Assessment of Risk for Tuberculosis in HIV-Infected Persons
Providers should maintain a high index of suspicion for active tuberculosis (TB) in any HIV-infected person, especially those with risk factors for acquiring Mycobacterium tuberculosis infection. These risk factors include birth or residence in an area endemic for TB infection, homelessness, injection drug use, and residence in a nursing home or correctional facility. Individuals with HIV infection are more likely to develop active TB if infected, but not more likely to transmit M. tuberculosis to others. Some HIV-infected patients, especially those with more advanced immunosuppression, develop TB rapidly after becoming infected with M. tuberculosis, with 37% developing active TB within 6 months of primary TB infection. In general, HIV-infected patients with a positive tuberculin test will develop active TB at a rate of approximately 8 to 10% per year.
Clinical Presentation of Pulmonary TB
In a large retrospective study of TB in HIV-infected individuals, 38% had isolated pulmonary involvement, 32% had pulmonary and extrapulmonary involvement, and 30% had disease only at extrapulmonary sites. This distribution contrasts with TB in HIV-negative patients, which is limited to pulmonary involvement in 85% of cases. Although TB can occur in HIV-infected patients irrespective of the degree of immunosuppression, those with more advanced HIV are more likely to have extrapulmonary involvement or other atypical features.[,] Among HIV-infected individuals who develop TB, common systemic symptoms include fever, weight loss, night sweats, anorexia, and malaise. Pulmonary disease usually manifests as cough, which eventually becomes productive of sputum. Hemoptysis, pleuritic pain, and dyspnea may also occur. Extrapulmonary manifestations in HIV-infected persons include lymphadenopathy, splenomegaly, mental status changes, and local pain, depending on the site of involvement.
Immunologic Testing in Suspected Pulmonary TB
The tuberculin skin test can play a role in the evaluation of patients suspected of having active TB. A positive result supports the diagnosis of TB in a patient with a compatible clinical presentation. A negative test, however, does not rule out the possibility of TB, as up to 25% of patients with active TB have a false-negative skin test, even without HIV co-infection. Such false-negative results in HIV-negative individuals are presumably due to poor immune function or compromised nutritional status caused by overwhelming TB disease. Testing with interferon gamma release assay (IGRA) can help confirm that a patient has been infected with M. tuberculosis at some point, but a positive test does not differentiate latent from active TB. Similar to a negative tuberculin skin test, a negative IGRA does not rule out active TB infection. In foreign-born persons from areas where TB is endemic, BCG status should be disregarded, as skin test reactions in adults who received BCG vaccine in childhood are highly variable and usually small (less than 10 mm induration). Prior receipt of BCG does not impact the IGRA test results. In addition, the BCG vaccine is not considered highly efficacious and receipt of the vaccine does not confer reliable protection against TB for adults.
Chest Radiographic Findings of Pulmonary TB
Among patients with a high CD4 cell count, TB typically presents as classic reactivation disease, with infiltrates located in the apical-posterior segment of an upper lobe or superior segment of a lower lobe; these infiltrates may be present with or without cavitation (Figure 1) and (Figure 2). Pleural effusions can also occur (Figure 3). In those with advanced immunosuppression, the chest radiograph may show intrathoracic lymphadenopathy, infiltrates in any lung field, or a combination of lymphadenopathy and infiltrates (Figure 4). In addition, those with advanced immunosuppression often develop diffuse or miliary infiltrates on chest radiograph (Figure 5), seen as innumerable small modules on CT scan (Figure 6). These diffuse infiltrates can resemble the infiltrates seen in patients with Pneumocystis pneumonia. Infrequently, patients will develop pneumothorax caused by pulmonary TB (Figure 7). Some HIV-infected persons can have a normal chest radiograph despite pulmonary involvement.
Diagnostic Tests for Suspected Pulmonary TB
Non-specific laboratory abnormalities can include anemia, leukocytosis, monocytosis, hyponatremia, hypoalbuminemia, elevated transaminases, and increased inflammatory markers. A minimum of three sputum specimens should be collected when pulmonary TB is suspected, with induction by an ultrasonic nebulizer using hypertonic saline if expectorated samples are inadequate.[,] These sputum samples should ideally be collected at least 8 hours apart and sent for AFB smear (Figure 8) and culture (Figure 9). The sensitivity of a sputum smear for detecting acid-fast bacilli (AFB) is approximately 50 to 80%. Nucleic acid amplification (NAA) tests may be useful if the AFB smear is negative and clinical suspicion high, particularly since results from the NAA test often are available within 48 hours. The CDC has published a suggested approach for the use of NAA tests in evaluating patients with suspected TB. For sputum samples, the sensitivity of NAA tests in smear-positive cases is approximately 95%, but it falls to 40 to 80% in AFB smear-negative cases; in both instances, the specificity of NAA is greater than 95%. The combination of a positive smear and NAA confirms the diagnosis of TB and excludes mycobacteria other than M. tuberculosis. The 2013 guidelines for the prevention and treatment of opportunistic infections recommends use of a NAA test on at least one sputum specimen from all HIV-infected patients undergoing evaluation for suspected pulmonary TB.
