Clinical significance of different virus load of human bocavirus in patients with lower respiratory tract infection

To assess the impact of human bocavirus (HBoV) virus load on epidemiologic and clinical characteristics in children with lower respiratory tract infection (LRTI). Clinical records of a total of 654 patients with HBoV infection during January 2013 and December 2014 were retrospectively reviewed. Patients with high HBoV virus load infection had a similar age distribution with the total HBoV infection, which had a peak age group of 6–24 months. Patients with high virus load are significantly younger (P < 0.01) than those with low load. The patients who had wheeze and tachypnea/dyspnea at presentation were more strongly affiliated with the patients with high virus load (both P < 0.01). Co-infection was found significantly more frequently among patients with low virus load than those with high virus load (57.0% vs 38.9%; P < 0.01). High virus load was a significant predictor of severe LRTI (P < 0.05). HBoV infections are found in an important proportion of the hospitalized children with respiratory illnesses (8.85% in our series). A high HBoV virus load could be an etiologic agent for LRTI, which may lead to more severe lower respiratory tract symptom and severe disease.

There were no significant differences between high and low virus load in different age groups less than 36 months (p > 0.05, Fig. 2B). However, the virus load decreased with their age, with a statistical significance for age distribution (p < 0.001) in patients older than 36 months (Fig. 2B). Fig. 3. It occurred throughout the year with a higher proportion from July to December and it peaked in October or November. The frequency of patients with high virus load also increased from July to December and peaked   in October and November. Interestingly, the frequency of patients with high virus load was higher from July to December, but lower from January to June.

Comparison of clinical characteristics and laboratory values by PCR status. Clinical character-
istics and laboratory values of children infected by HBoV with high virus load and low virus load are shown in Table 1. To exclude the interaction between parameters, mean age, fever, wheezing and tachypnea/dyspnea at presentation were tested by logistic regression analysis. On clinical manifestations, patients with high virus load were significantly younger (P < 0.01) than those with low load. The frequency of fever was significantly lower in patients with high virus load compared with low load (P < 0.01). The patients who had wheeze and tachypnea/ dyspnea at presentation were more strongly affiliated with the patients with high virus load (both P < 0.01). Co-infection was found significantly more frequently among patients with low virus load than those with high virus load (57.0% vs 38.9%; P < 0.01). Co-infection with RSV was more strongly affiliated with the patients with low virus load (both P < 0.01, Fig. 4).

Comparison of co-infection by
Risk Factors for disease severity. Fifty-one (7.8%) patients received O 2 (38 with high virus load and 13 with low virus load). None of the patients transferred to ICU. Requirement for O 2 was used to evaluate disease severity. In unadjusted analysis, requirement for O 2 was associated with young age, HBoV single infection and high virus load (all P < 0.05). Other variables (sex, white blood cells, percentage of neutrophils, platelets and C-reaction protein [CRP]) showed no difference.
The multivariable logistic regression model for requirement for O 2 is shown in Table 2. Controlling for 4 demographic and clinical characteristics, significant predictors of requirement for O 2 were age < 2 months and high virus load (both P < 0.05).

