CDC Grand Rounds: the TB/HIV syndemic
The world now faces a situation in which approximately 160 persons die of TB each hour, in which a quarter of all deaths in persons with HIV/AIDS are caused by TB, and in which the evolution of the bacteria has outpaced the evolution of its treatment to such an extent that some forms of TB are now untreatable.
This is another in a series of occasional MMWR reports titled CDC Grand Rounds. These reports are based on grand rounds presentations at CDC on high-profile issues in public health science, practice, and policy. Information about CDC Grand Rounds is available at http://www.cdc.gov/about/grand-rounds.
Since Robert Koch's 1882 discovery of Mycobacterium tuberculosis, substantial progress has been made in tuberculosis (TB) control. Nevertheless, in the latter part of the 20th century, a long period of neglect of both quality program implementation and research led to persistently high TB incidence rates and failure to develop new tools to adequately address the problem. Today, most of the world continues to rely on the same diagnostic test invented by Koch approximately125 years ago and on drugs developed 40 years ago. The world now faces a situation in which approximately 160 persons die of TB each hour (1.45 million died in 2009), in which a quarter of all deaths in persons with human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) (PWHA) are caused by TB, and in which the evolution of the bacteria has outpaced the evolution of its treatment to such an extent that some forms of TB are now untreatable (1). More recently, renewed attention has been given to reducing the global burden of TB (2), but much remains to be done.
Misconceptions Regarding TB
Misconceptions about TB infection and disease impede patient care, program implementation, and policy innovation. The first misconception is that TB infection and TB disease are the same. For TB disease prevention and control purposes, the global population can be divided into three discrete groups: those without TB infection, those with TB infection, and those whose TB infection has developed into TB disease. The lifetime risk that a person with TB infection will develop TB disease is 5%–10%; that risk is much higher among PWHA (3,4). A successful control strategy must, therefore, address each group.
A second misconception about TB is that it is no longer a major public health problem. In fact, of the 7 billion persons in the world, 2.3 billion are already infected with TB, and about 9 million develop TB disease each year. Furthermore, TB causes about 1.4–2 million deaths annually (Figure 1) (1).
A third misconception is that TB can be diagnosed easily by a physician or laboratory. To diagnose TB infection, only two tests are validated currently: the tuberculin skin test (TST) and the interferon gamma blood test. Unfortunately, TST is neither sensitive nor specific for TB infection, and both tests can be difficult to implement in resource-limited settings. To diagnose TB disease, most laboratories examine sputum with a microscope to look for TB bacilli, the same approach that Koch invented. In PWHA, the sensitivity of microscopic examination is low, approximately 40% (5–7). Given the high risk for death in PWHA who have untreated TB, this low sensitivity is a critical challenge that must be addressed. Culture of sputum for M. tuberculosis is considered the gold standard test, but it is difficult to use and, in resource-limited settings, challenging to implement. Culturing M. tuberculosis, a slow-growing airborne pathogen, requires laboratories that employ high levels of biosafety and specialized technicians. In 2010, the Xpert MTB/Rif assay, a sensitive, easy-to-use, polymerase chain reaction (PCR)–based test was validated. With no need for sophisticated biosafety or specialized technicians and a turn-around time of 2 hours for both TB diagnosis and detection of drug resistance, this assay has the potential to improve TB control in the developing world (8). Limiting its current use is the relatively high cost of the necessary equipment and supplies, a lack of evidence that the assay's use is feasible in routine practice, and the fact that it has not yet been demonstrated to improve patient outcomes in resource-limited settings.
TB and HIV act synergistically within a population to cause excess morbidity and mortality. PWHA are more likely to develop TB disease because of their immunodeficiency; HIV infection is the most powerful risk factor for progressing from TB infection to disease (4). Diagnosing TB disease among PWHA is particularly challenging because PWHA who have pulmonary TB frequently have negative sputum smears and up to one third might have completely normal chest radiographs (5). Furthermore, TB in PWHA often occurs outside the lungs, evading traditional diagnostic tests. Because TB is both common and difficult to diagnose, many PWHA feel ill but are unaware that they have TB. A recent review found that when systematic efforts were undertaken to diagnose TB, approximately 8% of patients who went to HIV care and treatment facilities were found to have TB disease (9), although the exact proportion varies substantially depending on the epidemiology of TB in the area. Finally, TB is a frequent cause of death for PWHA, particularly if HIV disease is advanced and antiretroviral therapy (ART) has not yet been initiated. Persons with both diseases must adhere to complex drug regimens that might interact with each other and might have overlapping toxicities.
