Treatment of drug-susceptible pulmonary tuberculosis in nonpregnant adults with HIV infection: Initiation of therapy

INTRODUCTION — Initiation of treatment of drug-susceptible pulmonary tuberculosis (TB) in adults with human immunodeficiency virus (HIV) must address potential drug interactions, adverse reactions including the immune reconstitution inflammatory syndrome, and optimal timing for initiation of antiretroviral therapy (ART) in ART-naïve patients [1-6].

Issues related to initiation of therapy for treatment of patients with HIV and TB infection are reviewed here. Issues related to follow-up after initiation of therapy are discussed separately. (See "Treatment of pulmonary tuberculosis in adults with HIV infection: Follow-up after initiation of therapy".)

General principles related to antituberculous therapy for treatment of drug-susceptible TB in HIV-uninfected adults are discussed separately. (See "Treatment of drug-susceptible pulmonary tuberculosis in nonpregnant adults without HIV infection".)

The clinical features and diagnosis of TB disease in patients with HIV are discussed separately. (See "Diagnosis of pulmonary tuberculosis in adults".)

Beyond issues related to ART discussed here, general principles related to ART are discussed further separately. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach" and "Overview of antiretroviral agents used to treat HIV".)

Issues related to drug-resistant TB are discussed separately. (See "Treatment of drug-resistant pulmonary tuberculosis in adults".)

GENERAL PRINCIPLES — Management of TB in patients with HIV infection should include attention to the following principles:

●Infection control – Issues related to TB transmission and control are discussed separately (see "Tuberculosis transmission and control in health care settings"). The principles are the same for patients with HIV infection as they are for patients without HIV infection.

●Administration of an appropriate regimen for pulmonary TB – The antituberculous regimen must be designed with close attention to potential drug interactions with antiretroviral agents. The initial regimen usually is empiric and must be tailored to drug susceptibility results when available. (See 'Regimen selection' below.)

●Clinical case management with directly observed therapy (DOT) – All patients with TB should be treated with clinical case management and DOT; it is the most effective way to maximize adherence and is recommended by the World Health Organization (WHO) [1,6]. DOT programs are very effective in decreasing rates of primary and acquired drug resistance [7,8]. In countries where TB and HIV are endemic, DOT has been associated with lower TB relapse rates and a higher percentage of patients who complete the full duration of TB therapy [9,10]. (See "Adherence to tuberculosis treatment".)

●Administration of appropriate antiretroviral therapy (ART) for HIV – Initiation or revision of ART must be made with close attention to the timing of initiation as well as potential drug interactions with TB medications.

●Administration of prednisone (for patients with baseline CD4 cell count <100 cells/microL and starting ART within 30 days of starting antituberculous treatment), to reduce the likelihood of immune reconstitution inflammatory syndrome (IRIS) – IRIS can present as a paradoxical worsening of a pre-existing infectious process (eg, TB) following immune recovery associated with administration of ART. The risk of IRIS appears to depend on the baseline CD4 cell count (baseline CD4 cell count <50 cells/microL confers greatest risk) and the timing of ART initiation relative to TB medications. (See 'Timing' below and 'Preventing IRIS' below and "Treatment of pulmonary tuberculosis in adults with HIV infection: Follow-up after initiation of therapy", section on 'Immune reconstitution inflammatory syndrome'.)

●Close monitoring for treatment response – The approach to follow-up after initiation of therapy is discussed separately. (See "Treatment of pulmonary tuberculosis in adults with HIV infection: Follow-up after initiation of therapy".)

ROLE OF EMPIRIC TB THERAPY — Establishing a definitive laboratory diagnosis of TB may not be possible in some circumstances. A presumptive clinical diagnosis may be based on epidemiologic exposure together with physical findings, radiographic findings, positive tuberculin skin test or interferon-gamma release assay, analysis of sputum or bronchoscopy specimens, and/or histopathology. In the setting of high clinical suspicion for TB, initiation of empiric therapy based on these findings is appropriate. (See "Diagnosis of pulmonary tuberculosis in adults", section on 'General diagnostic approach'.)

However, in the absence of specific clinical suspicion for TB, there is no role for routine administration of empiric antituberculous therapy among patients with HIV. Such an approach has been proposed in regions with high burdens of TB and HIV, where many patients with HIV infection are severely immunocompromised at presentation; diagnosing TB is often difficult in such cases, given the often atypical clinical manifestations and limited sensitivity of diagnostic tools [11,12]. However, this approach does not improve survival and confers increased risk of adverse drug reactions [13-15]. In one randomized trial including more than 1000 adults with HIV infection and CD4 cell count <100 cells/microL in four countries with high burden of HIV and TB, administration of empiric antituberculous therapy was not superior to test-guided treatment in reducing mortality at 24 weeks (rate of death from any cause or invasive bacterial disease 19.4 versus 20.3 events per 100 patient-years; adjusted hazard ratio 0.95, 95% CI 0.63-1.44) [13]. Patients who received empiric treatment had a lower rate of TB (3.0 versus 17.9 percent; adjusted hazard ratio 0.15, 95% CI 0.09-0.26), but higher rate of serious adverse events (17.4 versus 7.2 percent; adjusted hazard ratio 2.57, 95% CI 1.75-3.78).

TB DIAGNOSED PRIOR TO HIV — Patients with an established diagnosis of TB disease warrant prompt initiation of antituberculous therapy as well as HIV testing. Patients with TB who are HIV infected warrant initiation of antiretroviral therapy (ART) regardless of CD4 count [1,4,6]; the approach is discussed in the following sections.

