Current Status of Norovirus Infections in Children in Sub-Saharan Africa

Noroviruses are a leading cause of acute sporadic gastroenteritis worldwide. In Sub-Saharan Africa, information regarding norovirus infections in children is scarce. A systematic review of studies performed between 1993 and June 2015 was conducted to establish the genotypic distribution and prevalence of norovirus infections in children (≤17) in Sub-Saharan Africa. Analysis of data from 19 studies involving 8,399 samples from children with symptomatic and nonsymptomatic gastroenteritis revealed prevalence of 12.6% (range 4.6% to 32.4%). The prevalence of norovirus infections was higher in symptomatic children (14.2%) than asymptomatic children (9.2%). Genogroup II (GII) was the most prevalent genogroup accounting for 76.4% of all the reported norovirus infections. The rest of the infections were GI (21.7%) and GI/GII (1.9%). The most common genotypes were GII.4 (65.2%), GI.7 (33.3%), and GI.3 (21.3%). These statistics were calculated from studies carried out in 12 out of 48 Sub-Saharan African countries. Therefore, more studies involving several countries are required to determine fully the epidemiology of noroviruses and their contribution to childhood diarrhoea in Sub-Saharan Africa.


Introduction
Acute gastroenteritis is a common illness which is debilitating and life-threatening in children younger than five years of age [1]. Globally, diarrhoeal diseases claim the lives of 800,000 children under the age of five years every year [2]. Most of the deaths occur in Sub-Saharan Africa and south Asia [3]. Noroviruses account for 12% of all cases of sporadic acute gastroenteritis in children under the age of five years worldwide and approximately 200,000 children die from norovirus associated gastroenteritis in developing countries [4,5]. Noroviruses are small, single-stranded, positive-sense, RNA viruses belonging to the family Caliciviridae and the genus Norovirus [5]. They are divided into six genogroups and over thirty-one genotypes [5]. The epidemiology of these viruses is well described in developed countries. However, the exact epidemiology and contribution of noroviruses to acute gastroenteritis in children in Sub-Saharan Africa remain largely unclear. Therefore, a systematic scientific literature review of studies in Africa was carried out to establish the genotypic distribution and prevalence of symptomatic and nonsymptomatic norovirus infections in children in Sub-Saharan Africa.

Genome Organisation and Genetic Diversity of
Noroviruses. Noroviruses have a linear and polyadenylated genome surrounded by nonenveloped, icosahedral capsid of 27 to 40 nm in diameter [5]. The genome of noroviruses that infect humans is organised into three open reading frames (ORFs). ORF1 is the largest and encodes a polyprotein precursor for the seven nonstructural proteins [5,6]. ORF2 and ORF3 encode the major and minor structural capsid proteins VP1 and VP2, respectively [6]. Murine norovirus genomes comprise an additional ORF4, which encodes virulence factor I [7]. The genus, Norovirus, is divided into six genogroups (GI-VI) based on the amino acid sequence of the major structural protein VP1 [5,8]. Human pathogens are found in GI, GII, and rarely GIV. There is extensive genetic diversity among norovirus strains and new variants emerge almost every 2-3 years [5,6]. GI is divided into 9 genotypes and GII into 22 genotypes [5]. The (human) Norwalk virus is the prototype of the genus and is designated as GI.1. GIII comprises bovine and ovine strains and GV murine noroviruses [5]. Swine and canine strains are classified into GII and GIV or GVI, respectively [5]. In humans, GII.4 noroviruses predominantly cause the vast majority of gastroenteritis outbreaks worldwide [9,10].

Transmission and Clinical Characteristics of Norovirus
Infections. Acute norovirus infection is characterised by nonspecific symptoms such as vomiting, nausea, abdominal cramps, myalgias, and intense watery nonbloody diarrhoea that commonly resolves in 2-3 days [5]. Children and immunocompromised individuals may experience a more prolonged and severe disease course lasting for a few weeks to years [11,12]. Sequelae of prolonged norovirus associated gastroenteritis may include irritable bowel syndrome, necrotising enterocolitis, convulsions, and encephalopathy [13][14][15]. Transmission of noroviruses occurs through the faecal-oral route either by direct contact with infected individuals or via contaminated surfaces, food, water, and aerosolised particles from infected vomitus or stool [5]. Outbreak control and prevention strategies are limited to the use of disinfectants and hand sanitisers.