Combating the Dual Burden of Disease
TB disease and death can be prevented in PWHA by early TB diagnosis and effective treatment of both diseases. Early diagnosis and treatment ensure that TB treatment is provided before the illness reaches an advanced stage, thereby decreasing mortality, and ensures that the duration of infectiousness is limited, thereby reducing transmission of TB to others. TB disease also can be prevented by treating persons with TB infection. Treatment of TB infection requires reliably excluding the presence of TB disease to avoid the development of drug resistance; drug resistance could emerge if a patient receives a single drug to treat TB infection when the patient, in fact, requires a multidrug regimen to treat TB disease.
Until recently, no internationally accepted, evidence-based, sensitive approach existed to screen PWHA for TB disease, although some preliminary data had begun to suggest that commonly used approaches were inadequate. CDC investigators partnered with the U.S. Agency for International Development (USAID), ministries of health, and nongovernmental organizations in three Southeast Asian countries to derive a TB screening algorithm that would solve this problem. This study concluded that asking patients about three symptoms (i.e., cough, fever of any duration, or night sweats lasting longer than 3 weeks) accurately categorized PWHA for targeted interventions. Patients with none of these three symptoms can be considered free of TB disease and offered treatment to prevent TB disease, if indicated; patients with at least one of these symptoms should have further diagnostic tests performed for TB disease (5,6) These criteria mark a significant improvement over the 2007 World Health Organization (WHO) guidelines in which screening was based primarily on the presence of chronic cough (10). Screening for cough lasting more than 2 weeks was only 33% sensitive for TB disease in this study; screening for the combination of symptoms increased sensitivity to 93% (Figure 2) (5). The increased sensitivity under the new criteria will lead to fewer missed diagnoses of TB disease, at the cost of requiring TB diagnostic evaluation for more people.
Although this approach simplifies TB screening, a comparable approach for simplifying diagnosis of TB disease remains elusive. In the same study, investigators learned that adding liquid culture of two sputum specimens more than doubled the yield of TB case detection among PWHA, compared with microscopic examination alone of the same two sputum specimens, as recommended by WHO at the time (76% versus 31% sensitivity) (6). Unfortunately, liquid culture is not widely available in resource-poor settings and requires high levels of training, biosafety, and supervision. It is hoped that introduction of the Xpert MTB/Rif assay, which is more sensitive than smear but less sensitive than liquid culture, along with other emerging diagnostic techniques, will improve diagnostic accuracy in PWHA who have symptoms of TB (8).
In persons who screen negative for TB disease, treatment of TB infection should be considered. The tuberculin skin test (TST) identifies persons with TB infection who can benefit from isoniazid preventive therapy (IPT), a regimen that involves ingesting isoniazid daily for at least 6 months. In the pre-ART era, clinical trials confirmed that IPT was effective in reducing the development of TB disease in TST-positive PWHA by 64% (11). Subsequently, in 1998, WHO recommended that all PWHA living in TB-endemic countries receive 6 months of IPT, and that TST screening generally was not needed in countries with a high burden of TB. Follow-up studies found that the benefit of IPT waned as early as 6 months after completion of IPT. In 2009, only 0.3% of PWHA globally received IPT (1). ART also can reduce the risk for TB disease in PWHA by 54%–92% and might have a synergistic effect when used with IPT (12). In collaboration with the Botswana Ministry of Health, and with funding from CDC and USAID, CDC conducted a clinical trial in Botswana to evaluate how much better TB could be prevented with a 36-month regimen of IPT in PWHA who had access to government-provided ART. This study found that among those with positive TSTs, 36 months of IPT reduced TB incidence by 74%, compared with persons receiving only 6 months IPT. When the analysis was limited to TST-positive trial participants randomized to the 36-month IPT arm who successfully completed the initial 6 months of IPT, the reduction in TB was 92%. As with previous studies, no significant benefit from IPT was observed for TST-negative participants (Figure 3). ART provided an added benefit to IPT's protective effect, reducing TB risk a further 50% in all groups (13).