Antituberculous therapy

Initial treatment — Patients with pulmonary TB should begin antituberculous therapy promptly. Pending drug susceptibility testing results, a four-drug regimen should be initiated (isoniazid, a rifamycin [eg, rifampin or rifabutin], pyrazinamide, and ethambutol) with pyridoxine and daily therapy using directly observed therapy (DOT) (table 1) [4,6]. (See "Treatment of drug-susceptible pulmonary tuberculosis in nonpregnant adults without HIV infection".)

Regimen selection

●Our approach − For treatment of drug-susceptible pulmonary TB, options include the traditional regimen (≥6 months) or a shortened rifapentine-moxifloxacin (four-month) regimen. The traditional regimen remains standard of care; the shortened regimen may be used in a subset of patients who fulfill specific criteria as described below.

●Rifamycin nitrosamine impurities − In August 2020, the US Food and Drug Administration (FDA) announced detection of nitrosamine impurities in samples of rifampin and rifapentine [16]. Some such compounds have been implicated as possible carcinogens in long-term animal studies, with toxicity largely related to cumulative exposure. To preserve availability of rifampin and rifapentine for TB treatment, the FDA temporarily increased the maximum daily limits of these contaminants.

We favor continued use of rifampin or rifapentine if acceptable to the patient, as the risk of not taking rifampin or rifapentine for TB treatment likely outweighs any potential risk from nitrosamine impurities; this approach is consistent with the United States Centers for Disease Control and Prevention (CDC) guidance issued in September 2020 [17]. Precise levels of contamination for a given lot of drug are not provided to the consumer. However, the nitrosamine exposure is likely to be greater for the four-month rifapentine-moxifloxacin regimen than for the traditional regimen, given daily high dose, administration with a greater allowable limit for rifapentine [18].

Elevated levels of nitrosamine impurities have not been reported for rifabutin.

Traditional regimen (≥6 months) — This regimen including the drugs isoniazid, rifampin, pyrazinamide, and ethambutol (intensive phase of two months and continuation phase of at least four months); it remains standard of care.

●Antimycobacterial agents − The traditional regimen (intensive phase of two months and continuation phase of at least four months) includes the drugs isoniazid, rifampin, pyrazinamide, and ethambutol (sometimes referred to as "RIPE therapy"; outside the United States, this regimen is known as 2HRZE/4HR) (table 2 and table 3) [19]. (See "Treatment of drug-susceptible pulmonary tuberculosis in nonpregnant adults without HIV infection", section on 'Traditional regimen (≥6 months)'.)

●Dosing interval − Patients with TB and HIV infection should receive daily antituberculous therapy (five to seven days per week) during the intensive phase (initial two months of treatment) and during the continuation phase (at least four additional months after the intensive phase) (table 1) [1,6].

This approach is supported by a meta-analysis (largely comprised of cohort studies) in which thrice-weekly therapy in the initial phase was associated with higher rates of treatment failure than daily therapy (adjusted risk ratio 4.0, 95% CI 1.5-10.4) and relapse (adjusted risk ratio 4.8, 95% CI 1.8-12.8) [20,21].

Subsequently, a randomized trial including 331 patients with HIV infection in India treated for TB with a daily, part-daily (daily treatment during the intensive phase and thrice-weekly treatment during the continuation phase), or fully intermittent (thrice weekly throughout treatment) regimen noted a favorable response (all sputum cultures negative during the last two months of treatment) in 91, 80, and 77 percent of cases, respectively [22]. The trial was stopped prematurely based on the inferiority of the fully intermittent regimen and the observation of rifampin resistance in four of the patients in the fully intermittent treatment group (but in none of the patients in the other groups).

Patients with TB and HIV infection (particularly those with <100 CD4 lymphocytes/microL) should not be treated with regimens administered once or twice weekly [6,23,24]. Among such patients, intermittent TB therapy has been identified as a risk factor for relapse and acquired rifamycin resistance [23,25-29].

●Antiretroviral therapy − Depending on the ART regimen chosen, adjustment of the antituberculous regimen may be required (for example, rifabutin may need to be substituted for rifampin in some cases). (See 'Interactions between ART classes and the rifamycins' below and "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach", section on 'Regimen selection'.)

Rifapentine-moxifloxacin-based (four-month) regimen — This regimen may be used in patients >12 years of age with drug-susceptible pulmonary TB and HIV infection with CD4 count ≥100 cell/microL [2,30].

●Antimycobacterial agents − The four-month regimen consists of an intensive phase (eight weeks of high-dose rifapentine, isoniazid, pyrazinamide, and moxifloxacin administered once daily), followed by a continuation phase (nine weeks of rifapentine, isoniazid, and moxifloxacin also administered once daily) (table 4) [2,30,31].

●Antiretroviral therapy − Use of ART for patients on the four-month regimen is restricted to efavirenz-based therapy (with tenofovir disoproxil fumarate plus lamivudine or emtricitabine, as used in the clinical trial). Rifapentine may not be coadministered with any of the integrase inhibitors, protease inhibitors, or with tenofovir alafenamide.

Patient selection — The four-month regimen should not be used for patients in the following categories [2]:

●Patients with suspected or confirmed resistance to the medications in the regimen – Patients who have not received >5 doses of antimycobacterial therapy in the preceding six months may be started on the four-month regimen while drug susceptibility testing is pending. Rapid drug susceptibility should be available; molecular testing for drug resistance (available from local or state public health laboratory) is advised.

