Tuberculosis is the leading cause of death among people with HIV globally, accounting for almost 25% of all HIV deaths in 2008 (WHO, 2009i). The risk of acquiring TB is 21-34 times greater among people living with HIV than in the general population. In 2010, of 8.8 million incident TB cases worldwide, 1.1 million were among people living with HIV, with an estimated 350,000 deaths (WHO, 2011f). More than half a million women of child-bearing age die from TB each year (STOP TB Partnership, 2011).
"TB continues to be the leading cause of death among people living with HIV" (UNAIDS, 2009e: 3).In some countries in sub-Saharan Africa, up to 70% of people living with TB are also living with HIV (WHO et al., 2012d). Sub-Saharan Africa continues to account for the majority of people living with HIV and TB in the world with about 82% of the estimated total people living with HIV and TB in 2010. South East Asia, mainly India, accounts for 13% of the remaining cases. However, in 2010, only 34% of TB patients globally knew their HIV status and only 5% of people living with HIV were screened for TB (WHO, 2011f).
HIV Infection Fuels TB Epidemics
When an HIV epidemic is in an expansive phase, HIV and TB co-infection rates have rapidly increased. For example, HIV seroprevalence among TB cases in Chiang Mai, Thailand increased from 5% in 1989 to 40% in 1992, along with the rapidly growing HIV epidemic (Payanandana et al., 1995 cited in Raviglione et al., 1996; Kharsany et al., 2006). "The risk of TB increases with advancing immunodeficiency, so as the HIV epidemic in a community matures, the burden of HIV-associated TB may be expected to increase, even after the prevalence of HIV infection has stabilized" (Lawn et al., 2006: 1046).
TB Is a Serious Risk for Those Living with HIV
Not everyone exposed to TB has active disease. A person with latent TB infection, or LTBI, has been infected with the TB bacillus but has an immune system sufficiently intact to control the infection and will not permit the bacillus to cause disease. A person with LTBI is not ill and is not infectious. TB becomes a much more serious problem for someone with HIV. When a person infected with the TB bacillus cannot control the infection because of a compromised immune system, the bacillus is able to multiply so that there are millions of TB bacilli that then cause disease. A person with active TB disease becomes sick and is considered infectious to others.
Individuals living with HIV are up to 50% more likely to develop TB in than those not living with HIV (WHO, 2012d). Among individuals with latent TB, HIV is the strongest known risk factor for progressing to active TB (CDC, 2012b). In some sub-Saharan African countries, 70% of patients infected with TB also have HIV. Without proper treatment, nearly 90% of those living with HIV die within months of contracting TB and nearly 25% of deaths among people living with HIV are due to TB (WHO, 2012d). People whose immune systems have deteriorated due to advanced HIV disease are at greatly increased risk of developing active TB disease from a previously contained latent infection. Some immunocompromised people may not be able to contain a new TB infection and will immediately progress to active disease upon exposure to TB. Additionally people with HIV are more likely to develop extra-pulmonary TB (TB outside of the lungs) that may involve multiple organs and is harder to diagnose (Zvandasara et al., 2006).
Incidence of TB has fallen or stabilized in many developed countries in the last thirty years; however "not in Eastern Europe or countries of the former Soviet Union" (Altice et al., 2010: 374) where TB has emerged as a leading cause of morbidity and mortality in HIV-infected drug users.
Progress in TB Testing and Treatment is Being Made
Globally, HIV testing among TB patients has made progress with the number of notified TB cases with a known HIV status increasing from 16% in 2007 to 34% in 2010. In Africa, testing increased from 37% in 2007 to 59% in 2010 among notified TB cases (WHO, 2011f).
A study in Cameroon found that some patients preferred to treat their immediately life-threatening TB before testing for HIV to "become physically/mentally strong before facing the challenges of testing for HIV" (Njozing et al., 2010: 30). However, treating HIV at earlier stages, as now recommended by WHO, may reduce the incidence of TB/HIV co-infection (Gopal and van der Horst, 2010). Preliminary results from a recent study found that a household HIV counseling intervention reduced TB prevalence by 22% and increased HIV testing (Ayles et al., 2012), but further details on the intervention are awaited.
New technological advances, such as the GeneXpert MTB/RIF assay, may also contribute to progress with new diagnostics on the horizon. Although Xpert may miss some active TB cases among those with advanced immunosuppression (Theron et al., 2011 cited in Grant et al., 2011) and costs may be a concern (Grant et al., 2011; Abimbola et al., 2012 cited in Smart, 2012b), the technology is an improvement. Prior to the development of Xpert, the most widely used test to detect TB -- smear microscopy -- was 125 years old and routinely missed half of all cases (Small and Pai, 2010). Countries will, however, continue to require adequate laboratory services for microscopy and culture to detect drug resistance (WHO, 2011f).