These findings have enormous implications for controlling the TB epidemic in countries with a high burden of HIV. If 36 months of IPT were provided to all TST-positive PWHA in Botswana, countrywide TB incidence would decline 45%† (Figure 4). A cost-effectiveness model of 10,000 PWHA in Botswana demonstrated that providing 36 months of IPT for PWHA with a positive TST result, in addition to ART for those with CD4 <250 cells/µL, could avert more incident TB cases with fewer resources than increasing the threshold for ART initiation alone (CD4 <350 or 500), suggesting any cost-effective TB prevention strategy should include the provision of IPT for TST-positive PWHA.
From Evidence to Guidance to Global TB Control
The strong evidence provided by the studies described above has been combined with results from other studies to update the global guidelines for TB screening and prevention (14). A recent WHO publication outlines four updated recommendations for resource-constrained settings: 1) PWHA should be screened with the new symptom-based algorithm, and those who do not report current cough, fever, weight loss, or night sweats are unlikely to have active TB and should be offered IPT (a minor modification to the algorithm developed in the CDC Southeast Asia study); 2) PWHA who report any of the aforementioned symptoms are considered suspects for TB disease and should be evaluated further for TB and other diseases as clinically indicated; 3) PWHA who are TST positive or have unknown TST status and are unlikely to have TB disease based on symptom screening should receive IPT for at least 6 months; and 4) in settings where feasible, PWHA should receive IPT for at least 36 months, or even lifelong. Where feasible, TST should be used to help identify those who would benefit most from IPT (15).
TB control relies on an international strategy known as "DOTS" (directly observed treatment, short course) that includes finding as many highly infectious patients with TB as possible, initiating effective treatment, directly observing drug ingestion to ensure adherence, and standardized monitoring, evaluation, and reporting. DOTS has saved approximately 7 million lives globally since 1990 (1). In the United States, the experience in New York City provides an example of the progress that can be made through full implementation of the DOTS strategy (16). However, although TB prevalence and deaths around the world did fall in the period after widespread global DOTS implementation, treatment programs generally have not resulted in a rapid reduction in global TB incidence (17). Multiple factors explain this phenomenon: insufficient resources and commitment to implement DOTS, in part because TB occurs predominantly in the poorest populations; a focus entirely on treatment of TB disease but not TB infection; the HIV epidemic; the emergence of multidrug resistant TB strains; and limited attention to the social determinants of sustained TB transmission and reactivation. Modeling studies suggest that detecting more infectious TB cases and successfully treating them will, on its own, be insufficient to drive down TB incidence and prevalence quickly and that the global TB strategy must address the large burden of latent TB infection that exists globally (18). The simplified symptom-based screening approach derived in the Southeast Asian study and the effective approach to chemoprophylaxis documented in the Botswana clinical trial help address this need.
The Way Forward
In a 2010 "call to action," global leaders in TB control outlined crucial areas that must be addressed to accelerate the decline in global TB incidence to more than 1% per year and to meet the target for the 2015 Millennium Development Goal (Figure 5) (19). Achieving this will require fully implementing the DOTS strategy globally, and it will also require going far beyond that to address the limited impact that would be expected with DOTS alone, as outlined in WHO's latest STOP TB strategy (20). WHO calls for improvements in TB screening and diagnosis, including the use of newer TB diagnostic assays. In addition to these steps, treatment of latent TB infection also is needed (18). In settings with a high prevalence of HIV infection, implementing IPT can reduce TB incidence greatly. Finally, scientific advances are needed in three key areas to develop 1) an effective TB vaccine; 2) a shorter, simpler anti-TB drug regimen with efficacy against both drug-susceptible and drug-resistant TB; and 3) new diagnostic tests that can simply and accurately diagnose both TB infection and disease (21).
The fundamentals of TB control are early and accurate TB diagnosis, effective treatment, and prevention. The gap between what we know and what we need to know is large, but the gap between what we know and what we are implementing in practice is both larger and more harmful. By closing both our knowledge gap and our implementation gap, we can eliminate this deadly syndemic.
Haileyesus Getahun, MD, PhD, Mario Raviglione, MD, World Health Organization, Geneva, Switzerland. Jay K. Varma, MD, Global Disease Detection Br, Center for Global Health; Kevin Cain, Div of Tuberculosis Elimination, Taraz Samandari, Div of HIV/AIDS Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention; Tanja Popovic, MD, PhD, Thomas Frieden, MD, Office of the Director, CDC. Corresponding contributor: Kevin Cain, email@example.com, 404-639-2247.
* Additional information available at http://www.cdc.gov/nchhstp/programintegration/definitions.htm.
† Assuming provision of antiretroviral therapy to all PWHA if CD4 <200 cells/µL.