Search Strategy and Selection Criteria.
A systematic scientific literature review of norovirus studies in Sub-Saharan Africa published in peer reviewed journals in electronic databases such as PubMed, Medline, Scopus, and Google Scholar from 1993 to June 2015 was performed using keywords such as "gastroenteritis", "children", "norovirus", "Africa", and "name of country" alone and in various combinations. To avoid leaving out any studies that were not available in major scientific databases or nonindexed journals, Google search was also used. Master of Science theses whose contents have not been published in any journal were also included in the review. Studies outside Sub-Saharan Africa and those involving adult participants alone were excluded. For studies involving both children and adults, only data from children was extracted and included in this review. All studies describing norovirus infections in symptomatic and asymptomatic children were included. Due to the scarcity of information on norovirus infections in children in Sub-Saharan Africa, studies of all durations (≥1 month) were included in this systematic review.

Data Extraction and Analysis.
When reviewing the studies, frequencies of genogroup I and of genogroup II genotypes were calculated independently. Classification of genotypes into polymerase and capsid genotypes was not considered in the analysis due to lack of information. To calculate the approximate prevalence of norovirus infections in Sub-Saharan Africa, only reverse transcription PCR based studies were included. However, a reverse transcription PCR based study by Wolfaardt et al. that used calicivirus positive samples only to calculate the prevalence of Norwalk virus in South Africa was excluded from the calculation [16]. In studies, where the overall prevalence of norovirus infections was not included, prevalence was calculated using the information extracted from the articles. Seasonality of norovirus infections was accepted and included in this review only when the study duration was ≥12 months.

Results
Thirty-five studies were identified in Africa and only 23 were in Sub-Saharan Africa . Nineteen studies had sufficient data to determine the total prevalence of norovirus infections but only seventeen studies had sufficient data to allow the determination of the prevalence of genogroups I and II [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34]36]. Relative frequencies of norovirus GI and GII genotypes were determined from 11 studies only due to lack of data from the other studies [17, 19, 20, 22-25, 29, 31, 33, 36]. Only three studies had some information on the HIV status of study participants [18,22,33]. The duration of most studies was less than 12 months. Consequently, the seasonality of norovirus infections could not be established in many countries. Four studies reported the presence of norovirus recombinants [23,28,31,36] and one study investigated the role of genetic factors on the susceptibility to norovirus infections [29].

Mixed and Single Virus Infections in Norovirus Positive
Diarrhoea Samples. Four studies, from South Africa, Nigeria, Malawi, and Gabon, clearly reported the occurrence of noroviruses in diarrhoea samples as single or mixed infections [20,26,33,34]. In the study from South Africa, 77% (27/35) of the norovirus infections occurred as single