●Patients with history of cardiac morbidities (given risk for fluoroquinolone-induced toxicity), including:

•History of arrhythmias (especially bradyarrhythmias), uncorrected hypothyroidism or electrolyte imbalances

•History of prolonged QTc or concurrent use of other QTc-prolonging medications (table 5)

•History of ischemic heart disease

•Family history of sudden cardiac death

●Patients with advanced liver disease, renal insufficiency, and/or laboratory abnormalities including:

•Baseline transaminases >3 times the upper limit of normal

•Baseline total bilirubin >2.5 times the upper limit of normal,

•Baseline creatinine >2 mg/dL

•Baseline serum potassium <3.5 mEq/L

●Pregnant or lactating women

●Patients on drugs with potential for interactions with the medications in the regimen; specific interactions may be determined by using the drug interaction program included within UpToDate.

●Patients with incomplete access to medication or inadequate reimbursement to complete a full course of treatment

Clinical approach — Our approach to laboratory evaluation, electrocardiogram (EKG) monitoring, drug administration, chest radiography, and sputum monitoring is outlined below.  

●Baseline laboratory evaluation − Baseline laboratory evaluation should include:

•Complete blood count (hemoglobin ≥7 g/dL, platelets ≥100,000/microL)

•Transaminases, alkaline phosphatase <3 times the upper limit of normal

•Total bilirubin <2.5 times the upper limit of normal

•Creatinine <2 mg/dL

•Serum potassium >3.5 mEq/L

•Serum calcium and magnesium concentrations within normal limits

•Pregnancy testing if indicated  

●EKG monitoring – Moxifloxacin is a QT-prolonging agent and has been associated with cardiac arrhythmias which may be fatal.

We favor performing baseline EKG prior to starting treatment; we do not use a fluoroquinolone-based regimen for patients with baseline QTc >450 ms (calculator 1). For patients with normal baseline EKG, we perform subsequent EKG monitoring for QTc prolongation one to two weeks later, then at least monthly thereafter. We stop the regimen if QTc increases to ≥500 ms or increases by ≥60 ms over baseline [32,33]. (See "Fluoroquinolones", section on 'QT interval prolongation' and "Antituberculous drugs: An overview", section on 'QT prolongation'.)

Our approach differs from the CDC, which does not recommend EKG monitoring with the four-month regimen [2].

●Drug dosing and administration − The rifapentine-moxifloxacin four-month regimen consists of an intensive phase (eight weeks of rifapentine, isoniazid, pyrazinamide, and moxifloxacin administered once daily), followed by a continuation phase (nine weeks of rifapentine, isoniazid, and moxifloxacin administered once daily) (table 4) [2,30].

At least five of seven weekly doses should be administered under direct observation. (See "Adherence to tuberculosis treatment".)

Therapeutic drug level monitoring may be helpful if there is concern regarding malabsorption.  

●Chest radiography – Following initial evaluation, we obtain repeat chest radiograph at week 8 and at the completion of therapy [2].

●Sputum monitoring − Sputum should be obtained for acid-fast bacilli smear and culture at monthly intervals until two consecutive cultures are negative. A positive sputum culture at two months should prompt drug susceptibility testing of that isolate; patients with drug-resistant isolates should be evaluated for reasons for the emergence of drug resistance (eg, nonadherence, malabsorption).

Completing therapy — − Patients typically demonstrate clinical improvement (with regard to cough, fever, weight loss) within two to three weeks of starting appropriate treatment.

Completion of the regimen consists of 119 doses (56 intensive phase doses and 63 continuation phase doses). The intensive phase doses should be administered within 70 days from treatment initiation, and the continuation phase doses should be administered within 84 days from intensive phase completion.

If these targets are not met, the patients should be considered to have interrupted therapy. Further management should be individualized in consultation with a TB expert, and might require switching to the traditional regimen [1,2].

Regimen efficacy — The efficacy of the four-month regimen is discussed separately. (See "Treatment of drug-susceptible pulmonary tuberculosis in nonpregnant adults without HIV infection", section on 'Regimen efficacy'.)

Antiretroviral therapy

Timing

Clinical approach — We agree with the World Health Organization (WHO), which recommends ART in all patients with HIV infection and TB disease, regardless of CD4 cell count [4]. In addition, we agree with the United States Department of Health and Human Services (DHHS) and the American Thoracic Society (ATS)/CDC/Infectious Diseases Society of America (IDSA), which recommend "integrated ART" (eg, within two to eight weeks after initiating anti-TB therapy) rather than "sequential ART" (eg, after completion of antituberculous therapy) for all patients with HIV infection and TB disease [1,6]. ART initiation should not be delayed until the completion of TB treatment; this approach has been associated with increased morbidity and mortality [1,6]. (See 'Evidence from clinical trials' below.)

TB medications and ART should not be initiated simultaneously. The optimal timing of integrated HIV and TB therapy depends on the patient's immune status [1,3,4,6]:

●For patients with pulmonary TB and CD4 cell count <50 cells/microL, ART should be initiated as soon as possible but within two weeks after initiation of TB treatment. This approach reduces the combined risk of an acquired immunodeficiency syndrome (AIDS)-defining illness and death, despite an increased risk for TB immune reconstitution inflammatory syndrome (IRIS).

●For patients with pulmonary TB and CD4 count ≥50 cells/microL, ART should be initiated within eight weeks after initiation of TB treatment. In the absence of severe disease, early ART is not associated with a decreased risk of AIDS or death and later initiation of ART (eg, eight weeks) is associated with a lower risk of IRIS regardless of baseline CD4 cell count.

●For patients with TB involving the central nervous system (CNS), ART should be delayed for the first eight weeks of antituberculous therapy, regardless of CD4 count. Issues related to CNS TB are discussed separately. (See "Central nervous system tuberculosis: An overview".)

