Article Text
Abstract
A high prevalence of Dientamoeba fragilis is reported in faecal samples collected from patients attending complementary medicine practitioners in the British Isles. Specimens were collected directly after passing into sodium acetate–acetic acid–formalin (SAF). During two observation periods in 2002–04 and 2005–07, a D fragilis prevalence of 14.6% (n = 543) and 16.9% (n = 421), respectively, was recorded. These results confirm a surprisingly high prevalence of D fragilis among a selected population. Clinical information was only available for half of the patients with D fragilis; 50% of requests with clinical information reported gastrointestinal symptoms. For further work on its pathogenic role and prevalence among patients with gastrointestinal symptoms, immediate collection in SAF should be considered the optimal sampling modality for UK based laboratories.
Statistics from Altmetric.com
Leeds Teaching Hospitals NHS Trust provides a diagnostic service for complementary medicine practitioners in the British Isles. This service was founded in the early 1990s to provide faecal sample analysis that meets the needs of complementary medicine practitioners. Faecal samples submitted to our laboratory for parasitological analysis have been collected directly into sodium acetate–acetic acid–formalin (SAF).
Dientamoeba fragilis was first described in 1918 and has been taxonomically placed in the phylum Parabasala, class Trichomonadeae, family Trichomonadideae pending further taxonomonic work.1 Cyst forms have not been described and the organism is not flagellated. The trophozoite form is the only form described for this organism to date; this has significant implications for its optimal detection in that diagnostic faecal samples should be placed into a suitable fixative directly after passing a bowel motion to allow for trichrome staining or trophozoite forms. Unpreserved faecal samples are not recommended for detection, although attempts can be made to fix or stain unpreserved faecal samples in the laboratory on receipt of the samples. Axenic culture has not yet been established, but dixenic culture has been used for susceptibility testing and culture.2 3
We intended to provide epidemiological data on D fragilis in a highly self-selected population of patients who had most likely had gastrointestinal symptoms for some time before consulting a complementary medicine practitioner. We assumed that our population would not reflect the normal population of patients submitting faecal samples to diagnostic microbiology laboratories. However, the fact that samples were collected into SAF almost instantly offered a unique opportunity to gain additional information on the prevalence of D fragilis in the UK.
MATERIALS AND METHODS
Sample containers for parasitological and microbiological diagnosis were sent to complementary health care practitioners using our laboratory; written instructions for their patients were provided. SAF was used as fixative in one of the containers. Samples were returned to our laboratory by mail. We used the Bio-Bottle transport system (Bio-packaging, Coventry, UK) which complies with United Nations regulations for transportation of diagnostic specimens.
Trichrome staining was performed on all samples. A formol ethyl-acetate concentration method was carried out on the unpreserved sample. Results were recorded and reported back to the complementary medicine practitioner.
We conducted our prevalence study over two periods from 2002 to 2004 and from 2005 to 2007. In addition to D fragilis, we also recorded the presence of other parasites (pathogenic and non-pathogenic) in all samples.
RESULTS
A total of 964 samples were analysed over both three-year audit periods (543 and 421 samples, respectively, per period); 429 samples contained parasites, and only 3 samples contained pathogenic intestinal parasites (Capillaria phillipinnensis, Trichuris trichiura and Giardia lamblia).
The remainder contained non-pathogenic intestinal protozoa in various combinations, with Blastocystis hominis being the most commonly found parasite in 341 samples (196 and 145 samples per audit period) (fig 1). D fragilis was present in 150 samples (79 and 71 per period). The overall prevalence rate for D fragilis was 15.6% (14.6% and 16.9% for each period, respectively). Blastocystis hominis positivity rate was 35.4% for both audit periods (36.1% and 34.4% for each period, respectively).
We did not make any attempt to ascertain more clinical information than given on the received request forms. For over half of all received samples no history had been given, thus rendering this source of information unhelpful. Approximately half of those request forms with clinical information (25% of all requests) stated intestinal symptoms, ranging from acute gastroenteritis to chronic intestinal symptoms. The other half did not describe any intestinal symptoms.
