Giardia lamblia infections in children in Ghana

Introduction Though giardiasis is an important public health problem in Ghana, several aspects of its epidemiology, particularly the molecular epidemiology has not been investigated adequately. This could be a major hindrance to effective surveillance and control of giardiasis in the country. The study was carried out to determine the prevalence, risk factors and genotypes of Giardia lamblia infecting children at a paediatric hospital in Ghana. Methods A total of 485 patients including 365 diarrhoea and 120 non-diarrhoea children were enrolled into the study. Stool samples were collected and analysed for parasite presence using microscopy, ELISA and PCR. Positive samples were subsequently characterized into assemblages by PCR-RFLP, and further confirmed with sequencing of the glutamate dehydrogenase (gdh) gene. Epidemiological data on demographic, clinical and behavioral features of the study subjects were also collected. Results Prevalence of G. lamblia infections in diarrhoea and non-diarrhoea children were 5.8% and 5% respectively (P>0.5). Sequence data confirmed Giardia lamblia assemblage B as the predominant genotype in both diarrhoea and non-diarrhoea cases. There was no significant association of G. lamblia infection with any of the epidemiological variables investigated. Conclusion Our findings suggest that assemblage B could be the predominant genotype causing giardiasis in children. Increased public health education focusing on good sanitary practices, particularly among mothers and children, could decrease the risk of G. lamblia infection.


Introduction
Giardia lamblia is an enteric parasite that causes giardiasis, one of the frequent causes of diarrhoea, in humans, pets and livestock [1].
The parasite is distributed globally, and children are more at risk of infection than adults [2]. Several studies have associated these infections with socio-demographic, hygiene, nutritional and immune status of the host and strain of the parasite [1][2][3]. Giardiasis infections have also been associated with drinking of contaminated tap water, fresh water and the movement of individuals from a nonendemic region to an endemic region [4,5]. Antenatal studies in Auckland, New Zealand have shown that changing napkins of children was associated with a risk of giardia infection [6]. In developing countries, common risk factors associated with G. lamblia infection are socio-demography: improper sanitation, bad personal hygiene, eating of unwashed fruits and vegetables, and drinking of contaminated tap water [2,7]. Mukherjee et al., [8] showed that in Kolkata, India, there was a relationship between giardiasis and the socio-economic background of the study population. Majority of the diarrhoea patients were of lower socioeconomic status and lived in slums, suggesting that infection could be water and/or food borne. Further, children ≤ 5 years of age were most at risk of parasitic infections. A cohort study in rural Egypt suggested that more females were infected than male infants [9]. In Ghana, there have been reported prevalence rates of 9% of Giardiasis among diarrhoea children in the northern sector [10] and 9.7% in the middle belt [11]. A recent hospital-based study conducted in the capital, Accra, recorded a prevalence of 10.1% [12]. The disease seems to be more prevalent among children and thus, the importance of determining possible risk factors among this cohort cannot be overemphasized. Fast expansion of new settlements without adequate potable water supply, especially in urban communities, leave large populations to consume untreated water. This has consequently increased the risk of water-borne infections, including giardiasis in the country. Although no major outbreak of the disease has yet been reported in Ghana, the increasingly high incidence, 24-32%, reported in some hospitals and polyclinics [13] as well as 5.1-46.5%, in some community day-care centers [14,15] may suggest wide environmental contamination.
Giardia lamblia (syn: G. intestinalis or G. duodenalis) has 6 distinct genotypes; assemblages A and B are found in humans and other mammals, C/D in dogs, E in livestock, F in cats, and G in rats [16,17]. Genotyping of Giardia is seminal to determine sources of infection during outbreaks and also could help elucidate possible transmission routes. Restriction Fragment Length Polymorphism (RFLP) has been used as a useful typing tool in epidemiological studies to investigate possible association of identified genotypes of G. lamblia with severity of diarrhoea and other clinical symptoms.
Giardia assemblage A has been associated with asymptomatic diarrhoea [18] and B as the predominant genotype in diarrhoea cases [19][20][21]. There is poor epidemiological data on the genotypes of G. lamblia in Ghana, probably because studies investigating diarrhoea in the country have focused on enteric bacteria and other protozoans [22]. Though Cryptosporidium spp. is an important cause of diarrhoea in Ghana [22], several studies have reported high prevalence of Giardiasis in the country [11,12]. Using information from these studies, we reasoned that certain risk factors may predispose children, particularly those below 5 years to giardiasis, as reported in other countries. Thus, in the present study, we attempted to define certain risk factors associated with giardiasis in children < 5 years using data gather from a structured questionnaire. This border on education and socio-cultural practices of mothers, food and water sources, breast-feeding practices of mothers, pet ownership, age and sex of child. Using Restriction Fragment Length Polymorphism (RLFP), we genotyped both PCR and ELISA positive samples and confirmed the genotypes with sequencing. We present data, discussing risk factors of G. lamblia infection and suggest assemblage B as the predominant genotype causing diarrhoea in Accra, Ghana.