Other clinical considerations (apart from immune status) may influence clinical decisions regarding the timing of ART. Later initiation of ART (eg, eight weeks) may be preferred based on the patient's tolerance of TB medications and ability to swallow multiple pills, while earlier initiation of ART (eg, within two weeks) may be considered in a patient with malnutrition or wasting, regardless of CD4 cell count.

HIV drug-resistance testing should be performed in preparation for ART selection. The choice of ART depends on patient comorbidities, HIV drug-resistance patterns, and cost considerations. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach" and "Overview of HIV drug resistance testing assays".)

In patients with baseline CD4 cell counts <100 cells/microL and starting ART within 30 days of starting antituberculous therapy, we favor prophylactic administration of prednisone during the first four weeks after initiation of ART; this intervention may reduce the likelihood of IRIS. (See 'Preventing IRIS' below.)

Management of IRIS is discussed separately. (See "Immune reconstitution inflammatory syndrome" and "Treatment of pulmonary tuberculosis in adults with HIV infection: Follow-up after initiation of therapy" and 'Preventing IRIS' below.)

Evidence from clinical trials — The recommendation for integrated ART (initiation of ART during TB therapy) over sequential ART (initiating ART after completion of TB therapy) is based on data from a number of randomized trials:

●In a systematic review and meta-analysis within the 2016 DHHS/ATS/IDSA guidelines including eight randomized trials and 4390 patients, initiation of ART during TB treatment was associated with 24 percent overall reduction in mortality (8 studies; risk ratio 0.76, 95% CI 0.57-1.01) [1,34-41]. The overall risk of AIDS-defining illness or death was reduced by 34 percent with early or immediate ART (4 studies; risk ratio 0.66, 95% CI 0.47-0.91). Initiation of ART initiation during antituberculous therapy was associated with increased risk of IRIS (risk ratio 1.88, 95% CI 1.31-2.69). No increase in the risk of other adverse events was identified.

●In the SAPIT trial (Starting Antiretroviral Therapy at Three Points in Tuberculosis), a randomized trial undertaken between 2005 and 2010, the incidence of grade 1 or higher increases in aspartate aminotransferase or alanine aminotransferase among patients who received TB treatment plus efavirenz-based ART in the early-integrated, late-integrated, and sequential ART treatment arms was 34, 30, and 29 percent [42]. However, the rates of grade 3 or 4 liver toxicity were fairly low at 1, 4, and 2 percent, respectively.

The recommendation for early versus delayed ART initiation is based on findings from a number of randomized trials. A systematic review and meta-analysis published in 2015 included eight randomized trials (more than 4500 adults with HIV infection and newly diagnosed pulmonary TB) that evaluated early versus delayed ART initiation (1 to 4 weeks versus 8 to 12 weeks after initiation of TB treatment) [43]. Overall, lower mortality was observed with early ART compared with delayed ART (relative risk [RR] 0.81, 95% CI 0.66-0.99). Among patients with CD4 count <50 cells/microL, lower mortality was observed with early ART compared with delayed ART (RR 0.71, 95% CI 0.54-0.93). Among patients with CD4 count ≥50 cells/microL, no mortality difference was observed between early and delayed ART (RR 1.05, 95% CI 0.68-1.61). Early ART was associated with a higher incidence of TB-IRIS than delayed ART regardless of CD4 count (for CD4 <50 cells/microL: RR 2.50, 95% CI 1.84-3.40; for CD4 >50 cells/microL: RR 2.21, 95% CI 1.50-3.24).

Most studies on the timing of ART have been conducted in patients with smear-positive TB; this subset represents less than half of patients treated for pulmonary TB in resource-limited countries. For patients with smear-negative disease, it is not known if outcomes are comparable.

Preventing IRIS — Immune reconstitution inflammatory syndrome (IRIS) is a paradoxical worsening of a pre-existing infectious process (eg, TB) following immune recovery associated with administration of ART [44-49]. (See "Immune reconstitution inflammatory syndrome", section on 'Tuberculosis'.)

In patients with low baseline CD4 cell counts <100 cells/microL and starting ART within 30 days of starting antituberculous therapy, we favor prophylactic administration of low-dose prednisone (40 mg daily for two weeks, followed by 20 mg daily for two weeks) during the first four weeks after initiation of ART; this intervention may reduce the likelihood of IRIS.

This strategy is supported by a randomized trial including more than 200 adults with HIV infection (CD4 count ≤100 cells/microL) who were initiating ART and had started antituberculous therapy within 30 days before initiating ART; the incidence of TB-associated IRIS was lower among those who received prednisone (40 mg per day for 14 days, then 20 mg per day for 14 days) than those who received placebo (32 versus 47 percent; RR 0.7, 95% CI 0.51-0.96) [50]. There was no increased risk of severe infection or cancer; the trial was underpowered to detect a difference in mortality.

Regimen selection — The selection of the ART regimen depends on multiple factors including virologic potency, convenience, patient comorbidities, drug resistance, drug interactions with antituberculous therapy, and cost. When feasible, it is preferable to select an ART regimen that is compatible with a rifamycin-based antituberculous regimen. (See 'TB regimens based on the rifamycin component' below.)

The selection of ART regimens and the issues related to the various antiretroviral drug classes for the treatment of HIV are discussed in detail separately. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach" and "Selecting antiretroviral regimens for treatment-naive persons with HIV-1: Patients with comorbid conditions".)

HIV DIAGNOSED PRIOR TO TB — Patients with HIV infection who are already on antiretroviral therapy (ART) should continue ART and initiate antituberculous therapy with a rifamycin as soon as possible.

Antiretroviral therapy — If viral suppression has been attained and the current ART regimen is well tolerated, the ART regimen should be continued if possible and the antituberculous regimen adjusted accordingly.

