Daily cotrimoxazole better than three times weekly for infants with HIV

Use of co-trimoxazole preventive therapy (CPT) three times a week compared to daily use in infants was linked to more severe bacterial infections and longer hospital stays, while survival rates were similar, researchers from South Africa reported in the September 10th edition of AIDS.

So in resource-poor settings with high bacterial disease burden daily co-trimoxazole is preferable for HIV-infected children, they add.

Co-trimoxazole preventive therapy is cheap, widely available and effective against pneumonia, tuberculosis as well as other opportunistic and bacterial infections. In HIV-infected children co-trimoxazole has reduced morbidity and mortality.

The World Health Organization now recommends daily cotrimoxazole preventive treatment (CPT) in HIV-infected or exposed children in resource-poor settings to continue until at least five years of age.

United States guidelines suggest CPT given three times a week is as effective as once daily, but with fewer adverse events and increased tolerability. Studies in adults support this.

The increasing availability of antiretroviral treatment and the associated increased pill burden and potential for drug interaction or adverse events could make intermittent CPT a better choice than daily use in HIV-infected children.

Yet there are no studies on the best dosage for HIV-infected children, nor the efficacy or tolerability of different prophylactic regimens. Additionally little evidence exists about survival and morbidity among African infants getting CPT.

324 HIV-infected children at least eight weeks of age and attending the Red Cross War Memorial Children’s Hospital at the University of Cape Town or Tygerberg Hospital at Stellenbosch University  were randomised to get CPT either three times (165-51%) a week or once a day (159-49%).

The median age at enrolment was 23 months (IQR 9.5-48.6 months). About one third of the children were under 12 months of age. Median CD4 cell percentage was 20%. Malnutrition was common in both groups. Close to 90% (287) were Centers for Disease Control and Prevention clinical stage 2 or 3.

Mortality in both groups was similar. 24 out of 165 children (14%) compared to 29 out of 159 children (18%) in the CPT three times a week and once daily, respectively, hazard ratio: 0.75 (95% CI: 0.44-1.29).

However, infants under 12 months of age had a six-fold higher incidence of death (20 deaths per 100 child-years compared to 3.6 deaths per 100 child-years, incidence risk ratio (IRR): 5.91 (95% CI: 3.3-11.2) P<0.0001. The authors stress this supports recent recommendations of providing early antiretroviral treatment and CPT in HIV-infected infants.

Infant death rates in those under 12 months of age were similar in  the CPT three times a week compared to the once daily group, IRR: 0.84, (95% CI: 0.41-1.73).

The primary causes of death in both groups were sepsis, pneumonia or diarrhoea.

However, those receiving CPT three times a week compared to the once-daily group had an over two-fold increased risk of bacterial infections, IRR 2.6 (95% CI: 1.21-4.87), p=0.0006.

The authors note that in spite of increased infections in those receiving CPT three times a week, mortality was similar. Explained, they believe, by good access to care, timely hospitalisation, and prompt treatment with antibiotics.

However, they highlight that where access to care is poor there is an increased probability that bacterial infections are linked to more severe illness and death. So treatment regimens that prevent bacterial infections (once-daily CPT) are preferable.

About half of all the children had one or more hospital stays. The admission rates were similar for both groups.

Pneumonia was the most common diagnosis at hospitalisation followed by diarrhoea or sepsis. However those in the three-times a week group spent more days in hospital compared to those on daily therapy, 228.5 days per 100 child-years and 198.5 days per 100 child-years, respectively. IRR 1.15 (95% CI: 1.04-1.28, P=0.004).

The authors note that the area where the study took place had high levels of co-trimoxazole resistance. Nonetheless, they note CPT was linked to a significant reduction in bacterial infections. This apparent discrepancy has been reported elsewhere. It is not known why this happens. Possible explanations include a difference in the sensitivity of bacteria in the laboratory compared to in humans.

The dosing schedule allocation was not blinded, a limitation of the study, noted the authors. However, they added that the main outcome of the study, mortality, was not subject to bias.

In addition, independent hospital clinicians not associated with the study, made decisions about admission, discharge as well as tests they note. So the higher rate of bacterial infection and longer hospital stays seen in children on three times a week CPT is because these children have more severe illness.

While the final sample size is smaller than intended (324 compared to 400), the authors note that the significant differences in bacterial infection and inhospital stays between the groups shows adequate power for such comparisons.

The authors note that the study was done in a setting that allowed for excellent follow-up and adherence. So while the results may not be generalisable to other settings, the case for daily CPT is strengthened, they add. Tolerance was excellent in both groups.

The authors suggest that either regimen may benefit children who have limited exposure to bacterial disease.  “However, as the predominant burden of paediatric HIV is in resource-limited, high bacterial burden countries, our results remain globally relevant.” They add that in particular in subtropical Africa where CPT may further reduce illness and death by “preventing nontyphoid Salmonella or malaria infection.”

They conclude their findings support WHO recommendations for daily CPT in HIV-infected infants and children as an effective, well-tolerated, widely available and cost-effective intervention for reducing morbidity and mortality among this population.