Study site and subjects
This was a hospital-based prospective cross-sectional study The study population were children ≤ 5 years, who had been hospitalised at PML due to acute diarrhoea and dehydration.
Diarrhoea was defined as passage of loose or watery stools in the previous 24 hours and still present at the time of collection of faecal specimen. Patients who developed diarrhoea, after admission and as a result of food intolerance, were excluded from the study. Some children who were admitted in the same hospital, within the period of this study, for medical conditions other than diarrhoea, were included as controls. In all, 485 patients were included in this study; 365 patients with diarrhoea and 120 patients without diarrhoea

Collection of socio-demographic and clinical data
Socio-demographic and clinical data were obtained by a study nurse at the hospital ward, using a structured questionnaire. Sociodemographic data included study participants' residence, sex, age, parent/guardian occupation and educational background, source of drinking water, breastfeeding habits, whether food was prepared at home or obtained from the street, and presence of domestic animals at home. Clinical data included vomiting, malaise, fever, abdominal pain, frequency of passage and consistency of stools, drugs used for treatment or management, as well as duration, if admitted in a hospital.

Sample collection
A single stool sample was collected from each patient, by his or her parents or guardian, and placed into a clean disposable plastic container with tight fittings. The consistency of the stool was directly observed, classified and recorded by the study nurse as loose, semi-formed, formed, mucoid, slimy, or watery. Each sample was then divided into two portions, with one portion preserved in 10% formalin and the other stored at -20°C.

Stool microscopy and enzyme immunoassay test
Stool samples preserved in 10% formalin were processed for microscopy by the direct smear method after formal-ether concentration procedures, as previously described [

Detection of G. lamblia in stool samples by PCR and ELISA
Of the 365 diarrhoea samples tested, 21 and 4 were positive for ELISA and microscopy, respectively (Table 2). Interestingly, of the 120 non-diarrhoea samples tested, 1 and 6 samples were positive by microscopy and ELISA, respectively. In both cases, all ELISA positive samples also tested positive by PCR, which amplified a 461bp fragment of the gdh gene ( Figure 1). In total, 32 samples were positive for G. lamblia.

Discussion
Giardia lamblia is a medically important gastrointestinal protozoa associated with diarrhoea, especially in communities without proper sanitation and potable water. Acute diarrhoea is a major cause of mortality and morbidity among children, particularly those below 5 years, in developing countries [27]. In Ghana, previous studies have shown a steady rise in giardiasis with 9%, 9.7% and 10.1% Page number not for citation purposes 5 recorded prevalence in 2007, 2011 and 2013, respectively [10][11][12].
In this study, we recorded a prevalence of 5.8%. This reduction is encouraging and may suggest improved health conditions. Over the past 3 years, Accra has seen a major improvement in sanitation; , which was further confirmed with sequencing. This genotype is predominant in cases of giardiasis and has been associated with related symptoms, including diarrhoea [19][20][21].
However, among the non-diarrhoea cases (120), we detected G.
lamblia in 6. This may suggest that infection with Genotype B may not always lead to clinical symptoms. Prior exposure has been shown to confer some degree of protection to subsequent infections [30,31], and such cases may be reservoirs sustaining transmission to susceptible host, particularly at day-care centers, where children have frequent contact with each other [15,32,33]. Interestingly, our data is in contrast to a study by Opintan et al. [22] who identified Cryptosporidium as the commonest cause of diarrhoea in children at PML but did not find any case of Giardiasis. We do not rule out the possibility that there could be other causes of diarrhoea in children [22,34]. Assessment of clinical data on diarrhoea cases showed that over 70% of children presented with diarrhoea and vomiting, and more than 80% passed out loose watery stools. We suspect that there may be other underlying causes of these symptoms than G. lamblia. It would be interesting, in future studies, to determine if children presenting with diarrhoea may be coinfected with Cryptosporidium spp and G. lamblia. Also, we did not culture samples for presence of enteric pathogens, which have also been associated with diarrhoea [35].
Risk factors for giardiasis and other studies looking at the association between diarrhoea and some socio-demographic parameters, reviewed by Muhsen and Levine [36], have informed public health efforts in controlling infections and containing outbreaks. However, socio-economic, cultural and lifestyle differences imply that these findings may not necessarily be relevant to all social classes, particularly in developing countries where sanitation is a huge problem. In the current study, we wanted to understand how certain behavioural and socio-economic practices of mothers with children < 5 years, may pose as possible risk factors of G. lamblia infection. We observed that infection was independent of gender and age but seemed higher in ages 2-3 years (OR=3.456). These observations are parallel with two similar studies [37,38]. Elsewhere, in Lagos, G. lamblia infection was common among the age group 4-5 years [39]. In any case, during outbreaks, infections within the studied population may spread faster, as it is during this age that children tend to play more outdoors and are investigations into the role of breast milk in giardiasis. Berrilli et al. [45] suggested that infection of humans by zoonotic genotypes from domestic animals could be of low epidemiological significance, although possible. In this study, > 50% of cases had domestic animals at home but we did not find any association between pet/animal ownership by mothers and G. lamblia infection in their children. However, the zoonotic potential of genotype B cannot be ruled out [46][47][48]; we observed that >5% of cases who had animals were infected with genotype B, though not statistically significant.
We did not collect stool samples from animals to confirm possible presence of G. lamblia. On role of seasonality, we did not observe any significant difference between infections in rainy and dry • Assemblage B is probably the predominant genotype causing giardiasis in children in Accra

Competing interests
The authors declare no competing interests.

Acknowledgments
We acknowledge the nurses and physicians of the PML who assisted during the studies. We are also grateful to Dr. Anita Ghansah The funders of this study had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.