Antituberculous therapy

Regimen selection — The approach to selection of antituberculous therapy in adults with HIV infection is discussed above. (See 'Regimen selection' above.)

A rifamycin is a key component of TB treatment [6]. Rifampin, rifapentine, and rifabutin are considered comparable in terms of activity against Mycobacterium tuberculosis. For patients with HIV infection treated with the traditional regimen, the choice of rifamycin is governed by HIV drug interactions, availability, and cost considerations. For patients treated with the rifapentine-moxifloxacin-based (four-month) regimen, rifabutin may not be substituted for rifapentine. (See 'TB regimens based on the rifamycin component' below.).

There are a number of interactions between rifamycins and antiretroviral agents. The main consideration in choice of antituberculous therapy consists of selecting a rifamycin that will minimize the likelihood of significant drug interactions with ART and interference with HIV viral suppression (see "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach"):

●For patients whose ART regimen contains tenofovir alafenamide (TAF), rifampin may be used with caution, but the combination has not been tested to confirm virologic efficacy [51]. Use of TAF with rifabutin is not recommended; substitution of tenofovir disoproxil fumarate (TDF) should be considered.

●For patients whose ART regimen contains a protease inhibitor, rifabutin is the preferred rifamycin. Rifampin should not be given.

●For patients whose ART regimen contains a non-nucleoside reverse-transcriptase inhibitor (NNRTI), the rifamycin used depends on the NNRTI:

•For patients receiving efavirenz, rifampin is the preferred rifamycin; rifabutin could also be given.

•For patients receiving doravirine, etravirine, or rilpivirine, rifabutin should be administered with dose adjustments, as necessary. Rifampin should not be given.

•For patients receiving nevirapine, rifabutin is the preferred rifamycin; however, interaction between these drugs may increase rifabutin exposure and, with this, risk for rifabutin toxicity. We use rifampin with nevirapine in select situations for patients who are virologically suppressed [3].  

In general, rifabutin has fewer drug interactions than rifampin; however, rifampin is more cost effective than rifabutin in resource-limited settings. (See 'TB regimens based on the rifamycin component' below.)

●For patients whose ART regimen contains an integrase strand transfer inhibitor (INSTI): the dose of some INSTIs (dolutegravir, raltegravir) should be doubled if rifampin is given.

•Dolutegravir and raltegravir may be administered with either rifampin or rifabutin; the dose of these drugs should be doubled (ie, given twice daily) if given with rifampin (for example, dolutegravir dosing would be 50 mg orally twice daily rather than 50 mg orally once daily).

•Cabotegravir may be administered with rifabutin but not with rifampin or rifapentine.

•Other INSTIs (bictegravir, elvitegravir-cobicistat) should not be coadministered with rifampin or rifabutin.

DRUG INTERACTIONS

General principles — Careful review for drug interactions is essential when considering antituberculous therapy in patients taking antiretroviral therapy (ART). Such interactions can interfere with efficacy of therapy for either HIV or TB infection. A useful tool for HIV drug interactions may be found online.

TB regimens based on the rifamycin component — Rifamycins induce hepatic CYP3A4 enzymes that can accelerate metabolism of protease inhibitors (PIs) and some non-nucleoside reverse transcriptase inhibitors (NNRTIs) [6,52]. Rifampin is a much more potent inducer than rifabutin. Conversely, serum concentrations of rifabutin may be decreased by the concomitant administration of certain antiretroviral drugs (eg, efavirenz).

Despite the complexity of managing drug interactions, rifamycins are integral to the success of TB therapy and should not be substituted with other antituberculous medications based on concerns regarding drug interactions alone [6,52]. The following sections address selection of antiretroviral medications with either a rifampin- or a rifabutin-based TB regimen.

●Rifapentine-based regimen − Use of ART for patients on the rifapentine-moxifloxacin-based (four-month) regimen is restricted to efavirenz-based therapy (with tenofovir disoproxil fumarate [TDF] plus lamivudine or emtricitabine, as used in the clinical trial) [30]. Rifapentine may not be coadministered with any of the integrase inhibitors, PIs, or with tenofovir alafenamide (TAF). (See 'Regimen selection' above.)

●Rifampin-based regimens − A limited number of antiretroviral medications within a limited number of drug classes can be coadministered with rifampin-based TB treatment, due to drug interactions that adversely affect HIV treatment efficacy.

The ART regimen most commonly administered concomitantly with a rifampin-based TB regimen consists of the NNRTI efavirenz, together with two nucleoside analogs [6,53]. Regimens that include the integrase strand transfer inhibitors (INSTIs) dolutegravir or raltegravir (but not bictegravir, cabotegravir, or elvitegravir-cobicistat), may also be administered with rifampin at double the dose [6].

Drug interaction information related to specific ART drug classes and rifampin-based TB therapy is discussed below. (See 'Interactions between ART classes and the rifamycins' below.)

●Rifabutin-based regimens − Rifabutin is associated with fewer drug-drug interactions with antiretroviral agents than rifampin. Rifabutin is as effective as rifampin for treatment of TB when used in combination therapy. However, rifabutin is more expensive than rifampin and distribution is limited in resource-limited settings. The adverse effects of rifampin and rifabutin are comparable.

Rifabutin may be coadministered with nucleoside analogs (except TAF), PIs, NNRTIs, and INSTIs (except elvitegravir-cobicistat and bictegravir). The standard dose of rifabutin is 300 mg daily; however, when dosed with ART, rifabutin often requires dose adjustments depending on the coadministered drug. A useful tool for HIV drug interactions may be found online.

