV-positive patients was a higher viral load in <1-year-old HBoV-positive patients. This is in agreement with Jiang's report that patients with a high viral load were significantly younger. 29 In contrast, the duration of wheezing and hospitalization was longer in children with a high than a low HBoV viral load in the study by Deng,16 possibly due to inclusion of only patients with severe LRTIs.
HMPV and HBoV coinfections with other viruses are common in This study has the following strengths: a considerable duration, large number of patients, and comparison of the most common viruses. Its limitation includes the lack of a control group without clinical evidence of illness. Further studies should compare a symptomatic group with a control group and a symptomatic period with an asymptomatic period.
In summary, we present a prospective study of LRTIs caused by HBoV and HMPV. A further comprehensive and in-depth study of the role of HBoV and HMPV in LRTIs in China is warranted.