Treatment Adherence Is Critical to Curing TB and Reducing the Spread of Drug-Resistant Strains
Adherence to the full course of treatment--six months for first-line treatment--is essential to cure TB and avert the development of multidrug-resistant TB (MDR-TB) and extensively drug-resistant TB (XDR-TB). MDR-TB (where the disease has developed resistance to two of the first-line TB drugs, isoniazid and rifampicin, requires between 18-24 months of often complicated and expensive combination therapy. Of the 12 million estimated prevalent TB cases in 2010, it is estimated that 650,000 cases were multi-drug resistant TB. Treatment success varies by country, ranging from 23% in South Africa to 74% in Kazakhstan WHO, 2011f). XDR-TB is resistant to first and second-line drugs and significant efforts are often needed to identify a therapeutic regimen that works and to manage side effects. By the end of 2011, 77 countries had reported at least one case of XDR-TB. Globally, an estimated 9% of MDR-TB cases have XDR-TB (WHO, 2012e). WHO updated the tuberculosis treatment guidelines in 2010, confirming its prior recommendation of drug susceptibility testing (DST) at the start of all therapy for previously treated patients in order to find and treat MDR-TB; it also addresses the prevention of acquired MDR-TB, where new TB patients have isoniazid-resistant TB when they begin treatment (WHO, 2010b).
For people who are living with HIV, MDR-TB or XDR-TB co-infection results in high levels of early mortality. A retrospective observational study in Tugela Ferry, South Africa of 272 MDR-TB and 382 XDR-TB cases, of which 90% and 98% respectively were HIV co-infected, found that mortality at one year was 71% for MDR-TB and 83% for XDR-TB patients. The majority of deaths occurred within the first thirty days of sputum collection, with mortality rates worsening with greater degrees of drug resistance (Gandhi et al., 2009).
The high rate of mortality within the first thirty days is of concern considering it typically takes 6 to 8 weeks to diagnose drug-resistant TB by traditional culture and drug-susceptibility testing methods. "The majority of patients do not survive long enough to receive their drug-resistant TB diagnosis and to initiate treatment" (Gandhi et al., 2010: 83). Of note, "forty-three percent of MDR-TB cases and 56% of XDR-TB cases were women" (Gandhi et al., 2010: 81). However, a global review of MDR-TB did not find an association between MDR-TB and HIV, nor any statistically significant different rate of MDR-TB by sex (Zignol et al., 2011).
It is important to note that drug-resistant TB can be transmitted. "Overcrowded, poorly ventilated clinics that bring together large numbers of HIV-infected persons, some with active TB, will be a recipe for disaster" (IOM, 2005: 107). Therefore in health care facilities where people with HIV and/or TB access screening, treatment and support, and where TB transmission may be most efficient, measures should be taken to reduce the risk of primary transmission of TB, both drug-resistant and susceptible. Rapid drug susceptibility testing, prompt initiation of effective TB treatment and implementation of infection control measures such as separation of TB suspects, improving patient flow and increasing ventilation may reduce the risk of TB transmission in health care settings.
There Are Gendered Dynamics in TB Prevalence, Detection
"Traditionally, the majority of TB cases were reported in men, but the global HIV epidemic induced major changes in TB epidemiology. The preponderance of women living with HIV (women account for up to 70% of adults living with HIV in areas were heterosexual HIV transmission is dominant) may explain why more women than men receive a diagnosis of TB in countries where HIV infections prevalence is high" (Marais, 2011: 304). "By the end of 2007, relatively more women had TB detected than men in countries with a prevalence of HIV infection 1%" (Getahun et al., 2010: S204). A study in Rwanda found that even though the majority of adult TB cases reported to the surveillance system were male (60%), for women with smear-positive pulmonary TB, the risk of death was twice as great as the risk among men and more women were co-infected with HIV (Uwizeye et al., 2011).
A summary of the Stop TB symposium on the needs of pregnant women and children living with TB stated there is a considerable burden of TB among women and children. In 2010, there were approximately 3.2 million new TB cases among women, with 320,000 TB deaths among HIV-negative women and about 500,000 TB deaths occurred among women living with HIV (Smart, 2012a). In South Africa, the region with the highest global HIV prevalence, women aged 15-24 have rates of TB 1.5-2 times greater than men of the same age, and the pattern is consistent across each of the countries in the region (Deluca et al., 2009 cited in Smart 2012a), although "there are problems reliably estimating the burden of TB disease among women, because about 30% of countries are not disaggregating cases by sex" (Smart, 2012a: 3).
