Widespread use of HAART has reduced the risk of developing clinical TB among persons infected with HIV and may help bring about further declines when integrated into TB programmes. The effect of HAART on the risk of TB among persons infected with HIV has been examined in several studies. The risk of TB was up to 80% lower among persons prescribed HAART and 40% lower among persons prescribed other non-HAART antiretroviral therapy than the risk in persons not prescribed antiretroviral therapy. The protective effect of HAART was greatest in symptomatic patients and those with advanced immune suppression but was not apparent in those with CD4 counts >350 cells/uL.^31-33 Its effect is almost certainly related to improvements in systemic immunity (measured by an increase in the CD4 count) to a point where the risk of new or reinfection is greatly diminished.

There have been several short-term controlled trials in HIV positive persons showing the protective effect of chemo preventative therapy. ^98-110 

A protective effect of isoniazid is found only in those who are tuberculin skin test positive. This protective effect appears to only last 2 to 4 years as compared with 19 years or more in non-HIV populations. Such a short term effect in HIV positive patients studied especially in areas of high TB prevalence may be explained by the fact that the majority of the tuberculosis in HIV population arises from exogenous sources and thus are not from reactivation of latent TB but are new. Beyond recognized outbreaks, there is little evidence to suggest that re-infection (as opposed to reactivation) is a major factor in the UK.

 A pragmatic but still theoretical approach in those HIV patients who are at increased risk of TB, eg immigrants, is to give isoniazid prophylaxis until the CD4 count has risen to above a reasonable threshold, say 200–300 CD4 cells/uL on HAART and then it could be stopped. Data are needed on what the threshold might be as patients may need to be on isoniazid for more than 1 year and the effects of this are relatively unknown.

The routine use of such chemo preventative therapy in this setting is not recommended.  [DI]

11.1 The treatment of latent tuberculosis infection

The treatment of latent tuberculosis infection includes:

1. Isoniazid for a total of 6- 9 months
2. Rifampicin with isoniazid for a total of 3 months
3. Rifampicin alone for 4 months

Short courses of chemo preventative therapy using other drugs have been recommended to help overcome poor adherence.  Unfortunately rifampicin and pyrazinamide given three times a week for 2 months has been associated with severe and fatal hepatic reactions in 5 non-HIV patients with a total of 21 cases of liver injury reported to CDC. ¹¹¹

However this complication was not been seen in the studies of HIV positive patients taking this regimen. 

It is known from RFLP studies that many tuberculosis infections in HIV positive patients in TB endemic areas appear to be new infections rather than reactivation of the original TB. ¹¹²

Isoniazid may prevent such exogenous infection but would then have to be given long term or at least until there was a substantial CD4 rise on HAART^113-115.  There are no current data to support such a strategy.