Prevalence of Norovirus Infections
. Data from this review shows that the prevalence (12.6%) of norovirus infections in children in Sub-Saharan Africa is higher than previously thought. In most of the studies included in this review, the prevalence of norovirus was higher in the cases than in the controls [20,22,27,32,35]. Some studies in Botswana, Burkina Faso, and Cameroon reported higher prevalence in asymptomatic children [17,18,28]. The reasons for this phenomenon seen in these countries are unknown. It can only be speculated that the findings from such studies were a true reflection of the situation or some of the individuals included in these studies did not have truly asymptomatic infections or their parents/guardians had incomplete recall of diarrhoeal symptoms. Persistent shedding of the virus from resolved acute norovirus diarrhoeal infections could also have contributed to these events. On the other hand, the samples sizes for the studies conducted in Botswana (100 children) and Cameroon (54 children) were too small to draw serious conclusions [17,18]. Moreover, the study in Cameroon only involved asymptomatic children and thus it is difficult to really know the extent to which norovirus infections contribute to diarrhoea in Cameroonian children. Generally, most of the studies included in this review had smaller sample sizes. Only three studies had sample sizes above 1000 [19,20,22]. Therefore, norovirus prevalence observed in smaller studies may be enlightening but not conclusive. Longitudinal studies with large sample sizes are required to evaluate the exact norovirus prevalence and its contribution to diarrhoeal disease in children in Sub-Saharan Africa. The calculated 12.6% prevalence of norovirus infections came from studies in 12/48 Sub-Saharan African countries. There was no information about norovirus infections from the rest of the 35 countries. Therefore, this is not a representative prevalence rate. More studies in all the countries are required in order to come up with a representative figure.
Additionally, the duration of many of the studies was less than 12 months, and hence results from such studies are but just an eye-opener of the role of noroviruses in childhood diarrhoea in those countries [18, 19, 21, 25-29, 31, 35]. Consequently, the seasonality of norovirus infections in the country where the studies were carried out could not be established. Knowing the season within which norovirus infections frequently occur may help with proper planning for outbreaks and efficient utilisation of health resources.

How Much Diarrhoea Is Caused by Noroviruses in
Sub-Saharan Africa? Although the prevalence of norovirus infections is approximately 12.6% in Sub-Saharan Africa, the contribution of noroviruses to acute and chronic diarrhoea in this subregion is still obscure.
Detection of norovirus in a stool sample from a child with gastroenteritis does not always mean that this is the  cause; there may be mixed infections with other bacterial and viral agents. As true as this statement may sound, many studies included in this review did not investigate the occurrence of noroviruses as single or mixed virus infections in norovirus positive diarrhoea samples. For studies, such as those conducted in South Africa, Nigeria, and Malawi, that investigated occurrence of noroviruses as single or mixed infections with other viral pathogens, they did not investigate coinfections with protozoal and bacterial agents [20,26,33,34]. Therefore, it is still largely difficult to pinpoint the exact share of diarrhoea burden caused by noroviruses in Sub-Saharan Africa.

Norovirus Gastroenteritis in HIV Positive Children.
Noroviruses have been shown to cause chronic and persistent gastroenteritis in immunocompromised individuals such as HIV/AIDs patients, lasting several weeks to years [11,12]. By the end of 2013, there were approximately 35 million people living with HIV worldwide [40]. Sub-Saharan Africa accounted for 71% (approximately 24.7 million people) of the global HIV infections. With such a huge percentage of people infected with HIV, noroviruses may be silently causing undocumented severe and prolonged gastroenteritis in Sub-Saharan Africa. In this review, only two studies reported the prevalence of norovirus infections in HIV positive children [22,33]. In South Africa, more than 50% (8/15) of HIV positive children had chronic norovirus diarrhoea. The results were in agreement with the findings from across the globe that noroviruses tend to cause persistent diarrhoea in immunocompromised individuals. Although the results from these two studies may not be adequate to draw conclusions, they point to the potential devastating role of HIV in chronic norovirus gastroenteritis. Large scale research in Sub-Saharan Africa is required to correctly understand the impact of HIV/AIDS in norovirus associated gastroenteritis.

Risk Factors and Transmission Dynamics of Noroviruses.
Whilst all the studies included in this systematic review provided evidence on the prevalence of norovirus infections in children in Sub-Saharan Africa, none of them critically looked at the possible risk factors and transmission dynamics of noroviruses. It would be worthwhile to investigate the transmission dynamics and risk factors associated with norovirus infections in Sub-Saharan Africa.

Conclusion
The prevalence of norovirus infections in children with symptomatic and nonsymptomatic gastroenteritis in Sub-Saharan Africa is approximately 12.6%. However, most of the studies from which this figure was derived had small sample sizes and were of a shorter duration. It is therefore prudent that the epidemiology of noroviruses and their contribution to childhood diarrhoea and deaths in Sub-Saharan Africa should be investigated further using large scale longitudinal studies.

Disclaimer
The findings and conclusions in this report are those of the author.

Conflict of Interests
The author declares no competing interests.