Tuberculosis (TB) is one of the most important global infectious diseases, with an estimated 9.0 million global cases in 2013 and 1.5 million deaths. Among these 9 million cases 1.1 (13%) involved HIV-infected persons and an estimated 360,000 TB-related deaths occurred in HIV-infected persons. In the United States, tuberculosis remains a significant cause of morbidity in persons with HIV infection. In recent years, cases of tuberculosis in the United States general population have declined from 15,078 cases reported in 2002 to an estimated 9588 cases in 2013.[,,] Among the United States cases of tuberculosis, approximately 7% occur in persons who are also infected with HIV.[,] In the United States, an estimated 4% of the general population has latent tuberculosis. The transformation of latent tuberculosis to active tuberculosis occurs in approximately 3 to 16% of HIV-infected persons per year (if they are not on antiretroviral therapy), which is markedly higher than the 5% lifetime risk in persons without HIV infection.[,] Although the widespread use of antiretroviral therapy has significantly lowered the incidence TB in HIV-infected persons in the United States, the risk in persons on antiretroviral therapy still remains higher than with persons without HIV infection.
Testing for Latent Tuberculosis Infection
All HIV-infected individuals, regardless of history of exposure to TB, should undergo testing for latent tuberculosis as soon as possible after the diagnosis of HIV infection is made, unless they have recently had testing for latent tuberculosis. Two methods are used to diagnose latent tuberculosis: tuberculin skin testing (TST) and interferon gamma release assay (IGRA).[,,] Annual testing for latent tuberculosis testing is recommended for HIV-infected individuals who test negative for latent tuberculosis and remain in a high-risk category for exposure to M. tuberculosis, including persons incarcerated, those actively using drugs, person who live in congregate settings, or those who have other sociodemographic risk factors for tuberculosis. Practitioners should also repeat testing for latent tuberculosis in patients with advanced HIV disease (CD4 count less than 200 cells/mm3) who respond favorably to antiretroviral therapy and increase their CD4 count to greater than 200 cells/mm3, since the likelihood or response to testing for latent tuberculosis increases with improved cellular immunity. The tuberculin skin test and IGRAs can not distinguish latent tuberculosis from active tuberculosis.
Tuberculin Skin Testing
The traditional testing for latent infection with Mycobacterium tuberculosis is performed by the tuberculin skin test method, which consists of an intracutaneous injection of 0.1 ml of intermediate strength purified protein derivative (using the Mantoux method) (Figure 1). The site should be evaluated 48 to 72 hours later, and the diameter of induration, not erythema, should be measured (Figure 2) and (Figure 3). The tuberculin skin test evaluates the delayed-type hypersensitivity (type IV) reaction to the purified protein derivative antigens (tuberculins). For HIV-infected individuals, a positive tuberculin skin test is defined as at least 5 mm of induration 48 to 72 hours after injecting the PPD. The PPD contains some antigens that are in nontuberculous mycobacteria and in the Bacille Calmette-Guerin (BCG) vaccine; thus, false-positive test results can result from a patient's prior contact with nontuberculous mycobacteria or from receipt of BCG. The estimated interval between M. tuberculosis infection and reactivity of the tuberculin skin test is 8 weeks (range 2 to 12 weeks); this period is referred to as the window period. Overall, for the diagnosis of latent tuberculosis, the tuberculin skin test has an estimated sensitivity of 65 to 75% and a specificity of 56 to 95%.[,] Anergy is suggested by a response to the tuberculin skin test of less than 5 mm and a response to either mumps antigen or tetanus toxoid of less than 5 mm. A randomized, placebo-controlled, double-blind trial of isoniazid (INH) for the prevention of TB among anergic HIV-infected persons in the United States demonstrated no significant decrease in the rate of development of active TB, progression of HIV disease, or death (Figure 4). On the basis of this study, other studies performed internationally, and a meta-analysis of the available trials, anergy testing is not recommended as part of TB skin testing for HIV-infected individuals.[,,] The role of two-stage tuberculin skin testing (repeating the tuberculin skin test one to five weeks after an initial negative test) among HIV-infected individuals remains undefined and is not recommended in HIV-infected persons. In recent years, the tuberculin skin testing method has become less frequently used due to several problematic features, including the patient requirement to return for a second visit in 48 to 72 hours to have the test read, false-positive results in persons who have received BCG, and false-negatives in persons with advanced immunosuppression.
Interferon Gamma Release Assay
As a result of the logistical problems when performing TST and specificity issues with tuberculin skin testing, several IGRAs have been developed as an alternative method of testing for latent tuberculosis infection.[,,] As of March 2015, two FDA-approved IGRA assays were available for use in the United States to diagnose latent tuberculosis: QuantiFERON-TB Gold In-Tube test (QFT-GIT) or T-SPOT.TB test (T-Spot) (Figure 5).[,] With these assays, a patient's blood sample is obtained, incubated with specific mycobacteria antigens, and the release of gamma interferon from white blood cells is measured with an enzyme-linked immunoassay. The mycobacterial peptides used in these assays are not present in the BCG vaccine or in common nontuberculous mycobacteria, and the test is not affected by prior infection with most nontuberculous mycobacteria or receipt of BCG vaccine. In addition, because the IGRA involves testing of a blood sample without exposing the patient to any mycobacterial antigens, no boosting phenomenon occurs in patients tested with an IGRA. The FDA has established interpretation criteria for the QuantiFERON-TB Gold In-Tube test (QFT-GIT) (Figure 6) and and the T-SPOT.TB test (T-Spot) (Figure 7). For the diagnosis of latent tuberculosis in the general population, the IGRAs have an estimated sensitivity of approximately 75 to 88% and a specificity of 92 to 97%. Available data suggest similar performance of the IGRAs in HIV-infected persons, except perhaps a lower sensitivity in patients with low CD4 cell counts.[,,] Patients with HIV infection can have reversion of the IGRA test from positive to negative.