Discussion
As far as we know, this is by far the largest study focus on the clinical significance of different virus load of HBoV in patients with LRTI. We included a series of patients in the Suzhou region over two consecutive years. Also, this is the first study comprehensively describe the impact of HBoV virus load on epidemiology, clinical features, laboratory values and microbiological evaluation in children with HBoV LRTI.
During the last decade since its discovery, HBoV is found to be one of the most common pathogens that leads children to acute respiratory tract infection, especially LRTI 4,9 . Previous studies have shown that rates of detection of HBoV in ARI varied from 1.5% to 19.0%. In the present study, the overall detection rate of HBoV in LRTI patients was 8.85% which was similar to other regions of the world.
Age profile noted for individuals with HBoV was similar to RSV infections, with infections almost completely confine to infants and young children (< 24 months). Our study are in line with most of the previous studies 1,2,10 . However, our study disagree with the Canadian study, in which there was much less difference in the prevalence of HBoV according to age 11 . Interestingly, our study also demonstrated that patients with high virus load had a similar age distribution with the total HBoV infection, which had a peak age group of 6-24 months. This interesting phenomenon highlighted the demographic importance of high virus load in HBoV LRTI. Seasonal peaks of HBoV infection vary among different counties because of climate and geographic factors. Previous studies suggested that HBoV infection had a higher detection rate in winter 12,13 or in summer 2,5 . In our study, a higher frequency of HBoV was observed between July and December with a peak in November. Interestingly, the frequency of patients with high virus load was also higher from July to December, and its seasonal distribution correlated with the total distribution, which highlighted the importance of high HBoV virus load in the seasonal distribution of HBoV virus.
In our study, wheezing and tachypnea occurred more frequently in children with high virus load. Actually, wheezing and tachypnea have been recognized as important clinical manifestations in HBoV infection 2-4,14-16 , while the clinical significance of different virus load in manifestations has been less studied. Allander etal found that high virus load of HBoV were noted mainly in the absence of other viral agents, suggesting a causative role for acute wheezing 4 . Deng etal found that wheezing was one of the most common symptoms presented by patients with positive HBoV, and the days of wheezing correlated with virus load. Different from the previous study, we use logistic regression analysis to exclude the interaction between clinical parameters and supported the fact that wheezing and tachypnea were most common and important symptoms presented in high HBoV load LRTI.
Co-infection was found significantly more frequently among patients with low virus load than those with high virus load (57.0% vs 38.9%; P < 0.01), indicating that patients with low HBoV virus load were more likely to be co-infected with other pathogens, which is consistent with the studies of Brieu 17 and Kaida 18 . Although HBoV has been regarded as an infectious agent present, its pathogenic role in respiratory disease is still debatable. This virus is frequently detected in co-infection with other respiratory viruses of well-established pathogenic role 8,10 . In our study, high virus load were noted mainly in the absence of other respiratory viruses and suggesting a causative role for HBoV. However, low virus load were more detected in virus co-infection, and co-infection with RSV was more strongly affiliated with the patients with low virus load, suggesting that a low virus load in the nasopharynx   seemed to be associated with long term shedding of HBoV DNA, unrelated to current illness. Actually, a recent study has suggested that asymptomatic viral infection in infants is associated with low viral load 19 . Indeed, HBoV has been detected in infants and children with and without respiratory symptoms [19][20][21][22] . Byington reported that approximately half of the episodes of HBoV were asymptomatic. Furthermore, they found 14% of 151 episodes remained positive for 3 or more weeks 21 . Chonmaitree reported prolonged detection of bocavirus of up to 28 days 19 . HBoV DNA was also found in nasopharyngeal aspirates from 43% of asymptomatic children undergoing elective surgery 23 , and in lymphocytes from palatine tonsils of 32% of children undergoing tonsillectomy 24 , suggesting that HBoV could establish latent or persistent infection in respiratory tract. Actually, in our study, 18 patients readmitted within 30 days (median duration: 17 days) after discharge, while 14 (77.8%) of the readmitted patients had HBoV positive duration the second hospitalization (data not shown). This interesting phenomenon may also suggest that HBoV could establish persistent infection in respiratory tract.
Previous studies 6,25 showed that the high HBoV virus load played an important role in the severity of LRTIs. Deng etal also found that high HBoV virus load led to more severe lower respiratory tract symptoms and longer hospitalization 16 . In our study, we found that age < 2 months and high virus load were associated with severe LRTI (requirement for O 2 ), while virus co-infection would not increase disease severity. Therefore the more severe lower respiratory tract symptom presented in high HBoV virus load patients may solely depend on HBoV virus load. Different from the previous study, our study use multivariable logistic regression which highlight the role of high virus load in severe LRTI.
The present study has potential limitations. First, quantitative PCR combined with serology would give a better idea of whether the viral infection was active or incipient 26 , we used only endpoint PCR. We found a high proportion of HBoV infections that had coinfection with other viruses. However, we may have missed additional viral infections not detected by PCR or tissue culture. As diagnostic testing for bacterial pathogens was not performed, we do not know whether HBoV-bacterial coinfections occurred among children. Because all our study subjects were hospitalized patients with LRTIs, the results are not necessarily generalizable to outpatient clinics.
In conclusion, HBoV infections are found in an important proportion of the hospitalized children with respiratory illnesses (8.85% in our series). A high HBoV virus load could be an etiologic agent for LRTI, which may lead to more severe lower respiratory tract symptom and severe disease.