Regardless of the results of microscopy and any molecular methods used, evaluation of all initial specimens should include culture for M. tuberculosis. Culture is more sensitive than AFB staining techniques and is required for positive species identification and drug susceptibility testing. Other specimens should be obtained if sputum fails to yield a diagnosis. Fiberoptic bronchoscopy can play an important role in obtaining additional lung specimens via bronchoalveolar lavage or transbronchial biopsy. Use of strict infection control precautions can prevent transmission to health-care personnel during these procedures. Patient preparation for bronchoscopy should minimize the use of topical anesthetics, as these may inhibit mycobacterial growth.
Although obtaining pleural fluid by thoracentesis may help diagnose active TB in some cases, the sensitivity of AFB staining and culture of pleural fluid is low. In contrast, microscopic examination and culture of three pleural biopsy specimens can yield the diagnosis in up to 90% of cases. With a normal chest radiograph and negative sputum smears and culture, the evaluation may require biopsy of an extrapulmonary site, such as an enlarged lymph node.
Treatment Considerations for Pulmonary TB in HIV-Infected Persons
The 2013 guidelines for the prevention and treatment of opportunistic infections recommend a 6-month multi-drug regimen for treatment of drug-sensitive pulmonary TB in HIV-infected individuals (Figure 10).[,,] Patients with drug-sensitive pulmonary M. tuberculosis should initially receive 2 months of intensive phase therapy with a 4-drug regimen that consists of isoniazid plus rifampin (or rifabutin) plus pyrazinamide plus ethambutol. The ethambutol can be discontinued sooner if drug susceptibility testing shows the isolate is susceptible to ethambutol. This initial intensive phase of treatment is followed by a continuation phase that consists of 4 months of isoniazid plus rifampin (or rifabutin). If the TB culture is positive at 2 months, the continuation phase should be lengthened from 4 months to 7 months.[,]
Rifampin, and to a lesser extent rifabutin, can cause significant drug-drug interactions in a patient concomitantly taking antiretroviral therapy. Clinicians should carefully monitor patients for medication-related toxicity, including baseline testing of transaminases, bilirubin, alkaline phosphatase, creatinine, and platelet counts, and monthly monitoring during treatment if an abnormality is detected at baseline. In addition, patients should be told to immediately report any symptoms of hepatitis, namely abdominal pain, vomiting, and dark urine. Patients at higher risk for hepatotoxicity can be identified by testing for chronic hepatitis B and C and by screening for alcohol abuse. All HIV-infected patients who receive isoniazid should take pyridoxine (vitamin B6) dosed at 25 to 50 mg daily to prevent peripheral neuropathy. If ethambutol is used, visual acuity and red-green color discrimination should be tested at baseline and monthly thereafter while on ethambutol.
Adherence to therapy is of paramount importance to the success of TB treatment, and a patient-centered approach including directly observed therapy (DOT) is recommended. The total number of doses determines treatment completion, and patients should take all doses of the standard 6-month regimen within 9 months of treatment initiation. Intermittent therapy with the rifamycin drug rifapentine has been associated with the development of drug resistance in HIV-infected patients receiving TB treatment, especially in patients with lower CD4 cell counts. Therefore, rifapentine is not recommended for use in HIV-infected patients. Daily or thrice weekly DOT regimens are recommended for patients with CD4 counts less than 100 cells/mm3. Providers must report active TB to the local health department, which can help manage cases and evaluate contacts.
Monitoring Response to TB Therapy
The clinician should monitor response to treatment with sputum smears, sputum cultures, and chest radiography. The NAA tests should not be used to follow response to therapy, as these tests remain positive in the presence of dead organisms. As noted earlier, if the sputum culture remains positive after the 2-month intensive phase of treatment and the initial chest radiograph showed cavitation, the continuation phase of treatment course should be extend from 18 weeks to 26 weeks.
