A 24-year-old woman with a history of well controlled type I diabetes and hypothyroidism presented with a 2-month history of nonbloody diarrhoea and one stone of weight loss. She developed a transient widespread maculopapular rash and a painful knee. No history of abdominal pain or mouth ulcers was found and there was no family history of inflammatory bowel disease or coeliac disease. She had been on a holiday to Mexico 9 months previously.
The patient was admitted to hospital with worsening diarrhoea, dehydration, metabolic acidosis, electrolyte imbalance and had a single seizure. Her electrolytes were corrected, no further fits occurred and an MRI of her brain was normal.
The patient had a normocytic normochromic anaemia (Hb 10.1 g/dl) with significant hypoalbuminaemia (<20 g/l). Thyroid function, the remainder of her liver function tests and haematinics were normal. Coeliac serology (anti-tTG antibody) was negative and stool cultures were negative on multiple occasions including for Clostridium difficile toxin, ova, cysts and parasites. Although there were no signs of sepsis, her inflammatory markers remained mildly elevated (C-reactive protein >50 mg/l).
Initially, a colonoscopy showed mild patchy erythema throughout the colon with a few minor aphthous ulcerations. The terminal ileum seemed to have mucosal flattening. Colonic histology showed nonspecific focal inflammation but no granulomata. Terminal ileal biopsies were however normal. In view of the flattened ileal mucosa, an upper gastrointestinal endoscopy was performed. This also revealed macroscopically flattened duodenal villi, and distal duodenal biopsies showed loss of villous architecture and crypt hyperplasia with neutrophil infiltration. No increase in intraepithelial lymphocytes was observed. The findings were considered to be of an infective aetiology.
In view of this an exhaustive search for pathogens was commenced. No parasites were identified on stool microscopy following sodium acetate formaldehyde/ethyl acetate concentration. Yersinia or Histoplasma species were not isolated from stool or biopsy material. Serological tests for Toxoplasma gondii, cytomegalovirus and Epstein–Barr virus were negative, as was her HIV serology. Meanwhile, she was commenced on empirical antibiotic treatment with ciprofloxacin and metronidazole.
Abdominal computed tomography showed no splenomegaly or lymphadenopathy and a normal liver but suggested some distal small bowel thickening. A small, bowel follow-through was then performed (Fig. 1), which showed diffuse small bowel thickening and a ‘picket-fencing’ appearance of the mucosa suggestive of marked submucosal infiltration.
Endoscopies were repeated and colonic biopsies were this time normal, whereas ileal biopsies showed an atypical lymphoid infiltrate. Duodenal biopsies again showed villous atrophy but were not consistent with Whipples' or coeliac disease. Once again, a further extensive search for parasites in stool microscopy and culture were negative.
During the course of these investigations, her condition continued to deteriorate with continued watery diarrhoea, anorexia and weight loss, her BMI falling from 18 to 13 kg/m2. She developed a profound hypoalbuminaemia (<10 g/l) requiring total parenteral nutrition and her immunoglobulins were all low; IgA 0.08 g/l (reference range 0.8–3.5 g/l), IgM 0.35 g/l (0.45–2.5) and IgG 4.1 g/l (6.2–14). Her condition was complicated by the development of a left-sided above knee deep vein thrombosis requiring anticoagulation treatment.
The patient's small bowel histology was reviewed and in view of the doubt remaining over the diagnosis was referred for a further opinion. This reported subtotal villous atrophy and atypical lymphoid infiltrate within the lamina propria and focally within the submucosa. Morphologically and on immunophenotypic profile, this was felt to be consistent with an enteropathy associated T-cell lymphoma (EATL). The slides were therefore forwarded to two other tertiary centres for further histological opinion. The first of these confirmed the increased T-cell infiltration in the lamina propria but felt there was insufficient evidence on which to base a diagnosis of EATL. The second, found no evidence of lymphoma.
The patient was transferred to the Gastroenterology unit at Leeds General Infirmary, Leeds, UK, for further investigations. Push enteroscopy was performed, which demonstrated macroscopically abnormal proximal small bowel, and multiple further biopsies were taken. These biopsies proved to be negative for EATL. A bone marrow aspiration was also performed showing only reactive changes with a poor representation of B cells and only very occasional plasma cells.