Methods
Study Patients. All experiments were performed following the relevant guidelines and regulations of Soochow University. The methods were carried out in accordance with the approved guidelines. The study was approved by the Medical Ethics Committee of Soochow University. The parents of all study participants gave both verbal and written informed consent before study enrollment. A total of 7393 patients with a clinical and radiological diagnosis of LRTI in Children's Hospital of Soochow University during January 2013 and December 2014 were enrolled in the study.

Respiratory tract aspirates preparation and nucleic acid extraction. Nasopharyngeal aspirates
were obtained from all patients within 24 hours of admission. This involved passing a suction catheter through the nose with the intent of passing it into the lower part of the pharynx. The depth of penetration for the nasopharyngeal aspirate catheter was set at 7-9 cm. A total of 2 ml nasopharyngeal aspirates was obtained and centrifuged at 500 × g for 10 minutes and resuspended in 2 ml saline and divided into 2 aliquots for pathogen detection using direct immunofluorescence assay (DFA) and PCR. One of the equally divided samples of nasopharyngeal aspirate was centrifuged at 12000 × g for 5 minutes, followed by extraction of DNA from a 400-ul KL sample using DNA-EZ Reagents (Sangon Biotech, Shanghai, China) or TRIzol Reagent (Life Technologies, Carlsbad, USA) in accordance with the manufacturer's instructions. A final 200 KL of DNA was eluted and DNA sample was divided into 2 aliquots for HBoV and Mycoplasma pneumoniae (MP) gene amplification via PCR. Viral isolation. Direct immunofluorescence assay was done directly on nasal aspirate specimens by use of murine monoclonal antibodies (Chemicon). The specimens were tested for respiratory syncytial virus (RSV), adenovirus (ADV), parainfluenza virus (PIV) types 1-3, and influenza virus (IV) types A and B. All staining procedures were performed according to the manufacturer's instructions. Immunostained preparations were viewed with a fluorescence microscope (Leica 020-518.500, Germany). Sputum DNA was extracted as described above, and hBoV-DNA was detected by real-time fluorescent PCR.
Mycoplasma pneumoniae (MP) detection. Specimens were tested for the presence of MP by real-time PCR, acute IgM and IgG serology were also evaluated using enzyme-linked immunosorbent assay (ELISA). MP infection was confirmed if a positive PCR and an elevated IgM at admission or a fourfold increase in IgG at follow-up was detected.
Data collection. Demographic, clinical features and laboratory tests including white blood cells (WBC), neutrophils, platelet, C-reaction protein (CRP) and nasopharyngeal aspirates tests were routinely performed of each patient. Statistical analysis. Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS; version 17.0). Data were expressed as number with percentage, mean and standard deviation (SD) or median as appropriate. Normally distributed continuous variables were compared using the Student t test and non-normally distributed variables were analyzed using Mann-Whitney U test. Categorical data were analyzed using the chi-squared (χ 2 ) test or Fisher's exact test. P value < 0.05 was considered statistically significant. Multivariable logistic regression analyses were conducted to identify different clinical characteristics associated with different virus load, and to evaluate independent predictors of requirement for O 2 . Factors were considered