Visceral leishmaniosis (VL) is widely reported in Ethiopia, with about 30% of cases being associated with human immunodeficiency virus (HIV). In absence of antiretroviral treatment (ART), poor prognosis, high mortality and high relapse rates are characteristic of Ethiopian VL patients with HIV co-infection. Conversely, co-infection can be successfully managed via a combination of effective treatment of the initial episode, timely ART and prevention of relapses.

Actually, until cellular immunity returns with ART, the patient is at risk of VL relapses, which can result in death, severe illness, reduced ART efficacy, drug-resistance and possibly transmission of drug-resistant Leishmania donovani. Patients most vulnerable to relapses are those with high levels of immunosuppression, with previous VL episodes, or with opportunistic infections (OIs). The most important factor to prevent relapses seems to be the clearance of visible parasites.

Limited studies in Europe show that HIV co-infected patients may benefit from secondary prevention with antimonials (part of mainstay treatment for VL in Ethiopia) and pentamidine (PM), not used for VL treatment in Africa. Such maintenance treatment has not been studied in African VL, but the poor outcomes without secondary prevention highlight a need of better care to patients at risk of relapse.

This prospective cohort study aims at documenting the patient's outcomes of secondary prophylaxis with PM in VL-HIV co-infection, in terms of time to relapse or death, safety and feasibility, before it can be considered for general use in Ethiopia. A placebo group is not included, due to the clear advantages of the intervention to the patient population.

Visceral leishmaniosis (VL) in Ethiopia has been reported in different parts of the country, with approximately 30% of cases being associated with human immunodeficiency virus (HIV). The ruralisation of HIV epidemic in VL endemic areas will hamper efforts to control VL. Clinical experience in Ethiopia has shown that anti-leishmanial treatment in the absence of anti-retroviral therapy (ART) does not result in favourable outcomes: poor prognosis, high mortality and relapse rates are characteristic of Ethiopian VL patients with HIV co-infection. The effective management of the initial VL episode, timely ART, and prevention of relapses should be the cornerstones of effective management of HIV/VL co-infection.

However, parasitological cure of VL in HIV co-infected patients cannot easily be established, and until cellular immunity returns with ART, the patient is at risk of relapses of VL, which can result in death, severe illness, negative effect on ART efficacy leading to other opportunistic infections (OIs), emergence of drug-resistant parasites, and possibly to transmission of drug-resistant Leishmania donovani. Patients most vulnerable to relapses are 1) those with high levels of immunosuppression, 2) patients with previous VL episodes, and 3) patients with OIs.

ART reduces the risk of VL relapse/recurrence by ~50%, while the type of anti-leishmanial primary treatment has little effect on relapses; the most important factor seems to be clearance of visible parasites (if residual parasites are seen at the end of treatment, the relapse rate is 100%).

Limited studies in Europe show that HIV co-infected patients may benefit from secondary prevention, by significantly prolonging the relapse-free period. The drugs studied for secondary prophylaxis in Europe have been meglumine antimoniate and AmBisome, which are part of mainstay treatment for VL in Ethiopia, and pentamidine (PM), which is not used for VL treatment in Africa. The effect of such maintenance treatment has not been studied in African VL, but the poor outcomes without secondary prevention highlight a clear need to offer better care to patients at high risk of relapse.

Indeed, secondary prophylaxis is generally recommended in Europe and the United States (see the 2009 Center for Disease Control guidelines). PM 4 mg/kg intravenous (IV) every 3-4 weeks has been proposed as secondary prophylaxis, and it is already used in countries like United Kingdom and Spain.

Consequently, this prospective cohort study aims at documenting the patient's outcomes of secondary prophylaxis with PM in VL-HIV co-infection, in terms of time to relapse or death, safety and feasibility, before it can be considered for more general use in Ethiopia. A placebo group is not included, due to the clear advantages of the intervention to the patient population targeted herewith. Furthermore as other available VL treatments are used as main line treatments, they cannot be considered as alternative comparators, given the potential risk of rapid emergence of drug resistance and subsequent spread in areas of anthroponotic VL.

Inclusion Criteria:

• Patients diagnosed with Visceral Leishmaniosis (VL) during the recruitment period that are EITHER treated for VL relapse and have a documented negative test of cure (TOC), OR are treated for primary VL and have a documented CD4 <200 or WHO stage 4 disease during the recruitment period and have a documented negative TOC
• Patients treated for VL in the past with documented CD4 <200 or WHO stage 4 disease during the recruitment period AND documented negative TOC after the latest VL treatment and currently asymptomatic OR currently negative diagnostic test (microscopy)
• Patients agreeing to start or continue antiretroviral treatment (first or second line)
• Patients willing to provide written informed consent

Exclusion Criteria:

• Patients with known hypersensitivity to pentamidine
• Patients with known renal failure
• Patients with diabetes mellitus (type I or II)
• Patients unlikely to attend follow-up visits/comply with study requirements
• Pregnant and lactating women
• Any other condition that could increase the risk of toxicity of pentamidine to such an extent outweighing the expected benefit (eg severe cardiac dysfunction).

Adverse event data were collected over the whole study period of 30 months (primary study period (12 months), extended study period (6 months) and 12 months follow-up period).