In 2010, 64% of TB cases were among men compared to 36% among women (WHO, 2011f). The reasons for the higher global TB notification rates in men are not well understood and could result from a variety of biological or environmental factors such as the likelihood of producing a positive sputum sample, delays in health-seeking behavior, gendered dynamics within the family, stigma and access to care: women are less likely to produce positive sputum samples and more likely to have extra-pulmonary TB and/or be co-infected with HIV (Lawson et al., 2008; Karim et al., 2008; Sreeramareddy et al., 2008).
Preliminary data suggest that the implementation of the revised WHO case definition of smear-positive TB was associated with significant increases in case detection among women in Kenya (Ramsay et al., 2009). The revised guidance lowers the number of bacilli detected in a sample and reduces the number of reported smear-positive results from two to one required be classified as a TB case. Evidence from a Vietnamese study also suggests that women are slower to progress to smear-positive disease despite similar time from symptom onset to diagnosis as men (Thorson et al., 2007). Additionally, some studies have shown that physicians are less likely to conduct TB exams on women than men (Begun et al., 2001 cited in Theobald et al., 2006). However, a recent study from Thailand with 480 newly diagnosed TB patients of whom 86 were living with HIV, found very small differences between men and women in delaying a visit to a provider for TB, with a short median delay of 26 days (Pungrassami et al., 2010). Further research is needed to understand the role of sex in TB-HIV co-infection (Nakanjako et al., 2010).
Co-Infection is Particularly Deadly for Women During Their Childbearing Years
A summary of the Stop TB symposium on the needs of pregnant women and children living with TB stated that up to 11% of pregnant women living with HIV have TB and there is an increased risk of transmitting TB and HIV to infants born of mothers with TB/HIV co-infection (Mepham et al., 2011; Smart, 2012a). The U.K. Department for International Development (DFID) recommends integration of strategies to address TB/HIV co-infection in maternal and reproductive health services (UKAIDS, 2012). Women bear the greatest burden of HIV during their childbearing years, and similarly, the greatest burden of TB during those years as well. It is estimated that 15% of maternal deaths are among women co-infected with TB and HIV (Adhikari, 2009; Klotz et al., 2007; Mofenson and Laughton, 2007; Zwang et al., 2007; Ramogale et al., 2007; Gupta et al., 2011). The development of TB disease is associated with a four-fold increase in AIDS-related deaths among women co-infected with TB and HIV (Lopez-Gatell et al., 2007). A study in South Africa found young women to be at particular risk. Epidemiologic changes in TB notifications and the prevalence of HIV infection from 1996 to 2004 in a peri-urban community in South Africa of 13,000 found that annual TB notification rates among adolescents increased from zero cases between 1996-1997 to 436 cases per 100,000 in 2003-2004, with TB cases predominantly among female adolescents (Lawn et al., 2006).
The U.S. Centers for Disease Control and Prevention states that TB treatment for pregnant women should be the same as for nonpregnant women, but with special considerations for particular medications' effect on both the woman and the fetus (http://www.aidsinfo.nih.gov/Guidelines/GuidelineDetail.aspx?MenuItem=GuidelinesSearch=OffGuidelineID=211ClassID=4 (CDC, 2009a)). "Pregnant women on ART who have a diagnosis of active TB should have their ARV regimens adjusted as needed to accommodate their TB drugs. For women whose diagnosis includes concurrent active TB and HIV infection during pregnancy, TB therapy should be initiated immediately and ART should be initiated as soon as possible thereafter, usually according to the principles described for nonpregnant adults" (CDC, 2009a). Co-infection with TB can also require changes in ARV regimens for those living with HIV. A cohort study in Cte d'Ivoire of 2,012 adults living with HIV who started ART between 2004 and 2006 examined factors affecting changes in treatment and found that co-infection with tuberculosis, along with pregnancy and drug intolerance were the most common factors requiring drug modifications (Messou et al., 2010).
Because of the increased risk of maternal and infant mortality associated with TB and HIV co-infection during pregnancy and postpartum, there is an urgent need to implement TB screening as part of routine antenatal and postpartum care as well as treatment for latent and active TB for women (Mofenson and Laughton, 2007). Maternal and child health and HIV/AIDS prevention programs that include TB education and screening make these services more accessible to women of childbearing age. The NIH-funded IMPAACT trials network will be conducting a randomized trial, the TB Apprise Trial (IMPAACT P1078) looking at the safety and timing of isoniazid preventative therapy in pregnancy and postpartum period -- to provide some additional evidence on the risks and benefits of isoniazid preventative therapy in pregnant women (Smart, 2012a). There is also a need for additional research on MDR-TB in pregnancy, as there are less than 100 case reports (Gupta, 2009 cited in Smart, 2012a). Very few TB control programs have "successfully carried out isoniazid preventative therapy on a large scale" (Dye, 2011: 2230), due to yearlong treatment needed as well as other issues (Dye, 2011). However, new preventive therapies are under development, requiring larger studies to assess impact (Sterling et al., 2011 cited in Dye, 2011).
