Cutaneous and Presumed Visceral Leishmaniasis in a Soldier Deployed to Afghanistan

James P. Woodrow, MD; Joshua D. Hartzell, MD; Jim Czarnik, MD; David M. Brett-Major, MD; Glenn Wortmann, MD

Disclosures

November 30, 2006

Discussion

Both cutaneous and visceral leishmaniasis have been reported in troops returning from Operations Iraqi and Enduring Freedom,[1,2] but the current case represents the first case of leishmaniasis with concurrent manifestations of CL and presumed VL in our cohort. The simultaneous presentation of CL and VL is rare in the presence of an intact immune system. Manson-Bahr[5] reported in 1955 that some immunocompetant patients with CL may eventually develop VL through lymphatic spread; however, this theory has not been proven over time.[6] A more recent report of VL occurring 4 months following CL has been described in a child.[6] The parasites in this case were not typed, so it is possible that the patient was infected by different zymodemes. Another report from Israel described L donovani complex being isolated from both skin and bone marrow biopsies of a 41-year-old with simultaneous CL/VL manifestations.[7] In a Brazilian study of 18 patients with VL, 7 patients yielded skin cultures positive for L chagasi.[8] In that study, only 2 patients had cutaneous lesions, while 16 cases were from biopsies of clinically normal skin. Simultaneous cutaneous and visceral disease is a known phenomenon in the HIV patient population,[9,10,11] and it often is an opportunist of concomitant cutaneous morbidity such as Kaposi's sarcoma. It is reported that cutaneous/visceral overlap syndromes are seen in 2%-12% of HIV-positive cases.[12]

VL is not generally associated with cutaneous ulcers, and those affected do not usually recall any insect bite or lesion at the site of inoculation. However, the interplay of cutaneous and visceral disease is further clouded by the recognition of post kala-azar dermal leishmaniasis (PKDL). PKDL is an entity associated both with L donovani and infantum, and is manifest by diffuse skin lesions that typically develop after therapy for VL.

L donovani complex, which includes L infantum, L chagasi, and L donovani, is responsible for most cases of VL.[13] Though the majority of cases are asymptomatic, a fulminant, often fatal progression of disease may develop. Known in endemic populations as kala-azar (Hindi for "the black sickness"), this can present with severe fever, fatigue, and cachexia following an incubation period ranging from weeks to months.[14] Organisms are spread via the reticuloendothelial system and may be found in the spleen, liver, and bone marrow. Clinical features include hepatosplenomegaly, pancytopenia, and elevated liver-associated enzymes. Gastrointestinal involvement, including diarrhea, is a known manifestation of VL regardless of HIV status.[12] However, gastrointestinal symptoms are more common among children and HIV-positive individuals.[12,15,16]

Classically, VL is diagnosed by demonstrating the presence of Leishmania organisms in visceral tissue. The bone marrow, liver, and spleen offer the highest-yield biopsy sites.[17] Proving Leishmania infection is accomplished by direct visualization under light microscopy, tissue culture, animal inoculation, detection of DNA via PCR, or detection of parasite-specific antigens.[18] Invasive procedures confirm the diagnosis, but carry the risk of complications. Technical expertise presumably decreases the risk of complications, and 1 center has reported only 2 cases of fatal hemorrhage in a total of 9612 splenic biopsies performed over 10 years.[18]

Recently, a noninvasive immunochromatographic test strip for VL has been marketed in the United States. The test uses a recombinant k39 antigen cloned from L chagasi, which is immobilized on a nitrocellulose membrane, and will produce a colored band when exposed to sera with target antibodies.[18] This test is inexpensive, accurate, and can be performed in the field by inexperienced personnel. A specificity ranging from 93% to 99% and a sensitivity ranging from 89% to 97% have been confirmed in both retrospective[19] and prospective studies[20,21,22,23] using splenic and bone marrow biopsies as a gold standard. Furthermore, the test has been shown to not be falsely positive in patients with CL, although this finding was based on a study of patients with CL caused by L braziliensis and may not be applicable to patients infected with L donovani complex.[24] It is also important to mention that these studies used Leishmania-endemic populations, and so the test characteristics may not apply when used on US military personnel.

In the case described in the current report, the initial diagnosis of VL was made using the rK39 dipstick. The Leishmania parasite was not isolated by subsequent bone marrow or colorectal biopsy. The sensitivity of bone marrow biopsy in detecting VL has been reported at 60% to 85% in 1 study[17] and 55% to 97% in another.[15] Given a positive rK39 assay in the setting of strong clinical suspicion and negative results on other diagnostic assays, the patient was treated empirically for VL. The patient's rapid clinical and laboratory responses to Ambisome were reassuring and supportive of the diagnosis.

There are 3 alternative explanations for this patient's unusual presentation. First, it is possible that our patient was infected with 2 different zymodemes of L donovani complex species, which would explain his seemingly contradictory clinical manifestations (cutaneous and visceral).[25,26] An analysis of L infantum zymodemes isolated from patients with cutaneous disease suggested a correlation between isotype and clinical pattern of disease (visceral vs cutaneous).[25] Another study reported treatment failure in a patient with a mixed infection with the parasite L infantum.[26] The authors suggested that the mixed infections with different strains may explain the differences in clinical course experienced by patients.[26] Our patient's geographic exposure would be supportive of this supposition. L donovani is a more common cause of VL in East Africa and the Indian subcontinent, while L infantum is more commonly found in Southern Europe, North Africa, and the Mediterranean Basin.[27,28] Our patient was deployed along the Afghan/Pakistani border, which would potentially put him at risk for infection with either, or both, organisms.

Second, it is possible that the patient's clinical manifestations were a result of infection with Plasmodium vivax. Multiple blood smears for malaria parasites were examined during the hospital course, all of which were negative. In addition, the patient's normal platelet count during hospitalization argues against active malaria (the patient was thrombocytopenic when he later presented with documented malaria infection). Finally, the patient's clinical response to Ambisome (resolution of fever and laboratory abnormalities) suggests that his initial febrile episode was due to VL. Plasmodicidal properties of amphotericin have been reported,[29,30] though this has not been investigated in vivo. Therefore, we believe it is more likely that the presence of latent infection with P vivax was a confounding variable in this case. Both leishmaniasis and P vivax malaria are endemic to Afghanistan. Co-infection has been reported, as have suggestions of enhanced virulence by Leishmania in the setting of malaria.[31,32,33,34,35]

A final alternative explanation for this patient's clinical course would be an underlying, undiagnosed, immunosuppressive medical condition. Simultaneous presentation of cutaneous and visceral manifestations, as mentioned previously, is not uncommon in immunosuppressed patient populations (specifically patients with late-stage HIV infection). Our patient had negative HIV serologic tests performed during the initial evaluation and 6 months later. He was not taking any immunosuppressive medications, and there was no clinical/laboratory/radiologic suggestion of malignancy, connective tissue disease, or primary immunodeficiency. The patient had no prior history of unusual or recurrent infections, and remains well after treatment for presumed VL and malaria.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....