Portal de Periódicos da CAPES

Disseminated Mycobacterium ulcerans disease in an HIV-positive patient: a case study

2002; Lippincott Williams & Wilkins; Volume: 16; Issue: 12 Linguagem: Inglês

10.1097/00002030-200208160-00027

1473-5571

Roch Christian Johnson, David Ifebe, Aristote Hans-Moévi, Luc Kestens, Raoul Houessou, Augustin Guédénon, Wayne M. Meyers, Françoise Portaels,

Infectious Diseases and Mycology

Mycobacterium ulcerans disease, commonly called Buruli ulcer (BU), is a rapidly re-emerging disease in west Africa, particularly in Benin [1]. Severe disease is being reported with increasing frequency, with devastating complications [2,3]. Anecdotal reports suggest that HIV infection has little or no effect on the clinical course of BU, contrary to that seen with tuberculosis. Nevertheless, no long-term observations on the influence of HIV infection on patients with BU have been reported. Here we describe the evolution of BU in an HIV-positive patient seen at the Centre de Dépistage et de Traitement de Lalo, Benin, over nearly 2 years. This 27-year-old woman presented to the Centre de Dépistage et de Traitement de Lalo in October 1998 with a 10 cm diameter undermined ulcer on her right thigh. Indurated and hyperpigmented skin surrounded the ulcer. The lesion had appeared 3 months earlier as a painless swelling. After a clinical diagnosis of BU, the lesion was excised in November 1998, with revision in December 1998. Although the Ziehl–Neelsen stain for acid-fast bacilli and culture for M. ulcerans were negative, the polymerase chain reaction (PCR) was positive for M. ulcerans [4]. The wound was skin-grafted in January 1999, with additional grafting in February 1999. Two days after grafting, there was swelling at the site of an intravenous perfusion administered during surgery. This lesion ulcerated, and was excised in March 1999. Specimens for culture and PCR were positive for M. ulcerans. Rifampin (600 mg/day), ethambutol (800 mg/day) and clarithromycin (2 g/day) were given for 3 months. Radiological study of the swollen right leg in March 1999 revealed bone lesions requiring curettage. In May and June 1999, buccal candidiasis and intestinal salmonellosis were treated with nystatin and chloramphenicol. Initial serological HIV tests were positive at this time: Genscreen HIV-1 and −2 version 2 (Sanofi, Diagnostics Pasteur, Marnes la Coquette, France), with confirmation by Immunocomb II HIV-1 and -2 Bispot (PBS Orgenics, Strassbourg, France). A tuberculin skin test was negative, even though the patient had a Bacillus Calmette–Guérin scar. Other swellings and ulcers appeared on the left leg and wrist in June and July 1999. The PCR on specimens from the anatomical snuffbox area were M. ulcerans positive. From September 1999 to August 2000, multiple curettages of bone and excisions of skin were performed once or twice every month. The patient left the hospital against medical advice and died in September 2000, under the care of traditional healers. The cause of death is unknown, but she had widespread active BU when last observed. BU is primarily a disease of rural tropical wetlands. The aetiological agent proliferates in stagnant waters, and probable modes of transmission include contamination of the skin surface either directly or by insects such as waterbugs [5]. Trauma at contaminated skin surfaces frequently introduces the causative agent into the skin [6]. Various clinical forms may develop after M. ulcerans enters the skin and subcutis: localized necrotizing skin ulcers; contiguous dissemination over extensive body surfaces; and metastatic spread of M. ulcerans to distant sites, usually skin and bone. Minor forms heal spontaneously, ordinarily in 3–6 months by a delayed-type hypersensitivity response, followed by scarring. Devastating sequelae (scarring or amputations) often ensue from tissue destruction and osteomyelitis. Because of the geographical location of most BU disease foci in rural areas and the high frequency of BU in children, HIV infections are infrequent in at-risk individuals in endemic areas of BU. Few data are thus available on co-infections of HIV and BU. No prospective studies have been published, and none of the anecdotal reports suggests an effect of HIV infection on BU. Asiedu and Etuaful [7] in Ghana found three HIV-positive patients among 102 patients, but this correlation was insignificant as a risk factor for BU. Delaporte et al. [8] reported co-infection of HIV and BU, but did not believe that HIV affected the outcome of BU infection. Allen [9] drew similar conclusions from brief observations on three co-infected patients in the Democratic Republic of Congo. In Benin, five out of 108 BU patients were HIV positive, whereas none of 50 age-, sex- and origin-matched controls were HIV positive [10]. Although this suggests a correlation of HIV and BU, the difference is statistically insignificant. In our patient, the development and extent of dissemination of BU was unusually rapid and severe. The appearance of BU disease at an intravenous perfusion site 2 days after the perfusion is especially interesting. The shortest previously reported incubation period of BU lesions is 2 weeks [6]. Although the development of mycobacterioses by atypical mycobacteria is rare in HIV-positive patients in developing countries, we believe prospective studies are urgently needed on this issue for M. ulcerans [11]. R. Christian Johnsona David Ifebeb Aristote Hans-Moevia Luc Kestensc Raoul Houessoua Augustin Guédénond Wayne M. Meyerse Françoise Portaelsc Acknowledgements The authors would like to express their sincere thanks to the laboratory staff of the Lalo Center (Benin).