Prevalence of Human Bocavirus in Africa and Other Developing Countries between 2005 and 2016: A Potential Emerging Viral Pathogen for Diarrhea

Background Human Bocavirus (HBoV) is an emerging virus discovered in 2005 from individuals suffering gastroenteritis and respiratory tract infections. Numerous studies related to the epidemiology and pathogenesis of HBoV have been conducted worldwide. This review reports on HBoV studies in individuals with acute gastroenteritis, with and without respiratory tract infections in Africa between 2005 and 2016. Material and Method The search engines of PubMed, Google Scholar, and Embase database for published articles of HBoV were used to obtain data between 2005 and 2016. The search words included were as follows: studies performed in Africa or/other developing countries or/worldwide; studies for the detection of HBoV in patients with/without diarrhea and respiratory tract infection; studies using standardized laboratory techniques for detection. Results The search yielded a total of 756 publications with 70 studies meeting the inclusion criteria. Studies included children and individuals of all age groups. HBoV prevalence in Africa was 13% in individuals suffering gastroenteritis with/without respiratory tract infection. Conclusion Reports suggest that HBoV infections are increasingly being recognized worldwide. Therefore, surveillance of individuals suffering from infections in Africa is required to monitor the prevalence of HBoV and help understand the role of HBoV in individuals suffering from gastroenteritis with/without respiratory tract infection.


Introduction
Diarrhea is a leading cause of morbidity and mortality in children worldwide [1,2]. Diarrhea is the third major cause of childhood mortality in children less than 5 years of age especially in Africa and developing countries [3][4][5]. The modes of transmission include ingestion of contaminated food or water (e.g., via flies, inadequate sanitation facilities, sewage and water treatment systems, and cleaning food with contaminated water), direct contact with infected feces (fecaloral route), person-to-person contact, and poor personal hygiene [6,7].
According to WHO [1], approximately 90% of the estimated 2.2 million of deaths caused by diarrheal infections in children less than 5 years of age are related to poor sanitation and hygiene behaviors worldwide. While the mortality due to diarrheal diseases has declined significantly in children over the past twenty years in developed countries [8,9], the incidence of childhood diarrhea in developing countries has not decreased [1,2]. Those who survive these illnesses have repeated infections by enteric pathogens which remains a critical factor leading to serious lifelong health consequences [10] and eventually result in death [11]. Viruses are recognized as major cause of gastroenteritis, particularly in children, and the number of viral agents associated with diarrheal disease in humans has increased progressively. Viruses such as rotavirus, norovirus, astrovirus, and adenovirus that cause diarrhea have been reported worldwide [11].
These studies indicate that only HBoV2, HBoV3, and HBoV4 strains of the virus are mainly involved in gastroenteritis [37][38][39]. Currently, there is limited data on ELISA method for the detection of the virus. HBoV detection has been done by conventional PCR [17,28,40,41] and real-time PCR [42][43][44][45]. While HBoV epidemiological studies have shown evidence for widespread exposure to the virus, the causative role of HBoV in respiratory tract disease and gastroenteritis is still under investigation [46]. Proven evidence is difficult to obtain without an in vitro culture system and animal model [28,[47][48][49].
The prevalence of HBoV has been reported in Europe [50,51], America [17,41], Asia [34,52], Australia [40,53], Africa [54], and the Middle East [18], ranging from 1.5% to 19.3% [17,55]. This review is a summary of reported HBoV studies in individuals with acute gastroenteritis, with and without respiratory tract infections looking specifically to studies in Africa to determine the role of HBoV in diarrheal outbreaks.

Search Strategy.
A literature search of selected studies that investigated HBoV in Africa, in other developing countries and worldwide was performed using the following terms: Human + Bocavirus + Africa + Developing countries + Worldwide on PubMed, Google scholar and Embase. This search yielded 756 publications (Figure 1). The first search was performed for HBoV + Africa, the second search was HBoV + other developing countries, and the third search was HBoV + worldwide. Keywords used included Human Bocavirus, Bocavirus, and Human parvovirus combined for each (Africa; Developing country; Worldwide). To avoid leaving out any studies not found in major scientific databases, Google search was also used. After reviewing each article, studies were selected if they met the following inclusion criteria: (ii) Studies for the detection of HBoV in patients with or without diarrhea and respiratory tract symptoms. Diarrhea defined as the passage of loose or watery stools, at least three times in a 24-h period [56].
(iii) Studies using standardized laboratory techniques for detection of HBoV including PCR, real-time-PCR, and Multiplex PCR (m-PCR).

Data Extraction
Information extracted from the inclusion studies included country where study was done, time period of study, age range of participants, study setting (rural/urban/periurban), sampled population (number of included samples), method used for detection, clinical symptoms, sample type, and HBoV subtype.