Rifabutin is a substrate for CYP3A4. Thus, its serum concentration is affected by the degree to which CYP3A4 is inhibited or induced by the coadministered drug (table 6).

Drug interaction information related to specific ART drug classes and rifabutin-based TB therapy is discussed below. (See 'Interactions between ART classes and the rifamycins' below.)

Interactions between ART classes and the rifamycins

●Rifapentine

•Nucleoside analogs – Rifapentine may be coadministered with TDF. Rifapentine should not be coadministered with TAF given possibility of decreased TAF concentration.

•Non-nucleoside reverse transcriptase inhibitors – Rifapentine may be coadministered with efavirenz. Rifapentine should not be coadministered with doravirine, etravirine, nevirapine, or rilpivirine due to significantly decreased concentrations.

•Protease inhibitors – Rifapentine should not be administered with any protease inhibitor due to significantly decreased concentrations. Rifapentine also should not be given with cobicistat, a pharmacokinetic booster/CYP3A inhibitor often used with protease inhibitors.

•Integrase inhibitors – Rifapentine given daily should not be administered with bictegravir, cabotegravir, dolutegravir, elvitegravir, or raltegravir.

●Rifampin

•Nucleoside analogs − Nucleoside analogs may be coadministered with rifampin; a notable exception is TAF. Rifampin decreases drug levels of TAF, so TAF should generally not be given with rifamycins [6]. Rifamycins are potent inducers of P-glycoprotein and TAF is a P-glycoprotein substrate.

The nucleotide analog TDF does not have substantial interactions with rifampin, so TDF and rifampin may be coadministered.

•Non-nucleoside reverse transcriptase inhibitors − For patients receiving NNRTI-based ART with a rifampin-based TB regimen, the preferred NNRTI is efavirenz. Viral suppression rates are more favorable with efavirenz-based than nevirapine-based ART; this is because use of rifampin reduces nevirapine concentrations to a greater degree than efavirenz concentrations [6,53-57]. In addition, several trials of ART in treatment-naïve patients have shown that nevirapine is associated with higher rates of hepatotoxicity than efavirenz [58]; this is important since antituberculous agents can also cause drug-induced liver injury. (See "Treatment of pulmonary tuberculosis in adults with HIV infection: Follow-up after initiation of therapy".)

The appropriate dose of efavirenz with rifampin is somewhat controversial [6,53]. We are in agreement with the United States Department of Health and Human Services and United States Centers for Disease Control and Prevention guidelines, which favor the standard efavirenz dose (600 mg daily) among individuals treated with rifampin [6]. Efavirenz levels are generally higher in patients from Southern Africa, West Africa, and India due to varying pharmacogenetics of efavirenz among different ethnic and racial groups [59-61].

Initiation of nevirapine should be avoided in patients who are already on antituberculous therapy, due to risk of virologic failure. However, patients on nevirapine-containing ART who have achieved virologic suppression and develop TB may continue the same ART regimen [53]. This was illustrated in a cohort study including more than 2000 patients in South Africa; virologic outcomes were comparable among patients starting efavirenz-based ART with and without concurrent TB and in patients in whom TB developed while taking established nevirapine- or efavirenz-based therapies [53].

If nevirapine is administered to patients on rifampin, nevirapine dosing consists of 200 mg twice daily (with no lead-in phase) [6]. Increasing the dose of nevirapine further is not beneficial. In one trial comparing standard dosing of nevirapine (200 mg twice daily) with a higher dose (300 mg twice daily), increased rates of hypersensitivity were observed among those randomized to the higher dose [62].

The NNRTIs doravirine, etravirine, and rilpivirine should not be used with rifampin [6]. Rifampin substantially decreases concentrations of all three drugs.

•Protease inhibitors − PI-based ART should not be administered with rifampin [6]. Rifabutin, rather than rifampin, should be used when PIs are used for HIV therapy.

In patients with HIV infection initiating ART, PIs are usually dosed with the pharmacokinetic enhancer ritonavir, which boosts serum drug levels. Administration of rifampin with boosted PIs is associated with reduction of PI levels by as much as 95 percent [63,64].

Some studies have evaluated dose increments of a particular PI (eg, lopinavir or atazanavir) to determine if serum concentrations can be maintained with rifampin. However, this approach has been associated with significant adverse events including gastrointestinal intolerance and hepatotoxicity and is not recommended [63-66].

•Integrase inhibitors − Rifampin and raltegravir may be coadministered with caution; viral load monitoring is required. We favor doubling the dose of raltegravir when coadministered with rifampin [67]. Rifampin decreases serum levels of raltegravir by 40 to 61 percent [68]. Standard raltegravir dosing (400 mg twice daily) may be sufficient [69,70]; doubling the dose of raltegravir (to 800 mg twice daily) improves drug levels [67].

Rifampin and dolutegravir may be coadministered; in such circumstances, we favor doubling the dose of dolutegravir (from 50 mg once daily to 50 mg twice daily) [6,71,72], given a significant drug-drug interaction between dolutegravir and rifampin resulting in a 56 percent reduction in the area under the curve (AUC) of dolutegravir [73]. This approach is supported by a randomized trial including more than 100 patients with TB (on rifampin-based treatment for ≤8 weeks) and HIV (ART naïve) randomly assigned to initiation of ART with two nucleoside reverse transcriptase inhibitors and either dolutegravir (twice-daily dosing) or efavirenz [71]. Double-dose dolutegravir was well tolerated; the trial was not powered to compare virologic efficacy.

Rifampin should not be administered with the fixed-dose combination of elvitegravir, cobicistat, tenofovir, and emtricitabine, as rifampin is expected to decrease concentrations of both elvitegravir and cobicistat [6].