In view of her persistent symptoms computed tomography was repeated, which showed that she had developed marked colonic thickening (arrowed in Fig. 2). The clinical suspicion of an infective aetiology remained high and further sets of biopsies were sent for Leishmania, Histoplasma and mycology culture, and again these were negative. A further repeat flexible sigmoidoscopy showed that macroscopically the patient had now developed extensive linear ulceration in the rectosigmoid region (Fig. 3). Amoebic indirect fluorescent antibody test and Leishmaniasis direct agglutination test (DAT) and recombinant K39 tests were all negative. Rectal biopsies showed cells within crypts, which on Giemsa staining contained dark bodies suspicious of Leishmania amastigotes (cells arrowed in Fig. 4).
Almost 5 months after her initial presentation, a diagnosis of visceral Leishmaniasis was made from PCR of the rectal biopsies (Fig. 5). DNA extracts of her rectal biopsy were positive for old world Leishmaniasis using previously described primers . Species differentiation was achieved by Hha1 restricting amplicons of an IT1/IT2 nested PCR (not yet described). It must be noted that a series of different PCRs were used to report a Leishmania donovani complex only and further differentiation between L. donovani donovani/L. donovani chigasi/L. donovani infatum was not possible. The patient was treated with a course of intravenous liposomal amphotericin (Ambisome, San Dimas, California, USA) – dose 3 mg/kg days 1–5 and days 10, 21 – and her diarrhoea improved dramatically. Within 2 weeks she was maintaining her weight, total parenteral nutrition was discontinued and she was eventually discharged. At follow-up, 1 month later, she had gained further weight, which was independent of artificial nutritional support and her C-reactive protein and immunoglobulins had normalized.
Leishmaniasis is a vector-borne protozoan infection spread by the bite of the female sandfly (Phlebotomus and Lutzomyia species). Its clinical modes of presentation can be divided into three main types: Cutaneous Leishmaniasis which is often self-limiting, mucosal Leishmaniasis and visceral Leishmaniasis (VL). The mortality of untreated VL is 75–95%  and is endemic in 88 countries largely located in the tropics and subtropics .
VL (also known as kala-azar, Hindi for black sickness or fever) is a rare diagnosis in the UK but worldwide there are 500 000 cases diagnosed annually mainly in tropical and subtropical regions (Fig. 6) with 90% occurring in India, Bangladesh, Brazil, Nepal and Sudan [3,4]. L. donovani is the causative species in the Indian subcontinent and Africa (Old World) and L. donovani chagasi in the New World  and L. donovani infatum in the Mediterranean. In Mexico, there have been only a few cases of VL with no cases reported in the Yucatan Peninsula, a popular holiday region for foreign travellers .
In 20 years between 1985 and 2004, 39 patients were treated at the Hospital for Tropical Disease in London, representing 83% of the total reported cases in the UK (Fig. 7). Only one of those cases was from South America and 77% imported from the Mediterranean (Spain, Italy, Greece, Malta, Cyprus). Twenty of these cases were HIV negative and, interestingly all of those had positive DAT, rk39 and indirect fluorescent antibody tests .
VL has a widely variable incubation period, usually between 3 and 8 months, but periods of up to 34 months have been described . Although the majority of infections remain subclinical, common manifestations of the acute infection include fever, anorexia, weight loss and abdominal pain. Marked splenomegaly and hepatomegaly are common and anaemia, pancytopaenia and hypergammaglobulinaemia are all typical. The black skin pigmentation, after which it was named, is rare. The cause of death is often secondary infection. Disease manifestations vary widely between endemic regions because of variable host factors and local parasite species making diagnosis difficult and often delayed .
