Diagnosis and epidemiology of "Strongyloides stercoralis" in Cambodia

Background: The soil-transmitted helminth Strongyloides stercoralis is the most neglected of all neglected helminth infections. It is known to occur worldwide, more prominently in tropical and subtropical resource-poor areas. Environmental and socio-economic factors such as inadequate sanitation contribute to high infection numbers. Today’s estimates for the number of people infected with S. stercoralis globally are cited to be around 30-100 million people, however based on insufficient studies. Compared to the three widely known STHs, namely Ascaris lumbricoides, hookworm and Trichuris trichiura, in most countries, still only fragmentary information is available about the prevalence of S. stercoralis. A main factor contributing to this knowledge gap is the lack of high sensitivity diagnostic methods. The most commonly applied diagnostic methods for helminths have a very low sensitivity for detection of S. stercoralis. There are more appropriate diagnostic methods, yet they are not routinely applied. Furthermore these methods rely on multiple sampling for sensitivity increase, mainly because the irregularities with which S. stercoralis larvae are excreted in low-intensity infections. With Ivermectin, a safe and highly efficacious drug is available for the treatment of S. stercoralis, yet it is not routinely used in helminth control programs.
Cambodia, located in Southeast Asia, provides ideal conditions not only for STH-infections, but also other helminthic and protozoan infections. However, there is still only little data available on S. stercoralis. In the rural communities, where people mainly rely on subsistence farming, there is a constant exposure to risk factors for STH-infection.
Furthermore, specific details about the parasite, its biology and genetics, as well as transmission dynamics remain poorly understood. E.g., excreted larvae are detected in diagnostic tests, however, excretion dynamics are poorly investigated. Another aspect important for control is the possible zoonotic potential of S. stercoralis. Strongyloides infects animals and humans. If and how frequent infection between animals and humans occurs is unknown.
Aims and objectives: The work presented in this PhD was divided into three sections. The first section of the work consisted in examining today’s literature to draw a more accurate picture of the global prevalence, diagnosis and risk factors for S. stercoralis. Furthermore we aimed at analysing information about the treatments available for S. stercoralis today. The second section focused on S. stercoralis in communities in an endemic setting. We aimed of assessing co-infections, risk factors and transmission dynamics. The third section focused on the individual level of infection with S. stercoralis. We aimed to assess the excretion pattern in individuals before treatment and the impact of treatment, we genotyped larvae collected from humans and animals, to get more insight into possible transmission between humans and animals, and we validated a new PCR assay for the detection of S. stercoralis.
Methods: For the first section a literature search was conducted including all available studies reporting on the prevalence of S. stercoralis in communities and from hospitals. The data was combined and used for a meta-analysis to give prevalence estimates per country. Further were possible risk factors analysed with the help of a meta-analysis. A second review was conducted collecting information on the treatment of strongyloidiasis. Pooled cure rates were established for all drugs included in a meta-analysis.
The second and third section of this work were based on field work conducted in Kandal, Preah Vihear and Takeo province in Cambodia. To assess multiparasitism in humans and animals, a community-based survey was conducted. Inhabitants of rural villages were enrolled, and stool samples collected. The samples were analysed with several diagnostic methods: Kato Katz, Baermann, Koga Agar plate culture, formalin-ethyl acetate concentration method and PCR. Animals samples were analysed with the flotation method and PCR.
The studies reported in the third section were undertaken in the framework of our large-scale community-based surveys in Preah Vihear and Takeo province. For the study on the excretion pattern of S. stercoralis larvae, we used a modified Baermann method. By weighing the amount stool used, we established the larvae per gram stool that were excreted. For the genotyping of single worms, the worms were collected either from the Baermann or Koga Agar plate culture and fixed in 70% ethanol for subsequent molecular analysis. For the collection of S. stercoralis from animals, a veterinarian was appointed. This enabled us to collect the stool directly from the animals.
Results and conclusions: In total, 354 studies were included in the S. stercoralis prevalence review. Almost two thirds (63.3%) of these studies applied low sensitivity diagnostic methods. Serology was only used in 9.9% of the studies. Only two countries report 40 or more studies, namely Brazil (43 studies) and Thailand (40 studies). For many countries, potentially endemic for S. stercoralis, no data is available, e.g. in sub-saharan Africa where we could identify studies only in 20 (43.5%) of the 46 countries.
We conclude that the prevalence of S. stercoralis is still underreported. We demonstrated that there are still many gaps and “black spots” on the world map remaining today. Diagnostic methods most commonly applied in studies on STH infections today are not appropriate for the detection of S. stercoralis, which is one of the main factors for the underreporting of the prevalence.