Bictegravir should not be administered with rifampin because bictegravir levels are significantly decreased by rifampin [74]. Doubling the dose of bictegravir does not adequately overcome the inducing effect of rifampin; in one study, plasma bictegravir AUC was 60 percent lower when coadministered with rifampin, even when double the recommended dose was given [74].  

Cabotegravir should not be administered with rifampin because rifampin is expected to decrease cabotegravir concentrations significantly.

•Other ART classes − Maraviroc (a CCR5 receptor antagonist) should not be coadministered with rifampin [6]. Rifampin decreases serum levels of maraviroc by 78 percent; there is no published experience regarding efficacy or toxicity of this combination.

Fostemsavir (CD4 attachment inhibitor) should not be coadministered with rifampin, as rifampin significantly decreases fostemsavir concentrations [75,76]; however, rifabutin has only a weak interaction and can be coadministered [77].

Enfuvirtide (a fusion inhibitor) or ibalizumab (CD4 postattachment inhibitor) may be coadministered with rifampin; there are no interactions between these drugs.

●Rifabutin

•Nucleoside analogs − Nucleoside analogs may be coadministered with rifabutin; a notable exception is TAF. Rifabutin decreases drug levels of TAF, so TAF should not be given [6]. Rifamycins are potent inducers of P-glycoprotein and TAF is a P-glycoprotein substrate.

The nucleotide analog TDF does not have substantial interactions with rifabutin, so TDF and rifabutin may be coadministered.

•Non-nucleoside reverse transcriptase inhibitors − Efavirenz can reduce serum concentrations of rifabutin. Therefore, if efavirenz is administered, the dose of rifabutin should be increased from the standard dose of 300 mg to 450 to 600 mg/day [6,78]. A useful tool for HIV drug interactions may be found online.

There are no significant drug interactions between rifabutin and nevirapine.

Rifabutin can be given with doravirine, etravirine, or rilpivirine; dose adjustments may be necessary [6].

•Protease inhibitors − PIs can be coadministered with rifabutin without significantly compromising the serum drug concentrations, with or without the coadministration of ritonavir or cobicistat for pharmacologic boosting [79]. The dose of rifabutin should be decreased from 300 mg daily to 150 mg daily (or 300 mg three times per week). A useful tool for HIV drug interactions may be found online.

The rifabutin dose previously recommended for coadministration with ritonavir-boosted PIs (150 mg three times weekly) has been associated with insufficient rifabutin levels [80]. In addition, there have been reports of acquired rifamycin resistance among individuals receiving this dose [24].

There are no published data on coadministration of darunavir and rifabutin [81].

•Integrase inhibitors − Rifabutin does not appear to substantially affect raltegravir, dolutegravir, or cabotegravir concentrations (based on limited in vitro data) [72,82]. If coadministered with rifabutin, the doses of raltegravir, dolutegravir, or cabotegravir do not need to be increased [6].

Rifabutin should not be used with the fixed-dose combination of elvitegravir, cobicistat, tenofovir, and emtricitabine, given decreased elvitegravir levels. Likewise, rifabutin should not be used with bictegravir; in one study, rifabutin coadministration resulted in a 38 percent reduction in AUC [83].

•Other ART classes − There are no published data on the combined use of rifabutin and maraviroc. Dosing recommendations vary according to the concomitant medications [6]. A useful tool for HIV drug interactions may be found online.

There are no drug interactions between rifabutin and either the fusion inhibitor enfuvirtide, the CD4 attachment inhibitor fostemsavir, or the CD4 postattachment inhibitor ibalizumab.

TREATMENT FAILURE OR RELAPSE

●Definitions

•Treatment failure

-For the traditional regimen, treatment failure refers to positive sputum cultures after four months of antituberculous therapy [1].

-For the rifapentine-moxifloxacin four-month regimen, lack of clinical, radiographic, or microbiologic improvement at eight weeks of treatment should prompt complete re-evaluation of the patient and treatment regimen.

•Relapse − Relapse refers to recurrent TB at any time after completion of treatment with apparent cure.

Most relapses occur within the first 6 to 12 months following completion of therapy. Among patients with drug-susceptible TB, relapse occurs in approximately 5 percent of cases [84].

Relapse may occur as a result of relapsed infection due to the same M. tuberculosis strain (more common in low-incidence settings) or due to exogenous reinfection with a new strain (more common in high-incidence settings) [85-89].

Among patients treated with rifamycin-containing regimens using directly observed therapy (DOT), relapse generally occurs with susceptible organisms. If initial drug susceptibility testing was not performed and the patient fails or relapses with a rifamycin-containing regimen given by DOT, there is high likelihood that the organism was resistant from the outset.

●Risk factors − Risk factors for treatment failure and relapse include [90-93]:

•High burden of disease (presence of cavitary disease, bilateral disease, and/or extrapulmonary disease)

•Drug resistance

•Inadequate treatment adherence

•Malabsorption (if malabsorption is suspected, drug level testing may be useful)

•Malnourishment

•Alternative diagnosis

●Drug susceptibility testing − If treatment failure or relapse is confirmed or suspected, the M. tuberculosis isolate should be sent for drug susceptibility testing to first- and second-line agents. In clinical and public health laboratories, drug resistance is evaluated by assessing growth of in the presence of a "critical concentration" of drug (defined as the lowest concentration of drug that inhibit 95 percent of "wild-type strains") [94]. In the United States, specimens may be forwarded to the Centers for Disease Control and Prevention for molecular testing with relatively rapid turnaround time [95,96]. (See "Diagnosis of pulmonary tuberculosis in adults", section on 'Microbiologic testing'.)