Susceptibility testing should be performed on all initial isolates and should be repeated on any positive culture after 3 months of treatment or if the patient has a positive culture after a period of negative cultures. If the clinician suspects drug resistance, molecular methods can rapidly detect rifampin resistance. Patients with TB who have HIV infection may have a higher risk for developing drug-resistant M. tuberculosis, especially resistance to rifamycins (rifampin or rifabutin). If laboratory testing detects rifampin resistance, whether alone or in combination with isoniazid resistance, the medical provider should seek expert consultation.
Immune Reconstitution Syndrome
Among patients who start antiretroviral therapy soon after starting TB treatment, 8 to 36% will experience paradoxical worsening of their TB, characterized as new and unexplained fever, radiographic worsening, or an increase in lymphadenopathy or other lesions.[,] This scenario is referred to immune reconstitution inflammatory syndrome (IRIS) and usually occurs within the first 3 months of starting therapy for TB.[,] Factors that predict development of TB-related IRIS include baseline CD4 count less than 50 cells/mm3, high baseline HIV RNA levels, severe TB disease with a high mycobacterial burden, and initiating antiretroviral therapy within 30 days of starting TB therapy.
In the setting of clinical worsening caused by suspected immune reconstitution, it is important to evaluate the patient for TB treatment failure and for other opportunistic infections or malignancies. If causes other than immune reconstitution are deemed unlikely, either prednisone or methylprednisolone can be used to control immune reconstitution symptoms, with the dose depending on whether the patient is receiving rifampin or rifabutin. For patients on rifampin, the recommended prednisone dose is 1.5 mg/kg per day for 2 weeks, then 0.75 mg/kg per day for 2 weeks; for those on rifabutin, the recommended dose is 1.0 mg/kg per day for 2 weeks, then 0.5 mg/kg per day for 2 weeks. With either dose, patients may require a gradual tapering over several months. Patients who have immune reconstitution should not stop TB therapy or antiretroviral therapy. Mild immune reconstitution symptoms can sometimes be controlled with non-steroidal anti-inflammatory drugs.
Timing of Starting Antiretroviral Therapy in Patients with Newly Diagnosed TB
All HIV-infected patients with newly diagnosed TB should have therapy for TB started immediately, regardless of any considerations for antiretroviral therapy. Three major trials—SAPIT, CAMELIA (ANRS 1295), and STRIDE (ACTG 5221)—established the general principle that patients with pulmonary TB and CD4 counts less than 50 cells/mm3 benefit when starting antiretroviral therapy early after starting TB therapy.[,] These findings, as well as expert opinion led to recommendations in the HHS antiretroviral therapy guidelines that addressed the timing of initiating antiretroviral therapy in patients newly diagnosed with TB. These recommendations are based on the patient's CD4 cell count and clinical disease severity: (1) patients with a CD4 count less than 50 cells/mm3 should initiate antiretroviral therapy within 2 weeks of starting TB therapy (2) patients with a CD4 count of at least 50 cells/mm3 and presence of severe clinical disease (low body mass index, low Karnofsky performance score, low hemoglobin or albumin) should initiate antiretroviral therapy within 2 to 4 weeks of starting tuberculosis treatment; and (3) patients with a CD4 count at least 50 cells/mm3 and absence of severe clinical disease can delay starting beyond the 2 to 4 week period, but should start within 8 to 12 weeks of starting TB therapy.
Interactions of Antiretroviral Therapy Regimens and Medications Used to Treat TB
Major interactions can occur with the rifamycins (rifampin or rifabutin) and antiretroviral medications. These drugs do not have significant interactions with nucleoside reverse transcriptase inhibitors. However, rifampin is not recommended for use with protease inhibitors, etravirine (Intelence), rilpivirine (Edurant), or the fixed dose single tablet regimen tenofovir-emtricitabine-cobicistat-elvitegravir (Stribild). Rifampin can be used with efavirenz (Sustiva), raltegravir (Isentress), or dolutegravir (Tivicay), although raltegravir and dolutegravir doses need to be increased. In addition, some experts recommend increasing the efavirenz dose to 800 mg/day when used with rifampin. When rifabutin is used with antiretroviral medications, bidirectional dosing adjustments may be needed. For a more detailed discussion of drug-drug interactions that can occur with TB therapy and antiretroviral medications see the Drug Interactions tables in THe HHS Guidelines for Use of Antiretroviral Therapy.
Of note, the inducing effect of rifampin on cytochrome P450 enzymes persists for at least 2 weeks after its discontinuation. Intolerance, toxicity, or resistance to either HIV or TB medications may necessitate a change in the other regimen. Such decisions should be made in consultation with experts in both TB and HIV care.