Rifampicin and pyrazinamide treatment for latent TB in HIV patients does not carry high risk of liver injury, but editorial still urges caution

The treatment of latent tuberculosis in HIV-positive patients with a two month course of rifampicin and pyrazinamide does not carry a higher risk of liver damage when compared to standard isoniazid treatment, according to a retrospective US study published in the August 15^th edition of Clinical Infectious Diseases. Concerns had been expressed by US drug authorities about the hepatitic safety of this regimen following hospitalisations and deaths amongst HIV-negative patients taking it.

However, an editorial in the same issue of the journal states that, the even with the investigators’ findings, the concerns about the hepatic safety of pyrazinamide for the treatment of latent TB mean that it would be prudent to restrict its use to circumstances where standard isoniazid therapy is not feasible.

Data concerning the use of rifampicin and pyrazinamide were gathered by investigators during a large clinical trial which had already shown this regimen to be an effective treatment for latent TB in HIV-positive individuals. There were no reports of severe liver abnormalities in HIV-positive patients during this trial, and US drug regulators said that the regimen could be used as an alternative to standard treatment, twelve months of isoniazid.

However, shortly after this advice was issued, there were several reports of severe liver toxicities in HIV-negative patients receiving rifampin and pyrazinamide therapy for latent TB. This led the US regulatory authorities to amend its guidelines and recommend that the rifampicin and pyrazinamide regimen should “should generally not be offered” for latent TB in any population.

Investigators from the trial looking at the safety and efficacy of rifampicin and pyrazinamide in HIV-positive patients with latent TB conducted a retrospective review of the liver function tests obtained during their study to better understand the incidence of severe liver toxicities and their risk factors.

A total of 1,583 HIV-positive individuals were recruited to the trial. All the patients had a reactive tuberculin skin test suggesting latent TB infection. Patients were excluded from the trial if they had a bilirubin level above 2.5mg/dl or an aspartate aminotransferase (AST) level greater than five times the upper limit of normal at baseline, or acute hepatitis. Individuals had a mean age of 37 years, 86% were black or Latino, 28% were women and the median CD4 cell count at baseline was 436 cells/mm³.

Patients were randomised to receive a two-month regimen of rifampicin or pyrazinamide or a twelve-month course of isoniazid. At months one and two patients had their bilirubin and AST levels evaluated.

There were no significant differences in bilirubin or AST levels between the two arms of the study at baseline. Of the patients treated with isoniazid, 0.6% developed a grade III bilirubin level (>2.5mg/dl) at either month one or two compared with 1.8% of patients randomised to receive rifampicin and pyrazinamide (a non-significant difference, p = 0.06). Grade III elevations in AST (>250u/l) occurred in 1.6% of isoniazid-treated individuals and 2.1% of patients receiving rifampicin and pyrazinamide (again, this difference was non-significant, p = 0.056). There were no hospitalisations or deaths due to the use of the study medications.

Investigators performed a multivariate regression analysis to better understand the factors associated with bilirubin and AST increases. The use of rifampicin and pyrazinamide (p = 0.02), non-white race (p = 0.022), and female sex (p = 0.023) were all associated with an increase in bilirubin by >0.5 mg/dl. Only age above 40 years was significantly associated with an increase in AST levels by 40 u/l (p = 0.026).

“The data presented here support the low level of clinically meaningful hepatotoxicity associated with [rifampicin and pyrazinamide] in an HIV-infected population,” write the investigators.

The investigators could not easily explain why there was a lack of serious liver toxicity associated with the use of rifampicin and pyrazinamide in HIV-positive patients compared to higher rates seen in HIV-negative individuals. In particular, immune status did not appear to be a factor as the individuals enrolled in the current study had a good CD4 cell count, and only 7% had an AIDS diagnosis.

“The analyses presented here, as well as data from the other two comparative trials of rifampicin and pyrazinamide, support the safety of [this] regimen for the treatment of latent TB in HIV-infected persons,” conclude the investigators. However, they caution that individuals receiving this regimen should be closely monitored and support the US regulatory authorities’ recommendations that this regimen should not be used in HIV-negative patients.

An editorial accompanying the study however, states “several issues gnaw in the recent history of two-month courses of pyrazinamide treatment for latent TB.” The editorial’s author notes that conditions in clinical trials do not replicate those in clinical practice where “hepatotoxicity cofactors may be more widespread and less appreciated.” The editorial concludes, “it seems prudent to restrict the use of pyrazinamide for treatment of latent TB to very specific instances in which the use of isoniazid is not feasible.”