Prevalence and identification of Cryptosporidium species in paediatric patients with diarrhoea

Methods A total of 138 diarrhoeic faecal samples were collected between August 2011 and February 2013, from children under 12 years of age, admitted to paediatric units of Teaching Hospitals, Kandy and Peradeniya, Sirimawo Bandaranayake Childrens’ Hospital, Peradeniya and District General Hospital, Matale. One hundred faecal samples collected from healthy children were used as controls. All control and test samples were screened for the presence of Cryptosporidium oocysts with Modified Ziehl-Neelsen (MZN) method and PCR.


Introduction
Cryptosporidium species are intestinal coccidian parasites that infect a variety of animals including humans.

Abstract
Objectives To determine the prevalence of Cryptosporidium infection in children with diarrhoea, identify associated factors and identify the parasite using Polymerase Chain Reaction (PCR).
Methods A total of 138 diarrhoeic faecal samples were collected between August 2011 and February 2013, from children under 12 years of age, admitted to paediatric units of Teaching Hospitals, Kandy and Peradeniya, Sirimawo Bandaranayake Childrens' Hospital, Peradeniya and District General Hospital, Matale.One hundred faecal samples collected from healthy children were used as controls.All control and test samples were screened for the presence of Cryptosporidium oocysts with Modified Ziehl-Neelsen (MZN) method and PCR.
Results Prevalence of Cryptosporidium infection among children with diarrhoea was 5.7%.All the cases positive for Cryptosporidium were below 3 years of age.The majority (7 out of 8) of the positive cases had watery diarrhoea while none of the healthy children excreted Cryptosporidium oocysts in the faeces.Of the 8 positive cases, 6 had a history of animal contact.A large proportion of positive cases used pipe borne municipal water.The majority (66.6%) of positive cases did not consume boiled cooled water.We were able to identify C. parvum from one of the eight cases that had diarrhoea.

Conclusions
The current study shows that Cryptosporidium is one of the aetiological agents responsible for childhood watery diarrhoea in Sri Lanka, thus stressing the importance of routine stool examination for Cryptosporidium oocysts.This study recommends boiling water as an important measure to prevent the transmission of Cryptosporidium oocysts.Further molecular studies are needed to determine the other species of Cryptosporidium responsible for cryptosporidiosis in children in Sri Lanka.
The two main species of Cryptosporidium that infect humans are Cryptosporidium hominis and Cryptosporidium parvum, the former primarily infects humans [1].However, C. parvum naturally infects several animal species that serve as reservoirs for zoonotic infection, including cattle, sheep and goats.C. hominis is mostly associated with outbreaks of cryptosporidiosis in developed countries indicating human to human transmission [2].In developing countries cryptosporidiosis is a major cause of childhood diarrhoea.This disease is known to cause self-limiting infection in immunocompetent individuals.However, it causes severe and lifethreatening diseases in immune-compromised patients [3].Diagnosis of cryptosporidiosis is mainly based on the detection of Cryptosporidium oocysts in faeces.Modified Ziehl Neelsen (MZN) staining method is widely used as a reliable routine technique to detect Cryptosporidium oocysts.However, MZN staining is not carried out in routine stool analysis in Sri Lankan laboratories.Thus, cryptosporidiosis remains largely under diagnosed by the current routine laboratory practice in Sri Lanka.Environmentally resistant Cryptosporidium oocysts are passed in the faeces of asymptomatic and symptomatic animals and are acquired by ingestion [4,5].In developing countries, Cryptosporidium is responsible for 8% to19% of cases of diarrhoeal diseases, with significant mortality.Similarly, a few studies done in Sri Lanka have identified Cryptosporidium oocysts in diarrhoeic faecal samples of children, and results have shown a prevalence between 6% to 9% [6,7].In Sri Lanka, Cryptosporidium infection has been reported in cattle, buffalo, goats and monkeys indicating the presence of potential zoonotic transmission [8,9].As all species do not appear to cause infection in humans, and the majority of human infective species cannot be differentiated microscopically because of their identical morphology, both microscopic and molecular methods, which can differentiate Cryptosporidium species are required to identify potential sources of infection.To date there have been no studies to identify the Cryptosporidium species responsible for human infection in Sri Lanka despite the highlighting of possible zoonotic risk.Thus information about circulating Cryptosporidium Papers species and genotypes are not available in our country.Hence, the current study was undertaken to determine the prevalence of Cryptosporidium infection in children with diarrhoea, identify associated factors and identify Cryptosporidium species isolated from infected children with diarrhea.

Data collection
Between August 2011 and February 2013, faecal samples were collected from children with diarrhoea under 12 years of age admitted to paediatric units of Teaching Hospitals, Kandy and Peradeniya, Sirimawo Bandaranayake Childrens' Hospital, Peradeniya and District General Hospital, Matale.WHO criteria were used to determine the nature of the diarrhoeal episodes.An information sheet recorded age, sex, address and clinical symptoms related to the illness for each child.Information regarding the possible mode of acquiring Cryptosporidium infection such as source of drinking water, method of water purification, exposure to animals and personal hygiene were collected from parents or the guardian of the child after obtaining informed consent.Faecal samples of the control group were collected from age and sex matched children presenting with illness other than diarrhoea to the same units.

Detection of Cryptosporidium spp. oocysts in faecal samples
Half of each faecal sample was concentrated with modified formalin ethyl acetate technique as this method is more sensitive for the recovery of Cryptosporidium oocysts [10].Subsequently, each concentrated sample was stained with MZN method.If the faecal sample was positive for Cryptosporidium oocysts the rest was subjected to Sheather's sugar flotation technique and the concentrate was washed several times with distilled water and stored at -20°C until use.

