• Dientamoeba fragilis;
  • infections;
  • children


  1. Top of page
  2. Abstract
  3. References

A retrospective study of 87 patients diagnosed with the protozoan Dientamoeba fragilis was performed due to a recent increase in the number of patients diagnosed with this organism at the Unit of Clinical Parasitology, Huddinge University Hospital, Stockholm, Sweden. Medical records were reviewed. The highest incidence was found in pre-school boys, who also had the longest duration of symptoms, with a range of 1–630 weeks. A majority of the patients had symptoms of diarrhea, abdominal pain and flatus. The diarrhea varied from watery to loose, blood being reported only sporadically. Most patients had traveled outside Europe and had no other parasites in their stools. This study indicates potential pathologic properties in D. fragilis, and prospective studies are recommended.

Dientamoeba fragilis was discovered in 1909, and reported as a new species by Jepps and Dobell in 1918 [1]. The protozoan has a cosmopolitan distribution [2] and has been found in humans and monkeys [3]. It is now classified as a flagellate, as it shows ultrastructural and antigenic resemblance to Trichomonas and Histomonas[4]. It infects the whole colon from cecum to rectum. This protozoan has no cyst form but can be readily detected as a trophozoite. The mode of transmission is so far unknown, although fecal–oral transmission or the existence of a vector, possibly Enterobius vermicularis (pin worm), have been proposed [5,6]. It has been debated whether D. fragilis should be considered as a pathogen, despite recent studies supporting its pathogenic role [7]. Owing to an increase in the numbers of patients diagnosed with this protozoan at the Unit of Clinical Parasitology, Huddinge University Hospital, Stockholm, Sweden since 1996, an analysis of the patients was performed. The purpose was to describe symptoms, clinical findings, treatment and outcome in the infected patients.

Eighty-seven patients diagnosed with D. fragilis at the Unit of Clinical Parasitology Laboratory, Huddinge University Hospital, Stockholm, Sweden during the year 2000 were followed up. Clinical and laboratory findings were recorded. The stool samples (usually one per patient) were sent in to the laboratory sodium acetate–acetic acid–formalin solution. The modified Ritchie fecal concentration technique was used. The laboratory identification of D. fragilis was made by light microscopy and with staining (Trichrome; prepared by the laboratory) as the next step [8]. All parasites detected in these samples were reported, including Cryptosporidium and Cyclospora. Bacterial infections detected by stool culture were registered, but examination for viruses was not performed.

Forty-four of the subjects were female and 43 were male, their ages ranging from 1 to 77 years. Nineteen of the 43 males were less than 10 years of age, while there was little variation in incidence between the age groups among the females. Nine females were less than 10 years of age (Figure 1, Table 1).


Figure 1. Prevalence of D. fragilis in different age groups.

Download figure to PowerPoint

Table 1.  Bacteria and parasites found in 21 patients with D. fragilis infection
 Aeromonas caviae1
 Blastocystis hominis9
 Entamoeba coli7
 Giardia intestinalis4
 Enterobius vermicularis2
 Trichuris trichiuria2
 Entamoeba hartmanni1
 Endolimax nana1

Sixty-three per cent of the patients were likely to have been infected with D. fragilis outside Sweden. Forty-two per cent were reportedly infected outside Europe. Of the 23% infected within Europe, 8% were certainly infected in Sweden. Most patients were infected in Africa, South America and the Middle East.

Twenty-four per cent of the patients had findings of other parasites or bacteria in their stool samples. Giardia intestinalis and Entamoeba coli were the most frequently found parasites (Table 1). Two patients had Enterobius vermicularis in the stool samples.

Thirty of the 87 patients had results of blood tests measuring white blood cells, hemoglobin and C-reactive protein performed in association with the D. fragilis infection. None of the tested patients had any pathologic values in these tests.

The most frequent symptoms were abdominal pain, diarrhea, flatus, fatigue and overall looseness of stools (Table 2). Nausea, weight loss and vomiting were also found. Two patients were asymptomatic, and in six patients clinical symptoms were not recorded. The asymptomatic patients had stool samples taken as part of a routine medical examination. Five of the six patients with no symptoms were individuals seeking asylum for personal or political reasons.

Table 2.  Symptoms in 79 patients with D. fragilis infection
  • a

    As perceived by patients.

Abdominal pain33
Loose stools27
Sensitivity to lactosea19
Anal itching8
Reduced control over bowel6
Weight loss4
Mucus or blood in stool4

Nineteen patients were reported to suffer from intolerance of lactose. One boy reported feeling worse after eating his yoghurt breakfast and other lactose products. However, none of the patients with similar complaints improved after excluding lactose from their diets. Sixteen patients, two of whom were women, were subjected to clinical lactose intolerance tests. Two males were diagnosed with lactose intolerance.

Patients who were symptomatic had a duration of symptoms lasting between 1 and 630 weeks. The most important factors influencing the duration appeared to be sex and age. The duration was by far the longest in males, particularly in those under 10 years of age. A small peak of increased duration is found in males around the age of 40 years. The longest duration in women was found in patients infected during childhood, the childbearing years or after the menopause. In all age groups, males had a longer duration of symptoms.

Thirty-two patients infected with only the protozoan were treated with metronidazole. The drug was given at various dosages and for various lengths of time, but no correlation was found between dosage or length of treatment and clinical success. Four patients recovered after the metronidazole treatment, while 12 patients had recurring or reduced symptoms.