Statistical Analysis
All analysis were conducted using R programming environment for data analysis and graphics Version 3.5.0 [57] to calculate random and fixed effects. Function "rma" from the package Metafor [58] was used to calculate heterogeneity between studies and generate a forest plot. Heterogeneity was assessed by Cochran's Q test.
Meta-analysis was done to provide transparent, objective, and replicable summaries of the study findings. From all the 70 studies, 66 had sufficient information to enable statistical analysis. As shown in Figure 2 with the dispersion in study prevalence, there was a low heterogeneity among the studies (Cochran Q = 12.2800 [df = 65] P-Val = 1). Apart from the observed increase in the prevalence of HBoV, none of the other drivers (including age, setting, symptoms, method of detection, and hospitalization) achieved statistical significance. The test for overall effect was Z = 13.29 (P<0.0001) which was highly significant in the findings.
In Africa, the majority of studies (82%; 9/11) were done in urban settings while 18% (2/11) were done in rural settings. Ten (91%) of the studies reported on patients suffering from respiratory tract infections and one study (9%) reported on patients suffering from gastroenteritis. In these studies, a total of 383 (10.4%) samples tested positive for HBoV (Table 3). More studies reported HBoV in children less than five years of age (54%; 6/11) compared to children above the age of 5 and adults 45% (5/11) ( Table 3).
The prevalence of HBoV in Africa was 13% in individuals suffering from gastroenteritis with and without respiratory tract symptoms. The high detection rate of HBoV in Africa was consistent with the global increase of HBoV in children less than 5 years of age [4,59]. Children of all age group are most likely to experience HBoV infection as a result of poor sanitation and hygiene practices [59]. The most predominant HBoV subtype identified in Africa was HBoV1, which was detected in all the studies. Only one study (9%), from Kenya detected all subtypes (HBoV1-4) from children of all age group (Table 3). Not all the studies tested for all HBoV subtypes; this may be due to the fact that other HBoV subtypes have just been recently discovered compared to HBoV1 [37]. The most predominant HBoV subtype identified in other developing countries was HBoV1, which was isolated in all the studies. Only 10% (1/10) of the studies from china detected HBoV2 in the study population ( Table 2).
The results in Africa indicated that HBoV in children less than five years of age was high, 54% (6/11) compared to children above 5 years of age and adults 45% (5/11) ( Table 3). Schildgen and colleagues [60] showed that all age groups can be affected by HBoV, although severe infections requiring hospitalization occur primarily in patients with an underlying disease and children under 5 years of age [61][62][63][64].   Journal of Tropical Medicine 7    Journal of Tropical Medicine 9  Severe clinical cases (such as destruction of the epithelium of the respiratory system) have been described in children [61,[64][65][66] and adults with immunodeficiency [64] and other risk groups [67]. Studies in Africa (13%; 9/70) were mostly done in urban setting compared to other developing countries/worldwide (87%; 61/70) (Tables 1, 2, and 3). This could be due to the lack of laboratory resource capacity and technology for the detection of HBoV in rural settings.
The methods used for detecting HBoV have been conventional PCR [17,28,40,41,45,53,64] and real-time PCR [42-45, 54, 68], due to the limited success of serological and viral culture techniques. Real-time PCR is more sensitive and offers greater sensitivity, increased specificity with the addition of oligoprobes, and the added benefit of a closed detection system, reducing the likelihood of false positive results due to contamination with amplicon [33]. In Africa, 63% (7/11) of studies used conventional PCR for detection, 27% (3/11) used real-time PCR, and 9% (1/11) used Multiplex PCR which is also conventional PCR (Table 3).
In all eleven African studies, HBoV1 was detected (Table 3), similarly in other developing countries HBoV1 was detected in all the studies. HBoV Subtype 1 is mainly associated with respiratory diseases but can also be found in stool samples from patients suffering from diarrhea. Previous studies have reported prevalence of HBoV in symptomatic patients 1.5-16% worldwide [69,70]. Several studies have isolated HBoV from children with respiratory tract infection worldwide, and the prevalence of HBoV in these children was 1.5%-19% [32,33,35].
The reports on HBoV in Africa, other developing countries, and worldwide in individuals suffering from respiratory tract infection 78% (55/70) were higher compared to those suffering from gastroenteritis 21% (15/70) (Tables 1, 2, and  3). This could be due to the fact that most studies focused on HBoV in respiratory tract infection since its discovery in 2005 [28,33,36,71]. However recent studies are increasingly detecting HBoV in individuals suffering from diarrheal diseases due to the presence of the virus in stool samples of individuals suffering from gastroenteritis [37,[72][73][74].
Although the number of studies in Africa is limited, the HBoV prevalence rate of 13% indicates that this virus is one of the emerging viral agents in those suffering from diarrhea with and without respiratory tract infections. Currently there is no available reporting system for HBoV infection in the primary healthcare systems in Africa, suggesting that diarrheal cases with and without respiratory tract infection are likely to be underreported [47]. The high frequency of HBoV in children raises a potential health risk as these children may act as reservoir for other emerging epidemic HBoV strains [27,37,75].

Conclusion
More studies are required in Africa, especially in rural settings to monitor the prevalence of HBoV and help understand the role of HBoV in individuals suffering from gastroenteritis with/without respiratory tract infection. HBoV infections are likely to be underreported in Africa considering the costs of testing for the virus. This review was done to shed light on HBoV and its possible role in diarrheal incidence.

Conflicts of Interest
The authors declare no conflicts of interest.