For women with both HIV and TB who wish to avoid an unintended pregnancy, potential drug reactions with anti-TB therapy "can make the management of hormonal contraception more challenging" (McCall and Vicol, 2011: 196). [See Meeting the Sexual and Reproductive Health Needs of Women Living With HIV]
Active Case Finding Is Necessary to Increase TB Detection
Only about half of TB suspects seek out TB screening, and it is estimated that about half of these cases are misdiagnosed (Ayles, 2009). "Sputum smear microscopy, the tool for TB screening in most resource-limited settings, has low sensitivity to diagnose TB disease, especially in HIV-infected patients with advanced immunodeficiency" (Colebunders and Bastian, 2003, Siddiqi et al., 2003 and Kibiki et al., 2007 cited in Worodria et al., 2010). Many of the TB control strategies like passive case finding [as opposed to active case finding where health workers actively screen people for TB symptoms] and directly observed therapy (DOT), that are used today were developed in the pre-HIV era, and "[do] not take into account the profound impact of HIV on tuberculosis incidence" (Reid et al., 2006: 485). This situation is compounded by the fact that TB is more difficult to diagnose in people with HIV-related immune suppression.
Active case finding increases TB detection, particularly in sub-Saharan Africa, where HIV is driving the epidemic. However, as noted previously, TB symptoms may be different in people living with HIV and some people may be asymptomatic, which can complicate diagnosis. A prospective cohort study with 1,768 patients living with HIV from eight clinics in Cambodia, Vietnam and Thailand found that TB screening that includes questions about a combination of TB symptoms such as fatigue, fever and weight loss was significantly more effective in ruling out TB than asking about cough alone (Cain et al., 2010).
"Because dual infection with HIV and tuberculosis poses a life-threatening diagnostic and therapeutic dilemma... HIV programs must include capabilities for diagnosis, treatment, and prophylaxis of tuberculosis. Tuberculosis treatment programs should be supported as an important point of entry for HIV testing and consideration for HAART. It is critical to overall treatment success that these coexisting epidemics be addressed in parallel" (IOM, 2005: 6). It is estimated that half a million people with HIV/AIDS could be reached through existing TB programs (Kim, 2004 cited in IOM, 2005: 103). A modeling study based on nine-countries in sub-Saharan Africa found that if HIV-positive patients were started on ART within five years of seroconversion, the incidence of AIDS-related TB would be reduced by 48% (Williams et al., 2010b). "If ART was started even earlier, within one year of seroconversion, the direct effect on HIV-related tuberculosis would not be much greater, but such treatment would be expected to reduce HIV transmission substantially, resulting in fewer people infected with HIV and thereby greatly reducing the long-term risk of HIV-associated tuberculosis" (Harries et al., 2010b: 1909).
Guidelines Aim to Strengthen the Response to HIV/TB Co-Infection
WHO has released new guidance recommending twelve activities that should be carried out by the health sector response to HIV/AIDS, which focus on the intersection of the TB and HIV epidemics. The twelve activities aim to attain three overarching goals: 1) Establish and strengthen the mechanisms for delivering integrated TB and HIV services; 2) Reduce the burden of TB in people living with HIV and initiate early antiretroviral therapy through: a) isoniazid preventative treatment; b) intensified case finding for active TB; and c) TB infection control; and 3) Reduce the burden of HIV in patients with presumptive and diagnosed TB. The main elements of the new policy focus on routine HIV testing for TB patients, provision of co-trimoxazole for all HIV-TB co-infected patients, and starting all TB patients with HIV on ART as soon as possible. For a research agenda, please see Sculier et al. (Sculier et al., 2011).
Despite overwhelming evidence, many public sector health programs have failed to implement the activities that address reducing the burden of TB and HIV in populations affected by both diseases (Dong et al., 2007; WHO, 2009i). In many countries, the national TB program and the national AIDS control program run as parallel systems without a mechanism to link to one another (Reid et al., 2006; Williams et al., 2008). HIV programs play a vital role in identifying those with TB and interrupting transmission through active TB case finding and implementing infection control measures (Reid et al., 2006). Unfortunately, "delays with the scale-up of antiretroviral treatment have exacerbated the tuberculosis epidemic... and thousands of preventable deaths" (Chopra et al., 2009c: 3).
There is very little sex-disaggregated data on TB/HIV that is not pregnancy-related. As a result, there are a number of research and program gaps related to what works for women who are living with both HIV and TB.