CROI: Use of both cotrimoxazole and insecticide-treated bednets dramatically lowers the risk of malaria in HIV-positive children

Providing HIV-infected children in Uganda with insecticide-treated bed nets (ITNs) and cotrimoxazole prophylaxis (CTX) dramatically reduces their risk of acquiring malaria, according to results from the CHAMP (Children with HIV and Malaria Project) in Kampala, presented last week at the Fourteenth Conference on Retroviruses and Opportunistic Infections in Los Angeles.

This prospective study looked at malaria incidence within a cohort of HIV-positive children who received cotrimoxazole (CTX) and ITNs compared to a concurrent cohort of healthy children who don’t receive these interventions. Only nine episodes of malaria were reported in the HIV-positive children with CTX administration and use of treated bednets. This represented a 97% reduction in incidence as compared to what occurred in the cohort of HIV-negative children (who were not given CTX and did not have access to ITNs until more than midway through the study) where 440 episodes of malaria were reported.

HIV and malaria overlap substantially in sub-Saharan Africa. HIV is associated with an increased malaria incidence and morbidity in adults and pregnant women (particularly in those with advanced disease). Much less however is known about HIV’s effect on malaria in children, the population most at risk of infection, with severe illness. Over a million children die of malaria each year, 75% of them in Africa.

“This lack of information [in children] is a critical gap in our understanding of the overlap between HIV and malaria,” said Dr Anne Gasasira who presented the findings of CHAMP.

Nevertheless, interventions have recently become available in Africa that could significantly impact the risk of malaria in children with HIV. For example, CTX has activity against the malaria parasite and the use of CTX prophylaxis has been associated with decreased risk of malaria in HIV-infected adults.

At the same time, CTX is closely related to other antifolate antimalarials, such as Fansidar, and the use of CTX against malaria could be complicated by drug resistance. There is a possibility that CTX may not work in areas where resistance to Fansidar is widespread, or conversely, that widespread CTX use could increase the spread of Fansidar resistance and limit its effectiveness where it is still used (currently, primarily as a prophylaxis against malaria and placental malaria by pregnant women).

ITNs have also been shown to markedly reduce the risk of acquiring malaria — but the combined use ITNs with CTX has not previously been studied.

At Mulago Hospital in Kampala, 861 children aged one to ten were enrolled into the CHAMP study between October 2005 and September 2006 (300 HIV-infected children and 561 HIV-negative children randomly selected from the local community).

At the start of the study, 88% the children with HIV were already using ITNs (and, as per policy, most if not all should have already been taking CTX as well) while none of the children in the community cohort were on CTX and only 6% reported use of bednets at the start of the study.

Both cohorts were on identical assessment schedules for malaria. Parasite prevalence was measured at enrolment and every three months. Thick blood smears were performed on each child who presented in clinic with a new episode of fever. Clinical malaria was defined as fever (without other cause) and evidence of malaria parasites by microscopy.

Perhaps unsurprisingly (because many of the HIV-negative children were already receiving the interventions under evaluation) there was already a difference in malaria incidence at baseline: 113 (20%) of the HIV-negative children had positive malaria blood smears at study enrolment versus none of the HIV-infected children. Parasitaemia is quite common in this setting and not always associated with clinical malaria (especially in older children), but any child who developed clinical cases (with fever) would have been treated at the hospital and cured. However, children in areas with a high burden of malaria can get infected repeatedly.

Over the course of follow-up, all the HIV-infected children were all taking CTX and those who were not using ITNS at enrolment were given them, and those who qualified for antiretroviral therapy (ART) received treatment according to WHO guidelines. However, no HIV-negative child received cotrimoxazole, though by May and June, 2006, all were given bednets for ethical reasons.

By the end of follow-up 519 HIV negative children and 290 HIV-positive children remained in the study. The gender distribution was similar in both groups and the age of the children with HIV was slightly younger (5.6 vs 6.5 years old (standard deviation of 2.6 years for both). In the HIV-positive children, the median CD4% was 21% (interquartile range 15-28%) and 35 (12%) were on ART.

CTX reduced the incidence of malaria by 35%, though this was not statistically significant, while the use of ITN alone significantly reduced the incidence by 45% (p<0.001). The risk of malaria for children in the HIV-infected cohort who received CTX and used insecticide-treated bednets was decreased by 97% (IRR 0.03, 0.01 to 0.11; p<0.001).

Only nine episodes of malaria were reported in this cohort (incidence= 0.07/person-year). Five of these episodes occurred in small subset of HIV-positive children who were not initially using bednets (for an incidence rate of 64.3 per 100 person years) (and these occurred over a short period of time) while only four cases occurred in those with bednets from the start for an incidence of 3.4 cases per 100 person years.

In the HIV-negative community group without cotrimoxazole (or bednets for much of the study), there were 440 episodes of malaria (incidence = 0.90 per person-year, p <0.0001). This resulted in an incidence rate ratio of 0.08 (0.04 to 0.15) between the two cohorts. 356 episodes occurred among those without any intervention for an incidence rate of 104.6 per 100 person years. 84 cases occurred in HIV-negative children using bednets for an incidence of 56.0 per 100 person years.

This study establishes the efficacy of CTX administration and the use of treated bednets in preventing malaria in the pediatric population with HIV.

This is highly significant given that the study took place in a malaria-endemic area with high-level antifolate (Fansidar) resistance. The study had also investigated the incidence of markers for drug resistance to antifolate antimalarial in the children who developed malaria. The dhfr/dhps quintuple mutation, which has been associated with resistance to Fansidar, was found in blood smear samples from both study groups: 100% prevalence in patients taking CTX and 80% in those who were not taking the drug).

One final and useful clinical observation was that malaria accounted for only 4% of all fever episodes in the HIV-infected cohort. Usually, 30-40% of fevers in this area are attributable to malaria (in fact, in the HIV-negative children in this study, 33% of the fevers were associated with malaria). In fact, fever in children in the area is usually treated empirically for malaria.