Diarrhoeal disease is the world's second leading cause of death among young children (UNICEF, 2013). The primary cause of diarrhea is the ingestion of infectious agents, bacterial, viral, or protozoan, from human or animal feces. Transmission occurs through anthroponotic, zoonotic, foodborne, and waterborne routes (Moore et al., 2016). It consists of ingestion of contaminated food or water, direct contact with infected feces, person-to-person contact, and poor personal hygiene (Boithias et al., 2016).

Intestinal protozoa that are most commonly associated with diarrhea in children include Giardia duodenalis, Blastocystis hominis, Entamoeba histolytica, Cryptosporidium spp., and Dientamoeba fragilis (Maas et al., 2014). Symptoms of infection follow an incubation period of approximately 1 week and include watery diarrhea, cramp, and nausea that may lead to inappetence and undernutrition, as observed in children in West Africa and South America (Quihui-Cota et al., 2015). The World Health Organization (WHO) has identified Cryptosporidium spp. as globally the most common diarrhea-causing protozoan (Pedersen et al., 2014), and the Center for Disease Control and Prevention has reported that giardiasis is experienced by approximately 33% of people in developing countries (CDC, 2015). Accordingly, due to their significant public health and socioeconomic implications, both parasites Cryptosporidium spp. and G. duodenalis were included in the WHO's “Neglected disease initiative” in 2004 (Osman et al., 2016). A wide diversity of Cryptosporidium spp. and subtypes infect humans and molecular analysis of the GP60 marker has been useful for identifying contamination sources and thus inferring their transmission routes (Cacciò and Chalmers, 2016). The cosmopolitan intestinal protozoa B. hominis and D. fragilis are both also of relevance in this context (Stark et al., 2016). D. fragilis is frequently detected as a co-infection with multiple protozoa (Garcia, 2016) especially with B. hominis, which is highly prevalent in Qatar (Abu-Madi et al., 2016a).

Conventional diagnosis of these protozoa relies widely on microscopical examination of stool samples. However, coproscopic identification is limited by a low sensitivity as evidenced by earlier studies reporting a marked difference in the prevalences of B. hominis and G. duodenalis when comparing the results of coproscopic and molecular analyses (Abu-Madi et al., 2015; Gotfred-Rasmussen et al., 2016). Moreover, coproscopic analysis is unable to distinguish between the pathogenic and non-pathogenic species of Entamoeba (Ali, 2015), between different species of Cryptosporidium and different assemblages of Giardia (Haque et al., 2009) as these are morphologically identical within genera. Hence there is an urgent need to adopt more sensitive and specific methods for species identification (Boughattas and Salehi, 2014), as accurate diagnosis will lead to early and more appropriate treatment (Maçin et al., 2016). By employing real-time PCR (RT-PCR), the aim of the current study was to investigate the frequency of intestinal parasites in pediatric patients with chronic diarrhea and to clarify the importance of individual species and relevant risk factors in such cases.

Table 2 shows the distribution of subjects by region of origin and sex. Numbers in each age class are given in Figure 1. As can be seen, the largest category was children from Asia, and the smallest category was that referred to as “other” representing subjects from Europe and North and South America.

While a range of different pathogenic organisms can cause pediatric diarrhea in Qatar, including rotaviruses and adenoviruses (Al-Thani et al., 2013) and foodborne bacteria such as, Eschirichia coli (Weam et al., 2016), in the current study we have shown that a significant percentage of cases of pediatric diarrhea (22.9%) is also associated with the presence of intestinal protozoan parasites. The most common among these was C. parvum, accounting for 14.7% of cases.

Cryptosporidiosis is globally a widespread disease with prevalence in humans reported to range between 1.1% in Iran (Tahvildar-Biderouni and Salehi, 2014) to more than 60% (El Shazly et al., 2007; Abdel-Hafeez et al., 2012). Prevalence of B. hominis in humans has been found to vary from 0.5 to 76% (Abu-Madi et al., 2015) and even as high as 100% among Senegalese children (El Safadi et al., 2014) and that of D. fragilis from 0.3 to 89% (Garcia, 2016). In our current work, we observed a prevalence of 4% for D. fragilis which is similar to prevalence values reported from Tunisia 5·5% (Ayadi and Bahri, 1999) and Australia 5% (Stark et al., 2016). Moreover, we also found evidence for a positive association between this species and B. hominis and this is consistent with reports on pediatric populations in Lebanon (Osman et al., 2016), and Portugal (Júlio et al., 2015).