Cases of VL have also presented atypically, primarily involving the lungs, pleura, oral mucosa, larynx, oesophagus, stomach, small intestine, skin and bone marrow . VL confined to the gastrointestinal tract alone tends to occur in HIV coinfection and is highly unusual in immunocompetent individuals. A literature review revealed only four such case reports [9–11]. Although patients with VL may have symptoms of diarrhoea, the presence of enteropathy alone as the presenting feature is rare. Although the exact pathogenesis of the malabsorption is not clear, it is suggested that symptoms in enteropathic VL may be a combination of mechanical occlusion of the mucosa by parasites, bacterial overgrowth, partial villous atrophy, competition between host and parasite for nutrients, altered motility, bile salt deconjugation and lymphatic blockade [9,12]. Absence of organomegaly is rare in VL and only three other cases of VL without hepatosplenomegaly in immunocompetent adults have been reported . These three cases presented with pronounced fever and anaemia, whereas weight loss and diarrhoea were noticeably absent. A huge increase has been observed in the coexistence of VL in the HIV population in the Mediterranean thought to be as a result of reactivation of latent or subclinical disease and endemic population migration. In this patient population atypical visceral involvement, including the gastrointestinal tract, is more common, particularly in those with a CD4 count of less than 50 cells/mm3 [14,15].
Diagnosis in endemic regions relies on culture or direct visualization of the parasite at microscopy. It is often straightforward as parasites are plentiful in an enlarged spleen and liver. Smears from a splenic aspirate yield the highest diagnostic rate (as high as 98% sensitivity compared with <90% for other organs) but is associated with a small risk of intraabdominal haemorrhage. Bone marrow aspiration is often done but is painful and less sensitive .
Noninvasive testing to detect high titres of nonprotective antileishmanial antibody (e.g. with DAT) are detected during acute illness and for years afterwards. Serological assay for IgG antibody to rK39 (a recombinant leishmanial polypeptide) is also a useful and simple field test but its use is limited by cross-reactivity with other diseases and false-negative results in patients coinfected with HIV . In a recent meta-analysis, both DAT and rK39 tests had high sensitivities of 94.8 and 93.9%, with specificities of 85.9 and 90.6%, respectively . PCR also offers both excellent sensitivity (between 90 and 100%) and specificity (approaching 100%) and has the additional advantage of species identification [19,20].
Treatment options for VL are limited. Pentavalent antimonials are the first-line in developing countries but have large areas of resistance in India. Miltefosine is the first effective oral drug against VL and is well suited to widespread use in developing countries and resource-poor areas . Amphotericin B is effective in antimonial resistant cases, although it is toxic and needs prolonged inpatient administration. Liposomal Amphotercin B (AmBisome Gilead Sciences, Inc., San Dimas, California, USA) is an expensive but highly effective and well tolerated treatment for VL and is now the first-line drug in the UK. Intravenous therapy is short in duration and results in rapid resolution of clinical symptoms .
VL is a rare condition, especially in immunocompetent patients where it can present insidiously and culminate in intestinal failure. Diagnosis can be difficult and in such cases repeated investigations are often necessary. Using multiple diagnostic tools may improve sensitivity. Our case describes a very unusual presentation of VL in which both microscopy and culture of rectal biopsies, microscopy of bone marrow aspirate and serological testing (DAT and rK39 antibodies) were all negative. Furthermore, there was absence of hepatosplenomegaly. To our knowledge, this is the first case described in the UK of a solely gastrointestinal VL in an immunocompetent patient, and the first to be diagnosed by PCR alone. We advocate a high index of suspicion in considering the diagnosis of VL in travellers returning from endemic regions presenting with refractory diarrhoea despite the absence of classic signs of the disease.
1. Noyes HA, Reyburn H, Bailey W, Smith D. A nested PCR-based schizodeme method for identifying Leishmania kinetoplast minicircle classes directly from clinical samples and its application to the study of the epidemiology of Leishmania Tropica in Pakistan. J Clin Microbiol 1998; 36:2877–2881.
2. Piscopo TV, Mallia Azzopardi C. Leishmaniasis. Postgrad Med J 2006; 82:649–657.
3. Herwaldt BL. Leishmaniasis. Lancet 1999; 354:1191–1199.
5. Guerin PJ, Olliaro P, Sundar S, Boelaert M, Croft SL, Desjeux P, et al. Visceral Leishmaniasis
: current status of control, diagnosis, and treatment, and a proposed research and development agenda. Lancet Infect Dis 2002; 2:494–501.
