HIV seroconversion complicated by Mycobacterium kansasii infection

Abstract

HIV seroconversion can be associated with opportunistic infections when the CD4 cell count is low. Mycobacterium kansasii infection has been described in established HIV infection, often advanced, but not previously in seroconversion. We present a patient in whom seroconversion to HIV was complicated by M. kansasii infection. A 30-year-old Zambian man presented to our department in February 2004 with fevers, sweats, weight loss, and shortness of breath. A chest X-ray showed a right pleural effusion that was found to be an exudate. No organism was isolated on pleural biopsy. Given his symptoms and background, he was treated empirically for pleural tuberculosis. Testing for HIV antibodies at this time was negative. He responded well to 6 months of standard antituberculosis treatment, and was discharged from follow-up in July 2004. He re-presented to us at the end of April 2006 with a flu-like illness. He had a marked sore throat and also complained of fevers, diarrhoea, rash and a cough productive of sputum. On examination he had a macular rash over his torso, oral macular lesions and oral thrush. He was also noted to have lymphadenopathy in both groins and crackles at the right lung base. His chest X-ray was normal but he continued to have a cough productive of sputum. On direct questioning it transpired that his wife had just returned from a long period in Zambia, the patient denying any new sexual contacts. Initial blood tests showed a total lymphocyte count of 0.42 × 109 cells/l. His initial HIV serology was equivocal, with a reference laboratory report showing the absence of HIV antibodies, but the presence of p24 antigen. He subsequently became HIV antibody positive 6 days later. His HIV viral load was 694 843 copies/ml, with an initial CD4 T-cell count of 0.194 × 109 cells/l. Sputum microscopy at this time showed moderate numbers of acid fast bacilli. These samples were sent for testing by polymerase chain reaction for Mycobacterium tuberculosis. In the interim he was started on conventional antituberculosis treatment, with rifampicin, isoniazid, pyrazinamide and ethambutol. He made a good response to treatment, with resolution of his fevers and cough, his rash and diarrhoea having already settled. His polymerase chain reaction test, however, came back negative for M. tuberculosis. At this point we added clarithromycin to provide additional cover against atypical mycobacteria. His sputum subsequently grew M. kansasii sensitive to rifampicin, ethambutol and clarithromycin. At follow-up at 2 months he felt well with only a slight cough still present but no sputum production. He had no further fevers or sweats and his repeat CD4 T-cell count was 0.461 × 109 cells/l, with his viral load having reduced to 51 416 copies/ml. M. kansasii infection has been well reported in advanced HIV disease [1–6], but to the best of our knowledge this is the first reported case occurring in seroconversion. Given that our patient has been treated previously for tuberculosis, without microbiological confirmation, it is possible that he had M. kansasii infection originally, which became active again during seroconversion. The response to treatment, with resolution of his fevers and cough, led us to believe he was actively infected rather than merely colonized. He clearly had a seroconversion illness, as demonstrated by his previous negative HIV test and his high HIV viral load and low CD4 cell count at presentation followed by spontaneous improvement without antiretroviral therapy. Other opportunistic infections and complications of advanced HIV disease have previously been reported during seroconversion [7–9], and our case adds to this literature. Seroconversion is associated with a rapid temporary drop in the CD4 cell count and a reversal in the CD4/CD8 cell ratio, during which time opportunistic infections may occur [7,10]. M. kansasii infection in the context of HIV has been described in patients who tend to be moderately to severely immunosuppressed [1–4]. The introduction of antiretroviral therapy has reduced both the incidence and the mortality from M. kansasii infection [11]. It has also been noted that those with disseminated diseases do worse than patients only suffering from pulmonary disease [2,7]. As has been shown before, HIV patients respond well to standard treatment against M. kansasii[4]. Although the British Thoracic Society guidelines recommend lifelong treatment in HIV/M. kansasii co-infection [12], given that our patient has regained a good CD4 cell count, had local disease, and has responded well to treatment, we are intending to treat him for 9 months in the first instance (as is the practice for non-HIV-infected individuals) especially as his immune recovery has been marked. In summary, we have described a patient in whom an HIV seroconversion illness was complicated by M. kansasii infection. Clinicians should be aware that atypical mycobacteria can present during seroconversion and may require treatment.