Cotrimoxazole prophylaxis significantly reduces the risk of death in HIV-positive patients being treated for tuberculosis, according to the results of a randomised controlled trial conducted in Zambia and reported in the online edition of the British Medical Journal. The study’s findings “provide a strengthened evidence base for the UNAIDS/WHO guidelines issued in 2000” that all individuals with symptoms of HIV in Africa should receive prophylactic therapy with co-trimoxazole, say the authors.
Tuberculosis and HIV are leading causes of death in sub-Saharan Africa, and Zambia has one of the highest rates of tuberculosis in the world. Most of these individuals (70%) are HIV-positive.
Although tuberculosis can be successfully treated, it is associated with a high risk of death in HIV-positive individuals. There is evidence that a substantial proportion of these deaths is caused by bacterial sepsis. Furthermore, rates of bacterial pneumonia are several times higher amongst patients with HIV than in the general population.
In 1999 a randomised controlled trial found that prophylaxis with cotrimoxazole led to a significant reduction in the risk of death in HIV-positive patients receiving tuberculosis therapy in Côte d’Ivoire. On the basis of this and other studies, the World Health Organization (WHO) and UNAIDS issued provisional guidance in 2000 that all symptomatic HIV-positive patients in Africa should receive cotrimoxazole prophylaxis.
Although some countries have implemented this guidance, others have not, citing a belief that the situation in Côte d’Ivoire was not necessarily applicable to them. There have also been concerns that resistance to co-trimoxazole could compromise some malaria treatment.
Zambia has no policy on cotrimoxazole prophylaxis. Therefore an international team of investigators designed three placebo-controlled randomised controlled trials, one of which involved HIV-positive patients with tuberculosis. The main aim of the study was to see if such prophylaxis reduced the risk of death due to any cause in HIV-positive patients receiving tuberculosis treatment. A secondary aim was to assess the side-effects of cotrimoxazole treatment.
Two groups of HIV-positive patients with tuberculosis were recruited to the study. Patients in the first group were receiving tuberculosis treatment; the second had received successful tuberculosis therapy. None were taking anti-HIV drugs.
A total of 1003 patients were randomised to receive daily cotrimoxazole or a placebo. The investigators’ main analysis concerned the 835 patients who were currently receiving tuberculosis therapy. Follow-up information was available for 762 of these individuals.
A total of 1016 person years of follow-up were available for analysis, and during follow-up a total of 310 patients died (147 cotrimoxazole treated vs. 163 placebo). The rate of death was 27.3% per 100 person years for the patients receiving cotrimoxazole compared to 34.4% per 100 person years amongst those given the placebo. The investigators calculated that cotrimoxazole was associated with a statistically significant 21% reduction in the risk of death (hazard ratio, 0.79, 95% CI, 0.63 – 0.99, p = 0.04).
This finding was not affected when the investigators controlled for CD4 cell count, age and sex.
But the researchers did find that the benefits of cotrimoxazole prophylaxis varied over time. There was no reduction in the risk of death during the first six months of therapy. Cotrimoxazole treatment between months six and 18 led, however, to a 45% reduction in mortality compared to patients receiving the placebo. After 18 months there was once again no difference in the rate of death between the two groups of patients.
The investigators think the inclusion of rifampicin in tuberculosis treatment during the first six months would have protected both arms of the study against the risk of bacterial infections. Rifampicin has a broad anti-bacterial effect. The lack of a difference after 18 months could, the investigators believe, be explained by a trend to poorer adherence over time.
A total of 18 patients stopped treatment because of side-effects. Twelve of these individuals were receiving co-trimoxazole, the other six the placebo. Causes of treatment interruption included Stevens Johnson syndrome with anaemia, itchy rash, swollen face and lips, peripheral neuropathy, and anaemia.
Rates of death amongst patients who had received tuberculosis treatment in the past were almost identical at 14.5% per 100 person years amongst those receiving cotrimoxazole and 14.7% per 100 person years in the placebo arm.
“Prophylaxis with cotrimoxazole reduces mortality in adults with HIV infection who have pulmonary tuberculosis” write the investigators. They add, “cotrimoxazole was generally safe and well tolerated.”
Concerns about cotrimoxazole resistance should not, the investigators believe, prevent the provision of prophylaxis, particularly given recent guidance that coartemether should be used for malaria prophylaxis.
Linkage between HIV and tuberculosis programmes is “essential”, and the investigators believe that this could make “co-trimoxazole increasingly available.”
An accompanying editorial applauds the study’s findings but expresses frustration at the slow progress in the provision of cotrimoxazole prophylaxis in Africa. Its authors conclude “in sub-Saharan Africa, prophylaxis with cotrimoxazole and isoniazid are effective and these drugs should be prescribed. Trials are needed to assess their main alternative – starting antiretroviral therapy.”
Nunn A.J. et al. Role of co-trimoxazole prophylaxis in reducing mortality in HIV infected adults being treated for tuberculosis: randomised controlled trial. BMJ (online first), 2008.
Prophylaxis with co-trimoxazole for HIV infected adults in Africa: co-trimoxazole reduces mortality even in settings where bacterial resistance may be high. BMJ (online first), 2008.