Management of Tuberculosis Exposures
The risk of developing TB from close (prolonged, frequent, or intense) contact with a person who has active TB is increased risk of developing TB if they are HIV-infected. Thus, it is imperative to evaluate HIV-infected individuals for active TB if they have experienced close contact with a TB patient. All HIV-infected individuals who have had close contact with someone with active TB should receive treatment for latent TB infection once active TB is excluded. In this setting, results of the tuberculin skin test results do not alter the recommendation to treat for latent TB infection.
Evaluation of Patient with Presumptive Latent Tuberculosis
All HIV-infected individuals who have a positive tuberculin skin test or a positive IGRA should undergo evaluation for active TB, including a symptom review (cough, fever, and weight loss), and a chest radiograph. In addition, sputum samples stained and cultured for acid-fast bacilli (AFB) should be obtained if the initial evaluation reveals symptoms suggestive of TB, or if the chest radiograph is abnormal. Obtaining sputum samples in a symptomatic patient who has a normal chest radiograph is recommended because some HIV-infected individuals with active TB may have a negative chest radiograph. In certain circumstances, a patient may have a fibrotic lesion consistent with TB noted on a chest radiograph obtained for purposes other than evaluation of TB. Unless the patient has a history of previously fully treated TB, the incidental finding of a fibrotic lesion should warrant evaluation for latent tuberculosis infection, as well as for active TB, even if the patient has no symptoms to suggest active TB. If no evidence for active TB exists, the patient with who a fibrotic lesion seen on chest radiograph is presumed to have latent TB, regardless of results of the IGRA or tuberculin skin test for latent TB test. Asymptomatic patients who have a normal chest radiograph should not have sputum samples sent for mycobacterial culture.
Indications and Benefit for Treatment of Latent Tuberculosis Infection
According to the 2013 guidelines for the prevention and treatment of opportunistic infections, HIV-infected persons, regardless of age, should receive treatment for latent tuberculosis infection and they have any one of the following: (1) a positive diagnostic test for latent tuberculosis infection and no prior history of treatment for active or latent TB; (2) a negative diagnostic test for latent tuberculosis infection but close contact has occurred with a person who has infectious pulmonary TB; or (3) a history of untreated or inadequately treated healed TB (as suggested by old fibrotic lesions on chest radiography) regardless of diagnostic tests for latent tuberculosis infection.  The risk of progression to active TB from latent TB infection in HIV-infected individuals varies by geographic location, but a multi-center prospective cohort study in the United States found an overall rate of progression of 0.7 cases per 100 person-years. Among those persons who were tuberculin skin test positive at baseline, the risk was 4.5 cases of active TB per 100 person-years. The risk was higher among people who converted to being tuberculin skin test positive at 5.4 cases per 100 person-years. These numbers contrast sharply with the low risk of development of TB in tuberculin skin test negative HIV-infected individuals of 0.4 cases per 100 person-years (Figure 8). Injection drug use increases the risk further.. Treatment of latent TB significantly reduces the risk of progression to active TB. Based on a meta-analysis of international randomized controlled trials, treatment of latent TB infection with isoniazid among HIV-infected persons reduces the risk of developing active TB by approximately 60%. The risk of death was unaffected by treatment of latent TB infection in patients with HIV, regardless of tuberculin skin test results.
Treatment of Latent Tuberculosis Infection
In the 2013 guidelines for the prevention and treatment of opportunistic infections provides specific recommendations for the indications for the treatment of latent tuberculosis, the preferred regimens, and alternative regimens (Figure 9). The preferred treatment of latent TB infection consists of a 9-month regimen of isoniazid on a daily basis or isoniazid twice weekly (given by directly observed therapy); patients taking isoniazid should also receive pyridoxine (vitamin B6) 25 mg once daily to reduce the risk of peripheral neuropathy. Isoniazid does not have significant interactions with antiretroviral medications, but is associated with a risk of hepatitis, which increases with age, daily alcohol consumption, and pregnancy. Medical providers should inform all patients taking isoniazid about the symptoms and signs of hepatitis, such as abdominal pain, vomiting, and jaundice, and instruct them to promptly report any of these symptoms. Patients receiving isoniazid should undergo monthly visits to evaluate for adherence, side effects, and signs and symptoms of hepatitis or neuropathy. Baseline testing of hepatic aminotransferase levels is indicated for all HIV-infected persons. For patients at increased risk for developing hepatotoxicity, such as individuals with abnormal baseline tests, a history of hepatitis B or C virus infection, or those concomitantly receiving antiretroviral therapy, routine laboratory monitoring is indicated.[,] Patient exposed to drug-resistant TB, expert consultation should be obtained to select an appropriate regimen for treatment of latent tuberculosis. For patients who can not tolerate isoniazid, or who are exposed to a person with known isoniazid-resistant TB, the alternative treatment for latent tuberculosis infection consists of a 4-month course of rifampin or rifabutin. The use of rifampin or rifabutin can be complicated by major drug-drug interactions. In general, consultation with the public health tuberculosis authorities should be obtained in any patient exposed to an index case with known drug-resistant TB. The short 2-month regimen of daily rifampin plus pyrazinamide for the treatment of latent TB infection is not recommended due to the unacceptably high rate of severe liver injury resulting in hospitalization (3.0/1000 treatment initiations) and death (0.9/1000 treatment initiations).[,]