Researchers from South Africa and Canada have carried out a Cochrane review update to assess the safety and effectiveness of corticosteroids for treating tuberculous pericarditis.

Tuberculosis (TB) infection of the pericardium is not common but can restrict the function of the heart and is fatal in some cases. Current treatment involves doctors prescribing anti-TB drugs for six months, draining fluid from the pericardium or removing it in some situations, and in some cases corticosteroids are prescribed to reduce inflammation. However, there has been some concern that steroids have the potential to cause harm to patients with HIV, a common TB co-infection.

The review authors included seven trials, six looking at the use of corticosteroids and the other at different treatments. All trials were based in sub-Saharan Africa and included 1959 participants in total.

“Given the seriousness of TB pericarditis it was important that we were able to assess the effectiveness of using corticosteroids as a treatment,” said Professor Charles Shey Wiysonge, Director of Cochrane South Africa and lead author of the review, “it was also especially important to look at their safety in cases of HIV coinfections, as there has previously been an impression that treatment could cause a rise in HIV-related cancers.”

The review authors found that in people without HIV, corticosteroids drugs may reduce the number of people dying overall by 20%, as well as probably reducing deaths from pericarditis, although the confidence intervals include the possibility of both large beneficial effects and small increases in harm. The results also show that in HIV-positive patients, corticosteroids may reduce constriction of the pericardium membrane, but the confidence intervals include the possibility of both large beneficial effects and small increases in harm.

“This systematic review, which stratified patients into HIV-positive and HIV-negative participants and excluded those who received dual immunotherapy with corticosteroids and Mycobacterium indicus pranii, showed promising effects of corticosteroids on major outcomes but generally the confidence intervals include the possibility of both large beneficial effects and small increases in harm, and there was no evidence of an increase in cancer,” added Professor Bongani Mayosi, senior author of the review and lead author of the largest trial carried out on this topic. The patients on a combination of corticosteroids and Mycobacterium indicus pranii were excluded from the analysis because of the possibility of an interaction between the two agents resulting in the increased incidence of cancer in this group.