Host intrinsic factors including age and sex have been assessed to identify potential factors that may influence the risk of infection with B. hominis in pediatric patients elsewhere (Nithyamathi et al., 2016) and in agreement with earlier studies, prevalence of infection with this species in our study was also found to increase significantly with host age. An age-effect was also detected for D. fragilis, however, there was no evidence in our data for an age effect in the case of C. parvum. The increase of infection prevalence with host age that we detected in the cases of B. hominis and D. fragilis may be related to changes in breast feeding practices with a reduction in exclusive breastfeeding as the children grow older. Although we did not find a significant increase in prevalence with host age in the case of Cryptosporidium spp., failure to breast feed is considered to be a risk factor for cryptosporidiosis (Aldeyarbi et al., 2016). Indeed mothers provide protective passive immunity against different parasites infection (Abdel-Hafeez et al., 2013; Pedersen et al., 2014), as reflected for example in the high levels of anti-Cryptosporidium IgA in maternal breast milk (Korpe et al., 2013). G. duodenalis infection has also been reported to be significantly less frequent among breastfed children (1.9%) compared with those not on breastfeeding (6.5%) in Egypt as well as among Rwandan children (Tellevik et al., 2016), but in our study the number of cases was too few to allow statistical analysis of any age effect in this species.

Prevalence of B. hominis was found to differ significantly between the sexes being twice as high among female children in contrast to observations elsewhere reporting higher prevalence among male subjects (Nithyamathi et al., 2016; Osman et al., 2016). The contrasting findings suggest that the distribution of intestinal protozoan infections is spatially heterogeneous with respect to host sex, and differences between studies are most likely linked with variation in local environmental conditions in the communities, and as a consequence with physiological properties intrinsic locally to hosts.

Intestinal protozoan infections are considered to be transmitted via the fecal-oral route, contaminated water/food (Anvari-Tafti et al., 2016), close contact with animals, poor personal hygiene and inadequate sanitation, and are facilitated by the environmentally resistant cyst stage of the parasites. Drinking contaminated water, especially surface water, is a known and significant risk factor for B. hominis infection (Abdulsalam et al., 2013). Cryptosporidiosis is also primarily spread by ingestion of contaminated water, and was ranked fifth among the 24 most important food-borne parasites (FAO, 2012). Among gastroenteritis outbreaks reported in England and Wales during the 1990s, Cryptosporidium spp. were identified as the probable cause in 14 outbreaks associated with public water supplies and swimming pools (Insulander et al., 2005; Suppes et al., 2016) as these species are resistant to chlorine at the levels usually used in water treatment (Maçin et al., 2016). The usage of surface water for irrigation can lead also to human infections through the consumption of contaminated fresh raw vegetables (FDA, 2015). Indeed, Cryptosporidium spp. have been detected on Cilantro leaves in Costa Rita and on lettuce in Norway (Sunnotel et al., 2006) and outbreaks of cryptosporidiosis have been observed to be associated with the consumption of fresh pre-cut salad leaves in England (McKerr et al., 2015) and fresh parsley in Sweden (Insulander et al., 2008). Green leafy vegetables such as, lettuce, rocket, and parsley have uneven surfaces and this provides a convenient substrate for the attachment of parasite eggs, cysts, and oocysts, either at the farms where they are grown or subsequently when washed with contaminated water (Said, 2012).

Our analysis has shown that prevalence of combined protozoan infections, and C. parvum in particular, was highest among the Qatari, despite Qatari nationals representing only 12% of the total population in the country (Qatar's Ministry of Development Planning and Statistics; Abu-Madi et al., 2016b). The most plausible routes for infection are via day-care centers (Clavel et al., 1996), swimming pools (Suppes et al., 2016), and food (as the consumption of salads is frequent in the country even in fast food restaurants). As children in Qatar have little contact with farm animals and the country relies on desalinated sea water that is piped to houses and used for drinking, these two potential routes for transmission are unlikely (Nazemalhosseini-Mojarad et al., 2012). A Portuguese study reported that the risk of acquiring this infection was 12 times greater for children who attended school, suggesting probable human-to-human transmission (Júlio et al., 2015). Foreign workers may also play a role in the transmission of intestinal parasites. In Saudi Arabia, pediatric parasitic infections (including Cryptosporidium spp.) have been suggested to be associated with the widespread custom of employing foreign maids who usually have direct contact with children. A high prevalence (46.5%) of intestinal parasites among female Asian housekeepers in that study, suggests that members of families with foreign housekeepers may be at some risk of parasitic infections (Ghenghesh et al., 2012).

As the etiological agents of diarrhoeal diseases occur naturally in both temperate and tropical areas, epidemics arising from the different agents may show specific seasonality. Meteorological conditions are known to have different effects on the transport, spread, reproduction, and persistence of a range of pathogens (Boithias et al., 2016). Concurring with our results, a recent pediatric study in Thailand, found a higher prevalence of intestinal parasites in the cool season compared with the hot season (Tellevik et al., 2016). In neighboring countries, such as, Kuwait, where high temperatures and a very dry atmosphere are typically experienced during the summer, and are followed by a short cool winter as observed in Qatar (Boughattas et al., 2017), prevalence of cryptosporidiosis also shows seasonality with highest prevalence during winter months (Iqbal et al., 2001). There is no regular rainy season in Kuwait or in Qatar, and thus the possibility of water-borne transmission attributable to natural sources of water is minimal. However, most studies relate seasonality of outbreaks of diarrhea positively with relative humidity (Jagai et al., 2009), whereas others report an opposite effect (Kiani et al., 2016) with correlation to water shortage and to greater reliance on less safe water sources (e.g., stored water) at certain times of the year. Interestingly, the period of high prevalence of infection in Kuwait has been reported to correspond to the “winter desert camping” season in the country when a large number of families spend an extended period of time in desert camps supplied with hundreds of overhead water tanks (Iqbal et al., 2011). A Spanish study reported marked seasonal variation in diarrhea, with peak prevalence in winter among 1-to 3-year-olds, which seemed to be related to attendance at child care centers, as the attendance was lowest in summer due to holidays (Clavel et al., 1996). Thus, this lack of consistency among seasonality studies and the absence of a straightforward relationship between ambient temperature, water quality and incidence of diarrhea highlight the complex relationship between hydro-meteorological and ecological factors and the transmission of waterborne diseases.

