TB: Mass Isoniazid Preventive Therapy Is Ineffective

Laurie Barclay, MD

January 22, 2014

Mass screening and isoniazid treatment for latent tuberculosis (TB) did not significantly affect TB control in South African gold mines, even though isoniazid successfully prevented TB during treatment, according to findings from a cluster-randomized study published in the January 23 issue of the New England Journal of Medicine.

"[TB] is a leading cause of death in adults globally and was responsible for an estimated 1.4 million deaths in 2011," write Gavin J. Churchyard, MB, BCh, PhD, from the Aurum Institute and the School of Public Health, University of the Witwatersrand in Johannesburg, South Africa, and colleagues from the Thibela TB Study Team. "[HIV] infection, exposure to silica dust in ultradeep mines, and close working and living conditions predispose South African gold miners to [TB]. Despite standard control measures and annual active case finding, the escalating prevalence of HIV (29% in 2001) intensified the [TB] epidemic."

The goal of this trial was to assess an intervention to interrupt TB transmission by mass screening that was linked to treatment for active disease or latent infection. The investigators reasoned that this could be effective, as a randomized trial in the 1960s in Alaska, where TB was epidemic, showed that isoniazid preventive therapy that was delivered to all household members yielded a 55% decline in TB incidence over the course of 6 years.

The researchers randomly assigned 15 clusters with a total of 78,744 miners to either the intervention group (40,981 miners in 8 clusters) or control group (37,763 miners in 7 clusters). In the intervention clusters, all miners were offered TB screening and were referred for treatment if active TB were diagnosed. Miners without active TB were offered 9 months of isoniazid preventive therapy.

The main study endpoint was the cluster-level incidence of TB during the 12 months after the intervention ended; secondary endpoints included TB prevalence at study completion.

Intervention Did Not Reduce TB Transmission

Screening took place in 27,126 miners (66.2%) in the intervention clusters, of whom 23,659 (87.2%) started taking isoniazid. Depending on the cluster, 35% to 79% of the miners received isoniazid for at least 6 months.

The intervention was not associated with lower incidence or prevalence of TB or with lower rate of death from any cause. Incidence rates were 3.02 per 100 person-years in the intervention clusters and 2.95 per 100 person-years in the control clusters, yielding a rate ratio in the intervention clusters of 1.00 (95% confidence interval [CI], 0.75 - 1.34; P = .98; adjusted rate ratio, 0.96; 95% CI, 0.76 - 1.21; P = .71). Prevalence rates were 2.35% vs 2.14%, respectively, with an adjusted prevalence ratio of 0.98 (95% CI, 0.65 - 1.48; P = .90).

The investigators looked at the direct effect of isoniazid in 10,909 miners and found a lower incidence of TB during treatment (1.10 cases per 100 person-years in miners receiving isoniazid vs 2.91 cases per 100 person-years in control patients; adjusted rate ratio, 0.42; 95% CI, 0.20 - 0.88; P = .03). However, the protective effect declined rapidly after isoniazid prevention therapy was stopped.

"Mass screening and treatment for latent [TB] had no significant effect on [TB] control in South African gold mines, despite the successful use of isoniazid in preventing [TB] during treatment," the authors write.

Limitations of this study include variable participation in the intervention and retention in the study across clusters, a suboptimal proportion of miners taking isoniazid simultaneously, and limited data regarding the role of other factors that may have compromised the intervention. Other limitations included the inability to determine HIV prevalence; a relatively small number of clusters, which limited the study power; that varying routine TB control programs differed by mining company; and limitations regarding the analysis of direct effect.

"The best achievable implementation of the intervention is unlikely to have substantially changed the result, given that the best-performing cluster had excellent uptake and retention, but the intervention nevertheless had a modest and short-lived effect," the authors conclude. "Contributing factors include increased vulnerability to [TB] due to HIV infection and silicosis, along with the ongoing transmission of [TB]. [TB] infection control and strategies for controlling HIV and exposure to silica dust should be expanded."

The study authors also recommend finding systems that minimize the time from a positive microbiologic result to TB treatment, considering continuous isoniazid preventive therapy and strategies to maximize retention for persons at highest risk for TB, and using mathematical modeling to identify combinations of strategies that are more likely to control TB in gold mines.

Better Strategies Needed

In an accompanying editorial, Eric J. Rubin, MD, PhD, from the Department of Immunology and Infectious Diseases, Harvard School of Public Health in Boston, Massachusetts, speculates that the very large number of persons with latent TB makes prevention problematic. Such latent infections, he notes, represent a huge reservoir of potential reactivation.

The findings of this study were "disappointing and a bit surprising," he writes. They call into question the value of secondary prophylaxis with isoniazid. Even though the Alaskan trial in 1967 showed efficacy for this strategy, it is not without costs.

"Along with the direct expenses associated with buying and administering isoniazid, some patients with unrecognized active [TB] will receive inadequate therapy," Dr. Rubin writes. "[I]n such patients, treatment with isoniazid not only will fail but also will increase the risk of antibiotic resistance. Despite this caveat, isoniazid preventive therapy is increasingly recommended in an effort to control [TB], particularly in high-burden areas with high rates of [HIV] infection, such as Africa."

He attributes the lack of success of secondary prophylaxis with isoniazid in this study to rapid disappearance of protection after treatment completion and to unimpressive uptake of isoniazid preventive therapy.

For better TB prevention, Dr. Rubin recommends social interventions that increase compliance; for example, using a shorter regimen such as a 12-week regimen of isoniazid and rifapentine.

"[N]ew drugs such as bedaquiline, which may have theoretical benefits among patients with bacteria that are not growing rapidly, have not yet been shown to be safe enough to treat otherwise healthy persons," Dr. Rubin writes.

"[P]atients are more likely to comply with health-improvement advice when the consequences of noncompliance will be immediate and undesirable. For preventing [TB], it is likely that some combination of improved identification and treatment of patients with active [TB], technical improvements (e.g., an effective vaccine or a very short course of therapy for latent [TB] outside of settings with heavy [TB] transmission), and measures to increase compliance will be required for the intervention to be highly effective," he concludes.

The Consortium to Respond Effectively to the AIDS TB Epidemic supported this study. The consortium was funded by the Bill and Melinda Gates Foundation; the South African Mine Health and Safety Council; the Foundation for Innovative New Diagnostics, Switzerland; the National Institutes of Health, National Institute of Allergy and Infectious Diseases; the UK Department of Health; and the UK Medical Research Council. Sanofi-aventis donated isoniazid to the study. Full conflict-of-interest information is available on the journal's Web site. Dr. Rubin has disclosed no relevant financial relationships.

N Engl J Med. 2014;370:301-310, 375-376. Article abstract, Editorial extract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.