●Management − The approach to management of treatment failure and relapse is discussed separately. (See "Treatment of drug-resistant pulmonary tuberculosis in adults", section on 'Empiric treatment'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diagnosis and treatment of tuberculosis" and "Society guideline links: Opportunistic infections in adults with HIV".)

SUMMARY AND RECOMMENDATIONS

●The clinical approach to treatment of tuberculosis (TB) in patients with HIV infection depends upon which condition is diagnosed first. (See 'Introduction' above.)

●Patients with an established diagnosis of TB disease require prompt initiation of antituberculous therapy. Empiric treatment of TB infection may be appropriate in patients for whom there is high clinical suspicion for TB (based on epidemiologic exposure, physical findings, radiographic findings, and supportive laboratory data) but no definitive laboratory diagnosis. In contrast, we suggest against routine administration of empiric antituberculous therapy in patients with HIV infection (Grade 2B). In one trial, empiric treatment in these patients was associated with similar mortality rate but increased adverse events compared to test-guided treatment. (See 'Role of empiric TB therapy' above.)

●In patients with HIV and TB, an empiric four-drug regimen should be initiated (isoniazid, rifampin, pyrazinamide, and ethambutol) with pyridoxine and daily therapy using clinical case management and directly observed therapy (DOT) until drug susceptibility results are available (table 1). Depending on the antiretroviral therapy (ART) regimen chosen, adjustment of the antituberculous regimen may be required. (See 'TB diagnosed prior to HIV' above.)  

●For treatment of drug-susceptible TB in patients with HIV infection, we suggest treatment with the traditional (≥6 month) regimen (Grade 2C). The shortened rifapentine-moxifloxacin-based (four-month) regimen is a potential alternative regimen for patients >12 years and CD4 >100 cells/microL. Use of the shortened regimen requires attention to ART drug interactions (rifapentine may not be coadministered with integrase inhibitors or tenofovir alafenamide [TAF]) as well as more detailed programmatic considerations. (See 'Regimen selection' above.)

●For patients with HIV infection and pulmonary TB who are not already on ART, we recommend initiation of ART during TB treatment (eg, within two to eight weeks after initiating antituberculous therapy) rather than waiting until after completion of TB treatment (Grade 1B) (see 'Timing' above):

•For patients with CD4 cell count <50 cells/microL, we recommend starting ART within two weeks after starting antituberculous therapy (rather than at later time points) (Grade 1B). Studies suggest that this approach is associated with a lower risk of mortality, although higher rates of immune reconstitution inflammatory syndrome (IRIS).

•For patients with CD4 count ≥50, we suggest deferring initiation of ART to eight weeks after starting antituberculous therapy (Grade 2C). Studies suggest that in this group, delayed versus early treatment has an uncertain impact on mortality and a lower rate of IRIS.

●For patients with HIV infection and TB involving the central nervous system (CNS), ART should be delayed for the first eight weeks of antituberculous therapy, regardless of CD4 count. Issues related to CNS TB are discussed separately. (See "Central nervous system tuberculosis: An overview".)

●For patients with HIV infection and baseline CD4 cell count <100 cells/microL on antituberculous therapy and initiating ART within 30 days of starting antituberculous therapy, we suggest prophylactic administration of prednisone during the first four weeks following initiation of ART (Grade 2B); this intervention may reduce the likelihood of IRIS. (See 'Preventing IRIS' above.)

●Patients with HIV infection who are already on ART should continue the ART regimen and initiate antituberculous therapy with a rifamycin as soon as possible. If viral suppression has been attained and the current ART regimen is well tolerated, the ART regimen should be continued and the antituberculous or antiretroviral regimen adjusted accordingly with consideration for drug-drug interactions. (See 'HIV diagnosed prior to TB' above.)

●Pending drug susceptibility results, an empiric four-drug regimen should be initiated (isoniazid, a rifamycin [eg, rifampin or rifabutin], pyrazinamide, and ethambutol) with pyridoxine and daily therapy using DOT (table 1). The main consideration consists of selecting a rifamycin that will minimize the likelihood of drug interactions with ART and interference with HIV viral suppression. A useful tool for HIV drug interactions may be found online. (See 'Antituberculous therapy' above.)

●When feasible, it is preferable to select an ART regimen that is compatible with a rifamycin-based antituberculous regimen. However, there are a number of drug interactions between rifampin and antiretroviral agents. Rifabutin is as effective as rifampin for treatment of TB when used in combination therapy and is associated with fewer drug interactions with antiretroviral agents than rifampin. However, rifabutin is more expensive than rifampin, and distribution is limited in resource-limited settings. The adverse effects of rifampin and rifabutin are comparable. (See 'Regimen selection' above and 'TB regimens based on the rifamycin component' above.)

●For patients with TB and HIV who are receiving a rifampin-based antituberculous regimen, we suggest an ART regimen consisting of the integrase inhibitors raltegravir or dolutegravir (at double dose), or the non-nucleoside reverse transcriptase inhibitor (NNRTI) efavirenz, together with two nucleoside analogs (except TAF) (Grade 2B). (See 'TB regimens based on the rifamycin component' above.)

●For patients with TB and HIV who are receiving a rifabutin-based antituberculous regimen, we suggest either a protease inhibitor, NNRTI, or compatible integrase inhibitor, together with two nucleoside analogs (except TAF) (Grade 2C). (See 'TB regimens based on the rifamycin component' above.)

●Issues related to follow-up and duration of therapy are discussed separately. (See "Treatment of pulmonary tuberculosis in adults with HIV infection: Follow-up after initiation of therapy".)