Co-infection of HIV in patients with Buruli ulcer disease in Central Ghana

BACKGROUND: Previous studies have reported that presence and severity of Buruli ulcer (BU) may reflect the underlying immunosuppression in HIV infected individuals by causing increased incidence of multiple, larger and ulcerated lesions. We report cases of BU-HIV coinfection and the accompanying programmatic challenges encountered in central Ghana. METHODS: Patients with PCR confirmed BU in central Ghana who were HIV positive were identified and their BU01 forms were retrieved and reviewed in further detail. A combined 16S rRNA reverse transcriptase / IS2404 qPCR assay was used to assess the Mycobacterium ulcerans load. The characteristics of coinfected patients (BU+HIV+) were compared with a group of matched controls. RESULTS: The prevalence of HIV in this BU cohort was 2.4% (compared to national HIV prevalence of 1.7%). Eight of 9 BU+HIV+ patients had a single lesion and ulcers were the most common lesion type. The lesions presented were predominantly category II (5/9) followed by category I lesions. The median (IQR) time to healing was 14 (8-28) weeks in the BU+HIV+ compared to 28 (12-33) weeks in the control BU+HIV- group (p = 0.360). Only one BU+HIV+ developed a paradoxical reaction at week 16 but the lesion healed completely at week 20. The median bacterial load (16SrRNA) of BU+HIV+ patients was 750 copies /ml (95% CI 0-398,000) versus 500 copies/ml (95% CI 0-126,855,500) in BU+HIV- group. Similarly, the median count using the IS2404 assay was 500 copies/ml (95% CI 0-500) for BU+HIV+ patients versus 500 copies/ml (95% CI 500-31,000) for BU+HIV- patients. BU+HIV- patients mounted a significantly higher interferon-γ response compared to the BU+HIV+ co-infected patients with respective median (range) responses of [1687(81.11-4399) pg/ml] versus [137.5(4.436-1406) pg/ml, p = 0.03]. There were challenges with the integration of HIV and BU care in this cohort. CONCLUSION: The prevalence of HIV in the BU+ infected population was not significantly increased when compared to the prevalence of HIV in the general population. There was no clear relationship between BU lesion severity and HIV viral load or CD4 counts. Efforts should be made to encourage the integration of care of patients with BU-HIV coinfection