DISCUSSION
Our prevalence rate of D fragilis in faecal samples compares well with previously reported prevalence ranges. Talis et al, in a 10-year study from Israel, used a culture based technique to determine the prevalence of D fragilis; 30 609 (15.2%) of 201 750 samples were positive.3 The range for microscopic detection was 0–52%.1 Most papers report a prevalence rate of 8–15%. In 1999, Windsor found a D fragilis positivity rate of 5% in stools in the Sultanate of Oman.4
However, this study showed a surprisingly high prevalence compared with a recent study by Windsor et al that reported a rate of 1.3% by microscopic and 2.6% by culture detection among 976 routine stool samples.5 These samples were not, however, collected in SAF. An Australian study of an unselected population found a prevalence rate of 0.9% among 6750 patients who had collected samples directly in SAF. On the other hand, a recent report from Denmark found a surprisingly high D fragilis prevalence of 11.7% among a metropolitan population.6 This may suggest that there may be considerable local epidemiological differences.
Our study did not intend to analyse clinical symptoms of patients, but available clinical history was noted. Stark et al reported a close association with diarrhoea in patients with positive D fragilis microscopy.7 Restriction fragment length polymorphism typing of the D fragilis PCR amplicon was carried out on almost all isolates and appeared identical. This may suggest the prevalence of a virulent genotype in this particular population. Half of our obtained requests for D fragilis positive stool samples mentioned abdominal symptoms. The proportion of patients without gastrointestinal symptoms in our series is higher than reported in other studies.1 8 These studies reported findings for faecal samples in patients with diarrhoea, whereas complementary medicine practitioners often sent stool samples for reasons not related to suspected gastrointestinal infection. This may well point towards differences of virulence of the organism in the UK, and future work is required.
To aid such future work, improved diagnostic yield for diagnostic UK laboratories could be achieved by direct preservation of fresh stool samples in a suitable fixative such as SAF. This is the recommended methodology for microscopic detection of D fragilis. Various PCR detection methods have been described; genotyping was based restriction fragment length polymorphisms of PCR amplicons which would be suitable research methodologies.7–9
It is also noteworthy that over half of the D fragilis positive samples were found in combination with Blastocystis hominis as previously observed.7 10 11 The clinical significance of multiple intestinal protozoa remains unclear.
We did not follow up any therapeutic intervention carried out by the complementary medicine practitioners. There are a growing number of reports in support of symptom resolution on treatment.12 Eradication in response to treatment with secnidazole was seen in 97% of detected cases.13 Associations with acute gastroenteritis and chronic diarrhoea have been reported.14–17 Its pathogenic role in infectious gastroenteritis is, however, not established with certainty and many clinicians will not consider it as a true pathogen among immunocompetent patients. It has been recognised as causing gastroenteritis in the immunocompromised host.18
Possibly because of its uncertain pathogenicity, there are no UK treatment guidelines for D fragilis. Most laboratory and infectious diseases physicians will recommend treatment only in the immunocompromised patient. Initial therapy is metronidazole administration. Alternatives for failed initial therapy in the UK are doxycycline and paromomycin. There are other treatment options, such as iodoquinol for luminal intestinal therapy; this is not available in the UK.
This observational study of D fragilis prevalence among patients attending complementary medicine practitioners in the British Isles provides more information on the prevalence of this intestinal protozoon. It shows a higher than previously described prevalence for D fragilis in the UK in a selected patient population. Further studies are warranted on its prevalence and virulence in the UK, using stool samples directly preserved in SAF and molecular genotyping techniques.7 19
Take-home messages
Dientamoeba fragilis prevalence is higher among patients consulting complementary medicine physicians in the British Isles. The reasons for this are unclear.
Direct collection of stool samples into specimen collection containers containing sodium acetate–acetic acid–formalin (SAF) should be the best practice standard for collection to evaluate the presence of D fragilis.
Virulence and pathogenicity studies are necessary to assess the role of D fragilis as a causative agent for gastroenteritis in the UK.
REFERENCES
Footnotes
Competing interests: None.