The epidemiological impact of HIV antiretroviral therapy on malaria in children

Abstract

Objective: The objective of this study is to determine the epidemiological effectiveness of a first-line antiretroviral regimen with HIV protease inhibitor for preventing recurrent malaria in children under the range of HIV prevalence levels and malaria transmission intensities encountered in sub-Saharan Africa.

Design: A dynamic model of malaria transmission was developed using clinical data on the protease inhibitor extended posttreatment prophylactic effect of the antimalarial treatment, artemether-lumefantrine, in addition to parameter estimates from the literature.

Methods: To evaluate the benefits of HIV protease inhibitors on the health burden of recurrent malaria among children, we constructed a dynamic model of malaria transmission to both HIV-positive and HIV-negative children, parameterized by data from a recent clinical trial. The model was then evaluated under varying malaria transmission and HIV prevalence settings to determine the health benefits of HIV protease inhibitors in the context of artemether-lumefantrine treatment of malaria in children.

Results: Comparing scenarios of low, intermediate and high newborn HIV prevalence, in a range of malaria transmission settings, our dynamic model predicts that artemether-lumefantrine with HIV protease inhibitor based regimens prevents 0.03-0.10, 5.2-13.0 and 25.5-65.8 annual incidences of malaria per 1000 children, respectively. In addition, HIV protease inhibitors save 0.002-0.006, 0.22-0.8, 1.04-4.3 disability-adjusted life-years per 1000 children annually. Considering only HIV-infected children, HIV protease inhibitors avert between 278 and 1043 annual incidences of malaria per 1000 children.

Conclusion: The use of HIV protease inhibitor based regimens as first-line antiretroviral therapy for HIV is an effective measure for reducing recurrent malaria among HIV-infected children in areas where HIV and malaria are coendemic, and artemether-lumefantrine is a first-line antimalarial.

Fig. 1. Annual incidences averted per 1000 people vs. newborn HIV prevalence and treatment

EIR: Top 10 ibpppy, middle 112 ibpppy and bottom 562 ibpppy. Incidence averted is for all children, HIV-infected and HIV-uninfected.

Fig. 2. Disability-adjusted life-years saved per 1000 people/year vs. entomological inoculation rate and treatment

The newborn HIV prevalence is top 1.3% and bottom 6.7%. DALYs saved is for all children, HIV-infected and HIV-uninfected.

Fig. 3. Partial rank correlation coefficients, where (1/δ_PI) is the duration of the extended posttreatment prophylactic effect, (ω) is the proportion reporting for treatment to a healthcare setting, (β_H) is the human to mosquito transmission probability, (βH⁻) is the mosquito to human (HIV-negative) transmission probability, (1/γ) is the mean duration of malaria infection recovery, (τ_M) is the extrinsic incubation rate and (d₀) is the mean old adult mosquito life span

Partial rank correlation coefficients (PRCCs). EIRs for top, middle, and bottom panel are 10, 112, and 562 ibppy, respectively. Parameters were considered to be important in reducing incidence of malaria if their |PRCC| > 0.4. Significance tests assessed if PRCCs were significantly different from zero (see Table S2, http://links.lww.com/QAD/A618).