We included 68 studies for the review on treatment. Efficacies of 14 different drugs were reported. The four drugs tested the most include Ivermectin (in 30 studies), Albendazole (in 23 studies), Thiabendazole (in 21 studies) and Mebendazole (in 5 studies). A meta-analysis revealed the best pooled cure rate for Ivermectin with 0.97 (95% BCI: 0.94 – 1.00) and Thiabendazole with 0.95 (95% BCI: 0.92 – 1.00).
We conclude that Ivermectin is the best available treatment. Our studies further confirmed the effectiveness and safety in application of ivermectin. As ivermectin already has been applied in mass drug administration (MDA) programmes, for instance in the treatment of lymphatic filariasis, inclusion of ivermectin in helminth control MDA-programmes is an option to be considered in endemic areas. Other potential candidates we report for treatment of S. stercoralis need to be further evaluated, as the heterogeneity of available data (diagnostic methods applied, reports of adverse events) makes it difficult to assess and compare the established cure rates.
Our field survey conducted in rural villages in Preah Vihear province, Cambodia, revealed that intestinal multiparasitism was very frequent, in humans as in animals. From 218 humans examined, 191 (87.2%) had at least one intestinal parasitic infection. We could document 14 different parasitic species; eight helminthic and six protozoan parasites. In humans, most frequently diagnosed were hookworms (63.3%), Entamoeba spp. (27.1%), S. stercoralis (24.3%), Giardia duodenalis (22.0%) and Blastocystis hominis (18.4%).
In the 94 dogs sampled, the most common infections diagnosed were hookworm (80.8%), Spirometra spp. (21.3%) and Strongyloides spp. (14.9%). Altogether, we detected eleven parasitic species (eight helminhts and three protozoans). Seventy-seven (81.9%) had at least one infection diagnosed. In 76 pigs, 74 (97.4%) had at least one infection diagnosed. Of the twelve detected parasitic infection (eight helminths and four protozoan), the most frequent were Isospora suis (75.0%), Oesophagostomum spp. (73.7%) and Entamoeba spp. (31.6%).
The molecular analysis of hookworms revealed that 50.8% of the hookworm infected humans are due to A. ceylanicum. In the dogs positive for hookworm, 90.0% were infected with A ceylanicum.
In conclusion, we could demonstrate the high prevalence of intestinal multiparasitism in rural villages in Cambodia, not only in humans but also in domestic animals. We documented that infection with A. ceylanicum is rampant in humans in rural villages, and we showed the potential for zoonotic transmission between dogs and humans. We argue that for prevention and control, integrated approaches are needed, including not only the humans but the animals as well.
Our studies focusing on the individual level of strongyloidiasis infection, we analysed the excretion pattern of S. stercoralis larvae. Thirty-nine positive cases were followed for seven days prior to ivermectin treatment. Larvae per gram stool (LPG) were calculated. The highest and lowest average density per person observed were 151.2 LPG (range: 28.8 – 410.6 LPG) and 0.03 LPG (range 0 – 0.07 LPG), respectively. Local maxima estimation technique did not detect a pattern of excretion, indicating that larvae excretion is random and does not follow a cyclic pattern. High-intensity infections (>10 LPG every day) were correctly diagnosed S. stercoralis positive every day. In low-intensity infections, patients did not excrete larvae for up to two consecutive days. They would have been wrongly diagnosed as infection-free in a one day examination.
Those results underline that low-intensity infections can easily be missed if only single stool samples are analysed, as the individuals can have excretion-free days. This further emphasizes the importance of analysing samples of the same individual on consecutive days to increase sensitivity of the diagnostic methods and to subsequently better estimate the true prevalence
We evaluated the performance of real-time monoplex PCR for detection of S. stercoralis. We compared the PCR assay with the results of coprological diagnostic methods (Baermann and Koga Agar plate culture). We estimated an overall sensitivity and specificity of 61.0% and 92.7%, respectively. The cycle-treshold (ct) values indicate that PCR most often failed to detect infection in low-intensity infections. We conclude that it is a suitable option for individual diagnosis, but is difficult to apply in field settings.
We performed a molecular analysis of single S. stercoralis worms. We sequenced 269 individual worms collected from 29 individuals. We could identify three different genotypes of Strongyloides in our study population, co-existing sympatrically within the same host. Yet, we were not able to detect heterozygous worms, which would indicate that there is no interbreeding.
With our large-scale studies conducted in Cambodia, we could demonstrate the presence of high infection rates of S. stercoralis, never been reported at that high level in Cambodia. It shows that applying good sensitivity diagnostic methods on consecutive days gives prevalence rates that are a closer to the true prevalence and considerably higher than what has been previously reported. We conclude that this is true for most S. stercoralis endemic settings and hence, the true prevalence is likely to be higher than the numbers reported in studies which apply low sensitivity diagnostic methods.