The authors concluded that four months of rifampin was superior to nine months of isoniazid in the treatment of latent tuberculosis infection in terms of safety and compliance, but further research is needed into tuberculosis reactivation. Given the limited evidence base and potential for bias in the review, the authors’ conclusions regarding safety and compliance should be interpreted with caution.

To compare the efficacy, toxicity, and cost effectiveness of four months of rifampin versus nine months of isoniazid for the management of latent tuberculosis infection.

MEDLINE was searched for relevant publications up to July 2009. Search terms were reported. Reference lists of included studies were manually searched.

Studies that comparing treatment with four months of rifampin versus nine months of standard treatment (isoniazid) in patients with latent tuberculosis infection were eligible for inclusion.

The main outcomes of interest were the percentage of patients not completing treatments, hepatotoxicity (defined as the percentage of patients meeting grade three or four criteria for liver failure leading to discontinuing study treatment), and the number of treatment failures.

Included studies were single-centre studies (conducted in Canada or the USA) or multi-centre (conducted in Canada, Saudi Arabia and Brazil). Patients received 10mg/kg rifampin (maximum 600mg) or 5mg/kg (maximum 300mg).

Both reviewers independently screened studies for inclusion.

The authors did not state that they assessed study quality.

Both reviewers extracted outcome percentages to calculate risk ratios (RRs) and 95% confidence intervals (CIs). Discrepancies were resolved by consensus.

A fixed-effect model, or random-effects model where statistical heterogeneity was 50% or more, was used to pool risk ratios and 95% confidence intervals. Statistical heterogeneity was assessed using τ², X² and I² statistics.

Four studies were included in the review (n=3,692 patients, range 116 to 2,255). These included one historical study, two randomised open label trials and one retrospective study.

Patients that received four months of rifampin were statistically significantly less likely to discontinue treatment (RR 0.53, 95% CI 0.44 to 0.63; four studies; I²=54%) than patients that received nine months of isoniazid. Patients that received four months of rifampin were less likely to experience hepatotoxicity (RR 0.12, 95% CI 0.05 to 0.30; four studies; I²=0%) than patients that received nine months of isoniazid.

Two studies assessed treatment failure. One study reported that active tuberculosis on one patient receiving four months of rifampin. One study reported tuberculosis in one patient that received nine months of isoniazid.

Cost-effectiveness analyses showed that four months of rifampin was more cost-effective than nine months of isoniazid when doctor and/or nurse fees were included (saving of $213 per patient treated) or excluded (saving of $90 per patient treated).

Four months of rifampin remained more cost effective than nine months of isoniazid provided that the cost of rifampin remained below $127 per month and where laboratory costs remained at more than 80% of the baseline value.

Four months of rifampin was superior to nine months of isoniazid in the treatment of latent tuberculosis infection in terms of safety and compliance.

The review question was clear and was supported by appropriate inclusion criteria for intervention, comparison, participants and outcomes, but broad for study design. The literature search was limited to one database; it was unclear whether any language restrictions were applied or whether attempts were made to locate unpublished data, so potentially relevant studies may have been missed. Study selection and data extraction were performed in duplicate, reducing the potential for error and bias.

Study quality did not appear to have been assessed, so the robustness of the findings was unclear. Patient and study details were lacking, so it was unclear whether pooling of the results was appropriate. Only a small number of studies were included and there was some evidence of statistical heterogeneity.

The authors’ conclusions reflected the evidence available and their recommendation for further research seemed appropriate. However, as the evidence base was of a small number of studies of unknown quality and there was potential for some bias in the review, the authors' conclusions regarding safety and compliance should be interpreted with caution.

Practice: The authors stated that the low cost and short duration of four month rifampin therapy may be especially attractive to patients with isoniazid resistance, those with limited access to health care, and special populations such as prison inmates.

Research: The authors stated that a large trial is needed to determine the risk of tuberculosis reactivation in patients receiving four months of rifampin.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.