In many high-income countries, isoniazid preventive therapy has formed the cornerstone of tuberculosis prevention for nearly 50 years. Isoniazid is also effective in low-income and middle-income countries with a high tuberculosis burden, although widespread implementation has been slow. In The Lancet Infectious Diseases, Betina Durovni and colleagues (1) present a large pragmatic before-and-after assessment of a policy to increase tuberculosis screening and the use of isoniazid preventive therapy to reduce the incidence of HIV-associated tuberculosis (HIV/TB) in Rio de Janeiro, Brazil. The ambitious stepwise intervention was multipronged and delivered across the city's health service. Primary analyses did not show an overall reduction in tuberculosis incidence, although there was a small reduction in the combined outcome of tuberculosis and all-cause mortality. Adjusted for clinical characteristics, the intervention had a modest effect on outcomes. Effect sizes varied by adherence to clinical follow-up, and were greatest in the group of patients who remained in care.
There are issues inherent in this phased multifaceted intervention that mean the results warrant careful interpretation. Adherence to clinical follow-up was better during the intervention period; substantial person-time lost in the per-protocol analysis was from the control period. The study outcome is attributed to prescription of 6 months of isoniazid. However, independent effects of individual elements of the intervention are difficult to separate from each other. The intervention encompassed operational training of all levels of clinic staff—including social workers, nutritionists, and psychologists—to better recognise and treat tuberculosis, to increase the use of tuberculosis screening, tuberculin skin tests, and isoniazid preventive therapy, and to encourage periodic non-clinical supervisory contact with clinics. The intervention period also overlapped with the implementation of important economic, social, and health reforms in Brazil (2). These measures strengthened the health service, improved quality of care, reduced socioeconomic determinants of ill health, and improved overall health outcomes. (3, 4). Thus, overall improvement in health outcomes could have affected the results. (4, 5). The effect of this study would also be difficult to separate from that of secular trends arising from these national reforms (3, 6).
Despite design concerns and modest outcomes, Durovni and colleagues' study provides important insights for tuberculosis control. The outcomes reveal several challenges in prevention of tuberculosis in patients with HIV in moderate-to-high incidence countries: enrolment and retention of individuals likely to benefit in care can be difficult; simple algorithms to screen for tuberculosis in patients with HIV are insensitive; motivating otherwise healthy people to take lengthy courses of an antibiotic on the basis of a test that merely infers possible infection from evidence of immune sensitisation can be difficult; and scale-up of interventions to reach all those at risk and early can be challenging.
The authors suggest the following methods for scale-up of prevention of tuberculosis in HIV: use of interferon γ release assays to simplify identification of latent tuberculosis infection and to shorten the delay to treatment initiation; to provide isoniazid to all HIV-infected patients without testing for latent tuberculosis infection; or to offer antiretroviral therapy (ART) early with isoniazid preventive therapy.
There are several important considerations for scaling up treatment. Brazil's current isoniazid guidelines for patients with HIV are based on earlier trials, mostly done in high-incidence countries, in cohorts of patients who had not started ART (7). Patients who had a negative tuberculin skin test, possibly non-reactive because of immunosuppression, derived limited benefit. 73% of the cohort was receiving ART by the end of the study. In the context of ART-mediated immune reconstitution, the dynamic interaction between the host and Mycobacterium tuberculosis is unlikely to be similar to that in patients who have not started ART (8).This will affect the results of tuberculin skin tests, interferon-γ release assays, and the effect of isoniazid preventive therapy. Untargeted isoniazid is attractive, and it will increase the numbers eligible for treatment. However, large numbers of patients might never benefit but risk adverse events, and health systems could be overburdened. This is important for Brazil and similar low-income or middle-income settings where repeat infection is unlikely. Multivariate algorithms to identify people who should be prioritised for isoniazid preventive therapy are needed. These algorithms should not rely on imperfect tests for latent tuberculosis infection.
