Deadly Duo: Joining Forces to Fight TB and HIV

Extensively drug-resistant tuberculosis adds a new dimension to an old public health menace that can act in deadly synergy with HIV

By Catherine Zandonella

When a highly resistant form of tuberculosis emerged in a hospital in South Africa's KwaZulu-Natal province in 2005, its first victims were people living with HIV and AIDS. Within a month of diagnosis, extensively drug-resistant tuberculosis, or XDR-TB, had killed 44 HIV-infected people.

An astonishing one-third of the world's population is infected with the bacterium that causes TB, Mycobacterium tuberculosis. Most have a latent form but about 10% of those people will develop active TB disease within their lifetimes. Antibiotics can prevent and treat most of these cases, yet 8-10 million people develop active TB annually and 2 million die from it.

In recent decades the risk posed by TB has been far greater due to the relentless march of the HIV/AIDS pandemic (Figure 1). Stephen Lewis, the United Nations Special Envoy for HIV/AIDS in Africa, has called the two diseases a "combination made in hell." HIV-infected individuals have a 20-fold greater risk of developing active disease, and TB is now the leading cause of death among HIV-infected people around the world.

Figure 1. Disproportionate Burden of HIV, HIV-Related Tuberculosis Coinfections in Africa, for 2000.

View Larger Image

The global health community is responding by changing how it confronts these two diseases to promote collaboration between historically separate TB and HIV/AIDS programs. A major goal of the World Health Organization (WHO)'s new Stop TB Strategy, launched in 2005, is to decrease the burden of TB and HIV in populations affected by both diseases. The plan is endorsed by a coalition of organizations involved in TB and HIV care, including the Joint United Nations Programme on HIV/AIDS (UNAIDS), public-private partnerships, and non-governmental organizations. "Intensifying collaboration between the TB and HIV control programs in order to ensure the delivery of integrated TB and HIV services in a primary health care platform is crucially needed," says Haileyesus Getahun, secretary of the TB/HIV working group at the WHO.

Around the world the HIV epidemic is making it impossible to continue to confront TB through traditional approaches. The global health community has promoted a program called DOTS since the early 1990s, which stands for directly-observed therapy, short-course. The goal was to bring TB under control through enhanced case detection and short-course antibiotic treatment given under direct observation. This supervision was meant to roll back TB morbidity and mortality while curbing the emergence of drug-resistant strains.

Through DOTS public health officials hoped to achieve a 70% case detection rate and an 85% cure rate by 2005. But while the program has been successful in reducing case rates, according to the WHO, HIV is the main reason for the failure to meet control targets in high HIV prevalence regions. Even when DOTS runs as it should TB rates in HIV-endemic areas continue to rise. In the South African gold mining industry, for example, DOTS adherence is strong and includes yearly TB screens for all miners, and yet the TB case rates have risen four-fold since 1990. During that time, the HIV prevalence has increased from less than 1% to almost 30%.

The new STOP TB policy combines DOTS with a greater awareness of the synergy between HIV and TB. Together with the WHO's Department of HIV/AIDS, STOP TB has devised an interim policy that recommends more thorough surveillance and prevention both of HIV among TB patients and of TB in people living with HIV/AIDS. "TB screening for HIV positives or for those attending HIV services is not often done in many settings due to lack of recognition of the importance of TB in people living with HIV," says Getahun. "TB prevention, diagnosis, and treatment services need to be core functions of HIV prevention, treatment, and care services, and vice versa."

Immune consequences

M. tuberculosis and HIV are a deadly combination because both pathogens attack the immune system and induce a complex immunopathology. The precise molecular mechanisms are the focus of much current research. HIV ravages CD4+ T cells and cripples CD8+ T cells. M. tuberculosis induces anergy in T lymphocytes, rendering them unresponsive to antigen and effectively paralyzing the immune system. "TB strongly inhibits the immune system to make its way around the body," says Jerald Sadoff, president and chief executive officer of the AerasGlobal TB Vaccine Foundation. "Both diseases suppress the immune system independently."

The majority of TB infections are acquired when an individual inhales bacilli shed via the coughing or sneezing of a family member or other person with whom they are in frequent contact. The bacilli lodge deep in the lungs and are kept in check by the immune system, resulting in a latent infection that causes neither discomfort nor infectiousness.

Latent infection is far more likely to develop into active TB in individuals with compromised immune systems. In active TB the bacteria multiply and carve out cavities called tubercles in the lungs. The bacteria can break into the bloodstream and colonize other organs such as the kidneys and brain. Patients with active pulmonary TB develop a persistent cough and can spread the disease to others.