The distribution of Cryptosporidium spp. in humans is known to vary with geographic region and also with the socioeconomic conditions of the study population. Our findings are consistent with previous reports from other Middle Eastern countries (Nazemalhosseini-Mojarad et al., 2012). In Saudi Arabia, C. parvum was identified in 43 of 53 examined samples (Alyousefi et al., 2013) and in Kuwait C. parvum was the predominant causative agent of cryptosporidiosis in children (Sulaiman et al., 2005; Iqbal et al., 2011). Gp60 subtyping has shown that not all C. parvum infections in humans are of zoonotic transmission (Xiao, 2010). Within each subtype family, one subtype is frequently seen to dominate in specific communities/countries. For example, in industrialized countries most of C. hominis infections are caused by the Ib subtype family (Waldron et al., 2009), which appears to be the most virulent as reflected by the numerous outbreaks with which it has been linked. Worldwide there are two common subtypes within the subtype family Ib: IbA9G3 and IbA10G2. IbA9G3 is commonly seen in Malawi, Kenya, and India, whereas IbA10G2 is commonly seen in South Africa, Botswana, Jamaica, and Peru (Xiao, 2010). Using GenBank BLASTN, a similar IbA9G3 subtype to that identified in our isolates has been observed among children in North India (Gatei et al., 2007) and Bangladesh (Hira et al., 2011). The IbA10G2 subtype was reported to be involved in waterborne outbreaks especially in Sweden (Gherasim et al., 2012) and almost half of the C. hominis outbreaks in USA (Cacciò and Chalmers, 2016). It has been also associated with swimming pool outbreaks in Australia (Ng-Hublin et al., 2015).

Sequence analysis of the GP60 locus in the present study has identified only C. parvum subtype family IId. The source of the IId subtype in humans is not yet clear. It is the dominant C. parvum subtype family in humans in Middle Eastern countries such as, Iran, Syria (Kassouha et al., 2016), and Saudi Arabia (Nazemalhosseini-Mojarad et al., 2012). The most frequent subtype observed in our samples was the IIdA20G1 as identified predominantly in Egypt, Jordan (Helmy et al., 2013; Hijjawi et al., 2016), and among Kuwaiti children (Sulaiman et al., 2005; Iqbal et al., 2011). Recurrently identified in our population, subtype IIdA17G1, was reported to be the cause of an outbreak in Finland in 2012, with more than 250 individuals involved and it was associated with consumption of fries and salads (Cacciò and Chalmers, 2016), whereas subtype IIdA18G1 has been widely identified in Europe (e.g., Czech Republic, Spain) and in the Middle East (e.g., Turkey, Iran, Kuwait, etc.). Less frequently observed, subtype IIdA19G2 has been detected in France as a foodborne etiological agent (GenBank number KT716847.1), IIdA14G1 in Spain (Plutzer and Karanis, 2009), and IIdA16G1 in Egypt and Tunisia (Rahmouni et al., 2014). C. meleagridis, mainly isolated from poultry and birds, has been frequently observed in Africa (Aldeyarbi et al., 2016) as well as within mixed cryptosporidium infections (Ng-Hublin et al., 2013). Although, as stated earlier, Qatari children have limited contact with wildlife, such animal subtypes may infect them indirectly through housemaids who themselves may have been exposed to animals back home and/or through their presence in the food supply via remnants of fertilizers of animal origin on salad crops.

In our previous work, we suggested that transmission of some protozoa does occur in Qatar and our current results support this hypothesis based on the high rate of infection among Qatari subjects. Cryptosporidiosis has been shown to be the most frequent pediatric diarrheic infection. Food borne infective stages and a human-to-human role appear to constitute the most probable routes of transmission. In order to better understand this parasite's circulation among the population of Qatar, further studies targeting housemaids and food handlers are required.

SB carried out the experiments and drafted the manuscript; JB carried out the statistical analysis and participated in the preparation and refinement of the manuscript; KA collected the samples and their background information; AS contributed to the molecular detection; WA contributed in sequencing investigation; AA contributed to the interpretation of the results and participated in refinement of the manuscript; MA conceived the study, collected the data and revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript. The contents of this report are solely the responsibility of the authors and do not necessarily represent the official views of Qatar University and QRNF.