The study shows that tuberculosis or death in people with HIV could be reduced through policies that introduce isoniazid preventive therapy with complementary strategies in the context of reforms to modify social determinants of health. Our view is supported by preliminary results of trials of community-wide control interventions for HIV and tuberculosis, including isoniazid preventive therapy, in southern Africa, where similar reforms were not in place and which did not show an overall or sustainable effect (9—12). Per- protocol analyses in one study showed an effect for 6 month isoniazid which immediately waned, possibly because of tuberculosis reinfection (9, 11). Brazilian reforms might reduce exposure of individuals to repeated tuberculosis infections, thus 6 month treatment might be enough as repeated exposures from tuberculosis become less likely.
The most important lesson learned from Durovni and colleagues' study is that policies that focus on individual biomedical approaches to prevent tuberculosis in HIV, such as isoniazid preventive therapy, might not have a sustainable effect without a supporting package of interventions to address wider health system failings.
We declare we have no conflicts of interest.
References
1 . Effect of improved tuberculosis screening and isoniazid preventive therapy on incidence of tuberculosis and death in patients with HIV in clinics in Rio de Janeiro, Brazil: a stepped wedge, cluster-randomised trial. Lancet Infect Dis 2013. published online Aug 16. http://dx.doi.org/10.1016/S1473-3099(13)70187-7.
2 . The Brazilian health system: history, advances, and challenges. Lancet 2011; 377: 1778-1797. Summary | Full Text | PDF(2942KB) | CrossRef | PubMed
3 . Health conditions and health-policy innovations in Brazil: the way forward. Lancet 2011; 377: 2042-2053. Summary | Full Text | PDF(170KB) | CrossRef | PubMed
4 . Issues in the design and interpretation of studies to evaluate the impact of community-based interventions. Trop Med Int Health 1997; 2: 1022-1029. PubMed
5 . Evaluation designs for adequacy, plausibility and probability of public health programme performance and impact. Int J Epidemiol 1999; 28: 10-18. CrossRef | PubMed
6 . Successes and failures in the control of infectious diseases in Brazil: social and environmental context, policies, interventions, and research needs. Lancet 2011; 377: 1877-1889. Summary | Full Text | PDF(517KB) | CrossRef | PubMed
7 . Treatment of latent tuberculosis infection in HIV infected persons. Cochrane Database of Syst Rev 2010; 1. CD000171.
8 . Dissection of regenerating T-cell responses against tuberculosis in HIV-infected adults sensitized by Mycobacterium tuberculosis. Am J Respir Crit Care Med 2009; 180: 674-683. CrossRef | PubMed
9 Churchyard G, Fielding K, Lewis J, et al. Community-wide isoniazid preventive therapy does not improve tb control among gold miners: the Thibela TB study, South Africa. 19th Conference on Retroviruses and Opportunistic Infections; Seattle WA, USA; March 5—8, 2012. Abstract 150aLB.
10 Ayles H, and the ZAMSTAR Study Team. A household-based HIV and TB intervention increases HIV testing in households and reduces prevalence of TB at the community level: the ZAMSTAR community randomized trial. 19th Conference on Retroviruses and Opportunistic Infections; Seattle WA, USA; March 5—8, 2012. Abstract 149bLB.
11 Fielding K, Grant A, Lewis J, Hayes R, Churchyard G. Individual-level effect of isoniazid preventive therapy of risk of TB: the Thibela study. 19th Conference on Retroviruses and Opportunistic Infections; Seattle WA, USA; March 5—8, 2012. Abstract 150bLB.
12 Samandari T, Agizew T, Nyirenda S, et al. TB incidence increase after cessation of 36 months' isoniazid prophylaxis in HIV+ adults: Botswana. 19th Conference on Retroviruses and Opportunistic Infections; Seattle WA, USA; March 5—8, 2012. Abstract 147.
**********************************
Molebogeng Xheedha Rangaka a b c d g, Robert John Wilkinson a e f g
a Clinical Infectious Diseases Research Initiative, Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Observatory 7925, South Africa
b Centre for Infectious Diseases Epidemiology and Research, University of Cape Town, Observatory 7925, South Africa
c Department of Infection and Population Health, University College London, UK
d Department of Medicine, University College London, UK
e Imperial College London, London, UK
f MRC National Institute for Medical Research, London, UK
g Wolfson Pavilion, Institute of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, Unversity of Cape Town, Observatory 7925, South Africa
**********************************
Source: The Lancet