Intestinal parasites and diarrhoea in a childrens hospitalin Sri Lanka

The study population consisted of 940 children with acute diarrhoea and 260 children admitted for conditions other than diarrhoea. The first sample of stool was examined for bacteria (Sal­ monella spp, Shigella spp, Aeromonas spp, Campylobacter spp, Vibrio cholerae, Vibrio parahaemolyticus and Enteropathogenic E.coli), viruses (rotavirus and adenoviruses) and parasites (En­ tamoeba histolytica, Cryptosporidium spp, Giardia lamblia, Ascaris lumbricoides, Trichuris trichiura, Necator americanus and Enterobius vermicularis) us­ ing standard techniques.


Introduction
Diarrhoea is an important cause of morbidity and mortality in most developing countries and is the second commonest cause of morbidity among Sri Lankan children.Acute watery diarrhoea caused by bacteria and viruses can lead to death or de hydration.In contrast, intestinal parasitic infec tions associated with diarrhoea are less immedi ately dangerous, but the part they play in chronic diarrhoea and malnutrition is well documented (1,2,3).
The protozoans Entamoeba histolytica, Giardia lamblia and Cryptosporidium species and the helm inth, Trichuris trichiura are causally associated with diarrhoea.These organisms interact with the intestinal mucosa in several ways.Pathogenic amoebae can cause tissue damage through con tact cytotoxicity and by production of cytotoxins (4).Diarrhoea caused byT.trichiura is attributable to mechanical irritation and mucosal damage.Many of these parasites cause loss of body salts and minerals, but some (eg.E. histolytica and T. trichiura) cause loss of blood and proteins from damaged intestinal mucosa.

The influence of parasites on episodes of child hood diarrhoea has not been examined in Sri
Lanka.We report here our observations on the interactions and effects of parasitic infestation in children presenting with diarrhoea.The isolation and identification of Salmonella spe cies, Shigella species, Vibrio species, Aeromonas spe cies and Enteropathogenic E.coli (EPEC) were car ried out using standard laboratory techniques (6).Cold enrichment followed by culture on MacConkey agar was used for the isolation of Yersinia enterocolitica.

Each patient's clinical history, examination, nu tritional status treatment and response to therapy was recorded on a clinical history sheet by the
Senior House Officer.The nutritional status was determined by using the weight relative to age.Data were analysed using the standard error of difference of proportions.

Results and Discussion
Of 940 cases 87 (9.3%) had parasitic infection which was significantly higher than among con trols.Only 11 of 260 (4.3%) controls had parasites in their stools.Sixty one of the 87 cases (70.1%) had parsites which are known to cause diarrhoea among which Cryptosporidium species and T. trichiura were the predominant parasites (Table 1).In 39 of these 87 patients (44.8%), the parasite was the only pathogen that was detected.In 48 patients parasites were found concomitantly with bacterial and viral pathogens.
Cryptosporidium was the commonest parasite de tected indicating that it would be good practice to seek this pathogen routinely in stools of chil dren with diarrhoea (Table 1).This would pre vent unnecessary treatment with antibiotics.In fact it has been reported that treatment with an tibiotics prolonged the duration of diarrhoea due to Cryptosporidium (9).

Children affected with Cryptosporidium
and G.lamblia were younger than children with worm infestations.However this finding was not statis tically significant (Table 2).Malnutrition was ob served in a significantly higher proportion of chil dren infected with G.lamblia and T.trichiura (p < 0.05) than those who were infected with other parasites (Table 2).Diarrhoea associated with parasites was seen significantly more in malnour ished than in well nourished children (Table 3).
In the present series 28.7% and 9.7% of children with Salmonella and rotavirus diarrhoea respec tively had blood and mucus stools (Table 4).How ever in instances where rotavirus infection was associated with parasitic infections blood and mucus stools were observed in 50% of cases.Simi larly, when Salmonella diarrhoea was associated with parasitic infestations blood and mucus stools occurred in 87.5% of cases.Both these differences were statistically significant, p < 0.05.(Table 4).Therefore the presence of parasites seem to pro duce more severe disease in children with diar rhoea.Table 5 shows the associated parasitic in fections and the nature of the stool in diarrhoea caused by Salmonella species and rotavirus.

The Ceylon Journal of Medical Science
Although it is documented that Cryptosporidium infection caused either a watery or mucoid diar rhoea (5), in the present study blood and mucus stools was observed when Cryptosporidium was associated with bacterial or viral pathogens.Five out of 9 cases presented with blood and mucus diarrhoea (Table 5).
In this study 10 cases were observed with only T.trichiura infection among which 5 or 50% had blood and mucus stools.Six patients had either Salmonella or rotavirus in addition to T.trichiura infection, and all 6 had blood and mucus stools (Table 5).T.trichiura infections lead to blood and mucus stools only when the worm load is very heavy (10).We are unable to comment on the worm load, but, none of these children had rectal prolapse which is a feature of heavy worm load.The average duration of diarrhoea due to rotavirus was prolonged to over 14 days when associated with roundworm, whereas the mean duration of uncomplicated rotavirus diarrhoea is about 8 days (11).This may suggest that concomi tant roundworm infestation predisposes to a more prolonged and severe clinical illness.However the stools were not examined for all known bacterial and viral pathogens ie, Entero-haemorrhagic and Entero-invasive E coli (which give rise to blood and mucus diarrhoea), Norwalk virus etc, which is a limitation in the study.
From the above results and discussion it appears that parasites alter the course of an acute diar rhoea.This may be due to direct action of the para site or parasite products on the intestinal mucosa or due to indirect factors such as malnutrition and malabsorption.Further studies should be carried out to determine the pathogenic mechanisms ex erted by parasites on the intestinal mucosa.Fur thermore multiple aetiological agents may be re sponsible for severe and prolonged diarrhoea and should be investigated for, when facilities are available.
In most laboratories in Sri Lanka, diarrhoea stools are often examined for bacterial pathogens and sometimes for amoebae when there is blood and mucus.A simple faecal smear is adequate for screening for parasites.We recommend that in developing countries parasitological investigation should complement bacterial and viral diagnosis in patients presenting with diarrhoea illness be cause specific therapy instituted against the para site may alter the course of that diarrhoeal epi sode as well as future episodes.
examined 940 consecutive faecal specimens from children aged one month to twelve years who were admitted from August 1987 to April 1989 to Ward number 2 of Lady Ridgeway Childrens Hospital with acute diarrhoea and 260 age matched controls admitted for conditions other than diarrhoea.The first sample of stool obtained after hospitalisation was examined for the presence of bacterial, viral and parasitic agents.Trophozoi tes and cysts of £. histolytica and G. lamblia and ova of Ascaris lumbricoides.T. trichiura, Necator americanus and Enterobius vermicularis were de tected in fresh stools by observing unstained direct mounts and iodine stained mounts.Cryptosporidium oocysts were detected by stain ing faecal smears by a modified Ziehl-Neelsen technique (5).Rotavirus was detected by Enzyme Immunoassay (EIA) (DAKOPATTS, Denmark) and enteric adenoviruses, by latex agglutination (Adenolexorion Diagnostic, Finland).

Table 3 Nutritional status (Gomez classification) of children with diarrhoea in relation to the pathogen detected Organism Nutritional status
* Children in PCM2 \ PCM3 groups had more parasite associated diarrhoea compared to children in normal / PCM1 group (p < 0.