Although TB takes its greatest toll on people with low CD4+ T lymphocyte counts, the disease can affect people at all levels of immune status, making it one of the most common opportunistic infections. While more susceptible to TB, HIV-infected individuals are also less infectious. They are more likely to have extrapulmonary TB so they are less likely to spread the disease through coughing. Also, they quickly progress to advanced TB, giving them fewer years of life in which to transmit the disease.

By contrast, HIV-uninfected individuals may be infectious for several years and able to spread the disease to many immunocompromised individuals. A person with active TB can transmit the pathogen to 10-15 people a year. TB-infectiousness is greatest before diagnosis since most patients become noninfectious soon after starting treatment. This combination of vulnerability to TB among immunocompromised HIV-positive individuals and prolonged transmission from HIV-negative patients with TB is fueling the TB epidemic in areas with high HIV prevalence.

Microbe hunting

Improved diagnostics for TB will be imperative in stepping up surveillance and treatment. The complex interaction between the bacterium and the immune system means that the disease can take a number of forms, some extremely difficult to detect. While HIV can be detected through a relatively simple antibody test, diagnosis of latent or active M. tuberculosis infection presents particular challenges. Antibody tests for M. tuberculosis have proven useless because the microbe can cross-react with related mycobacteria in the environment. In many regions of the world the techniques used for diagnosing TB have remained relatively unchanged for the past hundred years.

The most common test for latent infection is a skin test where heat-killed M. tuberculosis proteins are injected subcutaneously to look for immune hypersensitivity that occurs in people with prior exposure to TB. Though quick and easy, roughly 25% of those with active TB may have negative skin tests, meaning that people who need antibiotics to prevent the transition from latent to active TB will not to get them.

The test does not work in newborns and babies due to their immature immune systems. Nor does it work well in immunosuppressed individuals, who may be unable to mount the immune response needed for the test to work. "People with HIV often cannot give the immunological reaction that you need to get a positive test," says Hans Rieder of the International Union Against Tuberculosis and Lung Disease in Paris.

The test can also return false positive results, especially in people who've received a vaccination against TB. The vaccine, known as bacille Calmette-Guerin or bCG, is used widely in areas with high M. tuberculosis infection rates to immunize infants but its protection is mostly against disseminated infection and lasts only throughout childhood. A new test that measures interferon-g returns fewer false positives and can be used in areas where people were vaccinated. However, it still returns false negatives and is expensive.

If latent TB is detected then treatment with the antibiotic isoniazid can dramatically reduce the lifetime chance of developing active TB. Yet it is important to make sure that an individual has latent and not active TB since the latter should be treated with an aggressive combination of antibiotics.

Active TB can be diagnosed with a chest X-ray, but it must be confirmed by other methods. The gold standard is the smear test, which involves taking a sample of coughed-up sputum and smearing it onto a microscope slide, fixing and staining it, and then examining for the distinctive cellular appearance of M. tuberculosis.

The smear test cannot detect extrapulmonary TB. This smear-negative TB has a worse prognosis than smear-positive disease in people living with HIV. People with HIV often present extrapulmonary cases of TB that manifest as febrile or other disorders that go unrecognized as TB by clinicians.

Growing the sputum sample in culture before searching for bacteria under the microscope can improve the detection of HIV-related TB, but this requires more training and special equipment. "In most resource-poor countries it is a challenge to obtain and maintain a sufficient number of microscopes, let alone more advanced technologies including laboratories for growing and testing cultures or rapid testing technologies," says Rachel Guglielmo, project director of Public Health Watch at the Open Society Institute. Such tests are especially difficult to implement for children, who have a difficult time coughing up sputum.

Keeping TB in check

Preventing the conversion from latent to active TB with isoniazid can reduce morbidity and improve survival. The WHO recommends that all individuals with latent infection, including people living with HIV/AIDS, take isoniazid for six to nine months, but more recent studies indicate that nine months to a year is more effective. In randomized controlled trials, isoniazid reduced the incidence of active TB by about 60% in HIV-infected patients with a positive skin test and 42% overall.

A more recent study sought to determine whether isoniazid could reduce active TB in a community setting. Roughly 700 HIV-infected South African miners were given isoniazid preventive therapy for six months, reducing active TB infection by 38% overall and by 46% in individuals not previously exposed to TB (JAMA 293, 2719, 2005).

A similar effect was seen in a randomized clinical trial involving about 250 HIV-infected South African children (median age 24.7 months). The group receiving isoniazid had a statistically significant lower incidence of TB (5 cases) than did the placebo group (13 cases) (BMJ 334, 136, 2007). The effect was so significant that the placebo arm was stopped and the study may have major public health implications, says Heather Zar, associate professor at the University of Cape Town, who conducted the study. "This could be recommended routinely for HIV-infected children who do not have access to antiretrovirals and who live in high TB-prevalence areas," she says.