6. Martín-Sánchez J, Navarro-Mari JM, Pasquau-Liaño J, Salomón OD, Morillas-Márquez F. Visceral leishmaniasis
caused by Leishmania infantum
in a Spanish patient in Argentina: What is the origin of the infection? BMC Infect Dis 2004; 4:20.
8. Mandell GL, Bennett JE, Dolin R. Mandell, Douglas and Bennett's principles and practice of infectious diseases. 6th ed. Philadelphia, Pennsylvania: Elsevier Churchill Livingstone; 2005. pp. 2428–2442.
9. Baba CS, Makharia GK, Mathur P, Ray R, Gupta SD, Samantaray JC. Chronic diarrhoea
and malabsorption caused by Leishmania donavani
. Indian J Gastroenterol 2006; 25:309–310.
10. Alvarez-Nebreda ML, Alvarez-Fernández E, Rada S, Brañas F, Marañón E, Vidán MT, Serra-Rexach JA. Unusual duodenal presentation of leishmaniasis. J Clin Pathol 2005; 58:1321–1322.
11. Bel Haj Salah M, Mekni A, Khanfir M, Bellil K, Benhaha-Bellil S, Chelly I, et al. Unusual presentation of visceral leishmaniasis
in an immunocompetent patient. Med Mal Infect 2006; 36:167–169.
12. Muigai R, Gatie DG, Shaunak S, Wozniak A, Bryceson AD. Jejunal function and pathology in visceral leishmaniasis
. Lancet 1983; 2:476–479.
13. Parija SC, Karki P, Koirala S. Cases of kala-azar without any hepatosplenomegaly. Trop Doct 2000; 30:187–188.
14. Rosenthal E, Marty P, Del Giudice P, Pradier C, Ceppi C, Gastaut JA, et al. HIV and Leishmania coinfection: a review of 91 cases with focus on atypical locations of Leishmania. Clin Infect Dis 2000; 31:1093–1095.
15. Laguna F, García-Samaniego J, Soriano V, Valencia E, Redondo C, Alonso MJ, González-Lahoz JM. Gastrointestinal leishmaniasis in human immunodeficiency virus-infected patients: report of five cases and review. Clin Infect Dis 1994; 19:48–53.
16. Singh RK, Pandey HP, Sundar S. Visceral leishmaniasis
(kala-azar): challenges ahead. Indian J Med Res 2006; 123:331–344.
17. Sundar S, Sahu M, Mehta H, Gupta A, Kohli U, Rai M, et al. Noninvasive management of Indian visceral leishmaniasis
: clinical application of diagnosis by K39 antigen strip testing at a kala-azar referral unit. Clin Infect Dis 2002; 35:581–586.
18. Chappuis F, Rijal S, Soto A, Menten J, Boelaert M. A meta-analysis of the diagnostic performance of the direct agglutination test and rK39 dipstick for visceral leishmaniasis
. Br Med J 2006; 333:723–726.
19. Rodríguez N, Guzman B, Rodas A, Takiff H, Bloom BR, Convit J. Diagnosis of cutaneous leishmaniasis and species discrimination of parasites by PCR and hybridization. J Clin Microbiol 1994; 32:2246–2252.
20. Salotra P, Sreenivas G, Pogue GP, Lee N, Nakhasi HL, Ramesh V, Negi NS. Development of a species-specific PCR assay for detection of Leishmania donovani
in clinical samples from patients with kala-azar and post-kala-azar dermal leishmaniasis. J Clin Microbiol 2001; 39:849–854.
21. Sundar S, Rai M. Treatment of visceral leishmaniasis
. Expert Opin Pharmacother 2005; 6:2821–2829.
22. Sundar S, Agrawal G, Rai M, Makharia MK, Murray HW. Treatment of Indian visceral leishmaniasis
with single or daily infusions of low dose liposomal amphotericin B: randomised trial. Br Med J 2001; 323:419–422.