DNA extraction
Oocysts stored at -20°C were defrosted at room temperature.Each oocyst sample was suspended in lysis buffer (1M NaCl, 0.5M EDTA, 1M Tris HCl and 10% SDS).Then, the suspension was subjected to 15 cycles of freezing and thawing by immersing in liquid nitrogen and thawing in a water bath at 60°C.Subsequently the debris was removed by centrifugation and the sediment was treated with proteinase K (20mg/ml).DNA was extracted by phenol-chloroform method and the extracted DNA was suspended in sterile water and stored at -70°C until use.

Results
A total of 138 diarrhoeal faecal samples were collected.Fifty were from males.Of the 138 faecal samples tested for Cryptosporidium oocysts, 8 (5.7%) were positive for MZN staining on microscopy.Cryptosporidium oocysts were not found in the 130 faecal samples of healthy children without diarrhoea.The majority of the children positive for Cryptosporidium oocysts were aged one to three years.Four of the positives were male.
In the study group, 67 (48.5%) gave a history of contact with animals.Of the 8 positives 6 gave a history of animal (dogs, cats, and goats, chicken) contact.In those who had animal contact, a large proportion (59.7%) had dogs and of these three cases were Cryptosporidium positive.Of the seven children who had contact with goats, two were positive for Cryptosporidium.Fever was the predominant symptom in these children (109 out of 138) and six out of eight Cryptosporidium positive cases presented with fever.
The study group obtained drinking water from municipal water supply, protected or unprotected wells and tube wells.The majority used piped borne water.Of the Cryptosporidium positives, a large proportion used piped borne water (Table1).Most in the study group used boiled water while three did not drink boiled cooled water.Of these three, two were Cryptosporidium positive (Table 2).

Discussion
The prevalence of Cryptosporidium in different countries of the world vary, however, the rate of infection is higher in children in countries [11].In this study, Cryptosporidium oocyst positivity of 5.7% was observed in children with diarrhoea.This rate of positivity is comparable to that reported previously [6,7].Similarly, the prevalence of Cryptosporidium infection in children in the neighboring countries, India and Nepal have been 5.6% and 4.1%, respectively [12,13].Much higher prevalence of this infection has been reported among children in Nicaragua (35.7%) and Egypt (33.3%) [14,15].In contrast, lower prevalence rates of 1.5% to 2.7% have been reported from Iran and Bangaladesh [16].These differences could be attributed to standards of living, socioeconomic status and availability of safe drinking water.
In the present study, the majority of the positive cases had watery diarrhoea.Similar findings have been reported in India where the greater proportion of cryptosporidiosis cases had watery diarrhoea [17].None of the age and sex matched control group in this study were positive for Cryptosporidium infection.This is in agreement with the previous studies [18,19].Contrary to this, asymptomatic carriage rate of 2% to 3% has been reported in India [20].
All the positive cases in this study group were below three years of age.Similar results have been reported in studies done in Ghana and Iran where the Cryptosporidium infection rate is high among children under five years of age [21,22].
Cryptosporidium is transmitted through water, person to person contact and from animals to humans.Cryptosporidium oocysts are highly resistant to common water disinfection practices including chlorination.The majority of Cryptosporidium infected children in the study group used pipe borne municipal water.Similarly, a study done in the USA shows high prevalence of cryptos-poridiosis among those who drink municipal water [23].Cryptosporidium oocysts have been recovered from irrigation reservoirs and canals in Sri Lanka [24].A large proportion of the study group used boiling as a method of water purification.Although boiling inactivates the Cryptosporidium oocysts in water, there were four Cryptosporidium infected children among this group.This may be due to improper boiling of water or due to consumption of contaminated food.Food such as fresh fruits, vegetables and raw or undercooked shell fish are increasingly recognised as sources of Cryptosporidium spp [25].It is interesting to note that one quarter of those who had Cryptosporidium infection did not consume boiled cooled water.This supports the explanation that municipal water purification is not effective in eliminating Cryptosporidium oocysts.Zoonotic transmission of cryptosporidiosis after exposure to infected animals has been documented [26].The majority of positive cases in the current study had a history of contact with animals.In Sri Lanka, calves and goats are considered to be major animal reservoirs [27].
Although cryptosporidiosis is prevalent in Sri Lanka no studies have been carried out to identify the Cryptosporidium species.This is the first study that reports of C. parvum causing diarrhoea in children in Sri Lanka.The rest of the MZN positive samples were negative with PCR using C. parvum specific primers.Those cases could be associated with other Cryptosporidium species.The majority of human infections are caused by the species C. hominis and C. parvum.In European countries, equal percentages of C. hominis and C. parvum have been documented, however, in developing countries the majority of human infections are due to C. hominis [1].In India too the commonest species isolated from the faecal samples of children with Cryptosporidium diarrhea is C. hominis [28].In this study only one isolate was identified as C. parvum and the rest was PCR negative using C. parvum specific primers.Since we only used C. parvum specific primers our species identification was confined to C. parvum.
In conclusion, Cryptosporidium is one of the possible aetiological factors for childhood watery diarrhoea in Sri Lanka necessitating routine stool examination for Cryptosporidium using MZN staining.Boiling water may be useful to prevent the transmission of Cryptosporidium oocysts.Molecular studies such as restriction fragment length polymorphism analysis and DNA sequencing are needed to determine the other species responsible for cryptosporidiosis in Sri Lanka.