But isoniazid preventive therapy is not as widely used as it could be. A study found that only 70% of Swiss physicians put their HIV-infected patients on isoniazid because of concerns over the possibility of a false-positive test, the drug's toxicity, or the development of drug resistance. In developing countries lack of funding and infrastructure compound the problem. Among individuals who comply with the treatment, protection wanes over time, especially in HIV-infected individuals. In areas with high rates of TB transmission, reinfection may be the reason for the failure of preventive therapy.

In addition to isoniazid, the WHO recommends the use of the antibiotic co-trimoxazole as a general preventive therapy for several secondary infections that target HIV-infected adults and children in sub-Saharan Africa. The WHO recommends the drug be given to children up to 18 months old born to HIV-infected mothers. Randomized controlled trials have found that co-trimoxazole preventive therapy reduces morbidity and mortality among HIV-infected smear-positive TB patients.

ARVs improve survival

Effective antiretroviral (ARV) therapy can dramatically improve the quality of life and survival time of HIV-infected individuals. The drugs' effective suppression of virus replication allows the immune system to rebound, and several studies have demonstrated that ARVs reduce the incidence of TB in HIV-infected people by greater than 80% (Int. J. Tuberc. Lung Dis. 4, 1026, 2000; Lancet 359, 2059, 2002; Clin. Infect. Dis. 34, 543, 2002). The effect is greatest among people with lower CD4+ T cell counts and those who start ARVs early in the course of their HIV infection. The WHO recommends that ARV therapy be offered to all HIV-positive TB patients if eligible.

A new study presented at the XVI International Conference on AIDS 2006 in Toronto shows that isoniazid plus ARVs may confer the best chance of preventing active TB disease in people co-infected with HIV and M. tuberculosis. The analysis of over 11,000 HIV-infected men and women in Rio de Janeiro found that isoniazid plus highly active antiretroviral therapy (HAART) is more effective than either therapy alone at preventing active TB disease—67% disease reduction among people treated with both drugs, while isoniazid or HAART alone reduced disease by 32% and 51% respectively.

The study is one of three ongoing projects by the Consortium to Respond Effectively to the AIDS/TB Epidemic, known as CREATE, led by Richard Chaisson of Johns Hopkins University. Chaisson is also looking at the question of how long HIV-infected individuals should take isoniazid. "We are doing a clinical trial funded by the National Institutes of Health that is looking at giving preventive treatment for an indefinite period of time to see if that is more effective in settings where there is more TB transmission," says Chaisson.

To address TB, some countries may have to revise their guidelines on when to start ARVs. TB occurs at higher CD4+ T cell counts than many other opportunistic diseases, around 250 cells/ml blood, which is higher than the cutoff of 200 where resource-poor countries start to prescribe ART.

Could ARVs exacerbate TB transmission? The possibility exists, since HIV-infected individuals who respond well to ARVs may remain highly susceptible to TB and remain healthier for longer, and therefore able to transmit and acquire TB for longer periods. Reducing this risk, a statistical model found, requires that ARV coverage start early, be comprehensive, and be combined with TB preventive treatment (AIDS 17, 2501, 2003; Science 301, 1535, 2003).

Extensively drug resistant TB

With XDR-TB now present in all regions of the world, global public health officials fear a deadly wave of TB that could spread first among HIV-infected individuals, and then within the general population. About 2% of all TB cases and 10% of multi-drug resistant TB (MDR-TB) cases are XDR-TB, defined by WHO as resistance to at least the two first-line TB drugs (rifampicin and isoniazid) in addition to any fluoroquinolone, and at least one of capreomycin, kanamycin, and amikacin. An estimated US$95 million is needed to stop the spread of XDR-TB, according to the WHO's Stop TB Department.

Identification of MDR-TB and XDR-TB in HIV-infected individuals is even more difficult than a simple TB diagnosis. Drug-resistant strains are usually diagnosed via sputum culture, yet smear-negative pulmonary TB and extrapulmonary TB are more common among HIV-infected individuals. The current tools are incapable of identifying MDR or XDR in a patient with extrapulmonary TB.

The success of programs to fight TB and HIV in a concerted manner rests ultimately on the availability and quality of services. Health officials in countries hit hardest by the HIV epidemic must adopt the WHO recommendations and work on coordination of testing for HIV and TB, counseling, and delivery of ARV and TB preventive therapies. While calling on donors to take a greater interest in funding these activities "it is particularly important to increase awareness about TB among people living with HIV, as a first step," says Guglielmo. A joint HIV-TB approach could range from referrals between services to the integration of HIV/AIDS and TB clinics. As the example of XDR-TB indicates, failure to implement HIV and TB prevention and treatment will only serve to worsen the toll on human lives.

*Catherine Zandonella, MPH, is a freelance writer whose work has appeared in Nature and New Scientist.