The present study shows that a cohort of patients diagnosed with D. fragilis suffered from clinical symptoms. These patients were in many cases puzzling to the clinicians, especially within the field of pediatric medicine. To date, the pathogenicity of this protozoan remains unconfirmed, despite recent studies seeming to indicate that D. fragilis has pathogenic properties [7,9–11]. Also, genetic diversity has been described in D. fragilis, implying the possible existence of pathogenic and non-pathogenic variants [12]. The absence of other pathogenic microorganisms in the stools of most patients in the present study supports the pathogenicity of D. fragilis. D. fragilis is known to be a cosmopolitan parasite [7,10]. In this study, the majority of the patients were abroad at the time of infection.

The mode of transmission of D. fragilis is still under debate. Theories concerning fecal–oral as well as vector-borne transmission have been presented [9,11]. The finding that a majority of patients had no other microorganisms isolated in their stools suggests that these patients had not been in close contact with water of poor hygienic standard. It could also strengthen the theory of a vector being involved in the transmission of D. fragilis. The high frequency of anal itching might be a symptom of infection with Enterobius vermicularis. However, the frequency of coexistence does not surpass that of D. fragilis with Blastocystis hominis. As pointed out by Preiss et al. [7], infection with Enterobius vermicularis can be resolved independently of D. fragilis infection.

The peak incidence was found in pre-school boys, and this finding is in contrast to previous studies [9], where the highest prevalence was found in young girls. The reason might be the behavior of this age and sex cohort. Whether the size or frequency of the infectious dose is of principal importance is unclear. It was not possible in this study to verify whether the children attended daycare centers, but our assumption is that the majority did (as this is the norm in Sweden). Close contact and crowding could be of importance for transmission, as mentioned by Windsor et al. [10]. The prolonged duration of symptoms in boys could again be dependent on the size of the dose, or the frequency of reinfection. Many patients reported having symptomatic family members, and the range of duration was significant (1–630 weeks).

In accordance with previous reports [7,9,10,13], the most commonly found symptoms were abdominal pain, loose stools or diarrhea, flatulence, anorexia, and fatigue. As these are well documented, it is of importance that fecal sampling becomes part of the routine investigation of this symptomatology.

The high number of patients tested for lactose intolerance may be due to frequent misinterpretation of the symptoms caused by D. fragilis. Though many patients reported feeling worse after all ingestion, both doctors and patients often suspected lactose intolerance, and tests for lactose intolerance were performed before parasitic infection was excluded.

In this study, the clinical effect of metronidazole was moderate, while Preiss et al. [7] reported a 70% success rate of 10-day metronidazole treatment. When this treatment was followed by oxytetracycline, the success rate rose to 90%.

Indications of doubtful effects of treatment, and of severe long-term health effects, emphasize the necessity for more thorough prospective studies of D. fragilis.


  1. Top of page
  2. Abstract
  3. References
  • 1
    Jepps MW, Dobell C. Dientamoeba fragilis ng n sp, a new intestinal amoeba from man. Parasitology 1918; 35267.
  • 2
    Garcia LS, Bruckner DA. Diagnostic medical parasitology, 2nd edn. Washington, DC: American Society for Microbiology, 1993: 3144.
  • 3
    Farthing MJG, Levallos A, Kelly P. Gut protozoa. In: CookGC, ed. Manson's tropical diseases, 20th edn. London: WB Saunders, 1996: 1270.
  • 4
    Camp RR, Maten CF, Honigberg BM. Study of Dientamoeba fragilis Jepps & Dobell. I. Electronmicroscopic observations of the binucleate stages. II. Taxonomic position and revision of the genus. J Protozool 1974; 2: 6982.
  • 5
    Burrows RB, Swerdlow MA. Enterobius vermicularis as a probable vector of Dientamoeba fragilis. Am J Trop Med Hyg 1956; 5: 25865.
  • 6
    Ockert G, Schmidt T. On the epidemiology of Dientamoeba fragilis Jepps and Dobell 1918. 4th communication: evidence of Dientamoeba fragilis in Enterobius eggs using isoelectric point determination. J Hyg Epidemiol Microbiol Immunol 1976; 20: 7681.
  • 7
    Preiss U, Ockert G, Broemme S, Otto A. On the clinical importance of Dientamoeba fragilis infections in childhood. J Hyg Epidemiol Microbiol Immunol 1991; 35: 2734.
  • 8
    Wheatley WB. A rapid staining procedure for intestinal amoebae and flagellates. Am J Clin Pathol 1951; 21: 9901.
  • 9
    Spencer MJ, Millet VE, Garcia LS, Rhee L, Masterson L. Parasitic infections in a pediatric population. Pediatr Infect Dis 1983; 2: 1103.
  • 10
    Windsor JJ, Rafay AM, Shenoy AK, Johnson EH. Incidence of Dientamoeba fragilis in faecal samples, submitted for routine microbiological analysis. Br J Biomed Sci 1998; 55: 1725.
  • 11
    Windsor JJ, Johnson EH. Dientamoeba fragilis: the unflagellated flagellate. Br J Biomed Sci 1999; 56: 293306.
  • 12
    Johnson JA, Clark CG. Cryptic genetic diversity in Dientamoeba fragilis. J Clin Microbiol 2000; 12: 46534.
  • 13
    Ayadi A, Bahri I. Dientamoeba fragilis: pathogenic flagellate? Bull Soc Pathol Exot 1999; 2: 299301.