Combination of antibiotics as effective and safe as co-trimoxazole at preventing infections, but too expensive to use where needed?

A combination of the antibiotics atovaquone and azithromycin is at least as effective and just as safe as the antibiotic co-trimoxazole at preventing serious bacterial infections in HIV-positive children, according to a study published in the January 1^st edition of Clinical Infectious Diseases. However, an editorial accompanying the article questions just how clinically significant these findings are given that HAART has largely replaced the need for prophylactic antibiotics in richer countries, and that cost is likely to preclude the use of atovaquone-azithromycin – which is 100 times more expensive of co-trimoxazole – in resource limited settings.

It is well established that co-trimoxazole is an effective prophylactic against PCP pneumonia in HIV-positive individuals. The antibiotic also offers broad-spectrum protection against other bacteria and has been extensively used in HIV-positive individuals around the world. However, co-trimoxazole can cause potentially serious side-effects in HIV-positive adults and children and its extended use can lead to bacterial resistance. Some have also questioned how useful it may be in settings where a high background level of bacterial resistance to the drug is present.

Investigators from the Paediatric AIDS Clinical Trials Group conducted a randomised, double-blind, placebo controlled trial comparing the efficacy of atovaquone-azithromycin against co-trimoxazole.

The primary object of the study was to determine if atovaquone-azithromycin was as effective as co-trimoxazole at preventing serious bacterial infections. Secondary objects were to compare rates of the AIDS-defining infections PCP and MAC and non-serious bacterial infections. Data were also gathered on adverse events.

HIV-positive children aged between three months and 19 years were recruited to the study. Doses of the study medication were provided according to weight. All the children had CD4 cell counts or percentages indicating that they were at risk of developing PCP.

Originally, it was planned to recruit 280 individuals to each arm of the study, but the study was terminated after only 366 eligible patients were recruited as the introduction of HAART meant that there were no longer sufficient children who met the entry criteria.

The median duration of follow-up was a little over three years. In total, 21 patients taking atovoquone-azithromycin and 18 individuals in the co-trimoxazole arm developed a serious bacterial infection, PCP, or MAC. Treatment related side-effects were recorded in nine patients taking atovoquone-azithromycin and four children taking co-trimoxazole.

In an intent-to-treat analysis, the investigators calculated that the incidence of serious bacterial infections was 17.3 events per 100 patients years amongst individuals taking atovoquone-azithromycin and 24.2 per 100 patient years in patients randomised to receive co-trimoxazole. When the investigators looked at data for both serious bacterial infections, as well as PCP and MAC they found that the regimen of atovoquone-azithromycin was associated with a lower incidence of these outcomes (p = 0.05).

Comparison of the data concerning the frequency of adverse events in the two study arms indicated that they were statistically equivalent. A potentially life-threatening rash can be a side-effect of treatment with co-trimoxazole, but the investigators found that it did not occur with increased frequency amongst individuals randomised to receive this drug.

“These results show that atovaquone-azithromycin is as effective as [co-trimoxazole] for the prevention of serious bacterial infection in HIV-positive children”, write the investigators. They believe that the marginal advantage shown for atovaquone-azithromycin could be because of a lower incidence of sinusitis amongst patients taking the drug.

Although they conclude that their study showed that atovaquone-azithromycin was at least as effective as co-trimoxazole at preventing serious bacterial infections “its use for routine antimicrobial prophylaxis in patients with AIDS should not supercede use of co-trimoxazole, primarily because atovaquone-azithromycin is 100 times more expensive. Nevertheless, atovaquone-azithromycin offers a useful alternative for patients who cannot take [co-trimoxazole] because of adverse events or intolerance.”

An editorial accompanying this article highlights the investigators’ conclusions regarding cost. However, they note that the cost of antiretroviral drugs has dropped significantly in recent years and speculate that similar cost reductions for “adjunctive drugs are certainly not out of the question.”

They also speculate that the lower rate of sinusitis seen amongst patients taking atovaquone-azithromycin could have been because they were less likely to have resistant bacteria than patients taking long-term co-trimoxazole. Nevertheless, they draw attention to the findings of a study conducted in Zambia which established that even in the presence of resistant bacteria, co-trimoxazole remained effective at preventing pneumonia.