D23 Tuberculosis in HIV/AIDS


Worldwide, tuberculosis (TB) incidence has been declining at about 2% per year, while a quarter of the world’s population is infected with Mycobacterium tuberculosis (MTB). In 2017, there were 10.0 million new TB cases, of which 0.83-1.01 million (9%) were people living with HIV/AIDS (PLWHA). Out of 1.6 million TB deaths, 0.30 (0.27-0.33) million were HIV-associated. (Global tuberculosis control report 2017, World Health Organization www.who.int/tb) As regards bacillary drug resistance, 0.6 million cases of rifampicin-resistant (RR) and multidrug-resistant TB (MDR-TB) were estimated to emerge in 2017. The proportion of MDR-TB cases with extensively drug-resistant TB (XDR-TB), defined as MDR-TB with additional resistance to fluoroquinolones and one or more of the three injectable drugs (kanamycin, amikacin and capreomycin), was 8.5% (95% CI: 6.2-11%), an increase from the 6.2% in 2016. Among countries with a high TB or MDR-TB burden, the proportion of MDR/RR-TB cases with resistance to any fluoroquinolone was about 22%. Drug-resistant TB and HIV are causing a deadly syndemic in a number of parts of the world including some countries in Africa.

Hong Kong is classified as a place of ‘intermediate’ TB burden with good health infrastructure in the Western Pacific Region. The local TB notification rate decreased from a peak of 697.2 per 100,000 in 1952 to 58.1 per 100,000 in 2018. The death rate was 2.4 per 100,000. Voluntary HIV testing is routinely offered to all patients diagnosed of TB in the public service in Hong Kong. The proportion with a positive HIV antibody result among those tested ranged from 0.6 to 0.8% in the past five years. Overall, HIV-associated TB cases constitute around 1% of TB notifications. In contrast to areas with ‘low’ TB burden, only the extrapulmonary form of TB (at any CD4 count; except cervical lymphadenopathy) and pulmonary form of TB that occurs at a CD4 count below 200/μL are considered AIDS-defining conditions for surveillance purpose.

Manifestations of TB in HIV-infected patients

Compared to HIV-negative persons, the risk of progression to active TB in HIV/MTB co-infected persons is increased about 20-100 fold, at a rate of 5-10% per year.[1] Although active TB can occur at all ranges of CD4 cell counts, its clinical features vary according to the degree of immunosuppression. With higher CD4 counts (e.g. >350/μL), the manifestations generally appear more ‘typical’, such as apical or upper lobe pulmonary infiltrations and cavity formation. With lower CD4 counts (e.g. <50/μL), extrapulmonary TB, with or without concurrent pulmonary involvement, becomes more common (in 40-80% of cases compared with 10%-20% of HIV-uninfected persons). Multiple lymphadenitis, pleuritis, pericarditis, spondylitis, meningitis, cerebral tuberculomas, miliary or disseminated infections may be the presenting features. For pulmonary diseases, ‘atypical’ features like lower or middle lobe infiltration without cavitation are common and can confuse the diagnosis. Moreover, a decrease in CD4 cell count and a 5 to 160-fold rise in viral load have been demonstrated in TB, suggesting a possible effect of TB in accelerating HIV progression.

Diagnosis of active TB

Diagnosis of active TB disease in HIV-infected patients requires a high index of suspicion. World Health Organization has recommended a TB symptom screening algorithm (WHO.2016. http://www.who.int/hiv/pub/arv/annexes-5Sep2016.pdf). The screening tool is based on the self-reported presence of any of 4 symptoms (weight loss, fever, night sweats, and cough of any duration) associated with active TB. Having a high negative predictive value, the tool is mainly used as a “rule-out” test. Further clinical scoring that classifies subjects with a positive WHO-TB screening result for the likelihood of TB have been developed, which may help to reduce the number of patients in need of further TB investigations before starting antiretroviral therapy (ART).[2] While clinical vigilance is important for the detection of TB, the absence of symptoms may NOT reliably exclude disease, especially among those with compromised immunity.

For most patients, the initial investigations should include a chest radiograph and sputum examination (see below). Notably, up to one third of TB patients with HIV coinfection may have a “normal” chest radiograph.[3] Lowering the diagnostic threshold to any radiographic abnormality may increase the sensitivity; and sputum examination for acid-fast bacilli (AFB) should still be considered if the clinical presentations are suggestive of TB, regardless of radiological appearances. Three separate sputum specimens should be collected. Sputum induction by inhalation of hypertonic saline (under suitable safety precaution) may be considered if sputum cannot be produced spontaneously. When clinically indicated, bronchoalveolar lavage, transbronchial biopsy, pleural fluid or lymph node aspiration/biopsy, and early morning urine can be obtained for AFB examination. Imaging studies like ultrasound or CT scans may be helpful in localising the site and extent of infection, and guide biopsies.

Smear and culture

Although sputum microscopy is simple, rapid, readily available and inexpensive, it has a low and variable sensitivity (20-60%). Its sensitivity is usually higher in cavitary disease. Fluorescent microscopy using conventional florescent light sources or fluorescent light-emitting diode (LED) is more sensitive but more expensive. Specificity of AFB smear is affected by the prevalence of non-tuberculous mycobacterial infections (NTM, particularly Mycobacterium avium complex, or MAC) because of their indistinguishable morphologies. Further laboratory testing is required for differentiation.

The gold standard of diagnosis is the isolation of MTB by culture. Culture requires only 10-100 viable organisms per ml to become positive. However, culture on solid media takes up to 8 weeks to yield positive mycobacterial growth. Automated rapid liquid culture reduces the turn-around time of mycobacterial culture to an average of 3-4 weeks, and may be more sensitive than culture on solid media. Adding two liquid cultures to direct smears on the same specimens could increase sensitivity to 76% among HIV-infected patients.[3] The algorithm with liquid culture is more complicated than using solid medium, and mixed growths may be more difficult to detect in liquid culture as colony morphology cannot be observed, thus limiting its overall utility for drug susceptibility testing (DST).

Newer rapid culture and DST systems, e.g. microscopic-observation drug-susceptibility (MODS), have been evaluated and reported to show higher MTB detection yield especially in smear negative HIV patients. For miliary TB, positive rate of conventional sputum culture is only about 25%; collection of samples from multiple sites (e.g. tissues, body fluids, bone marrow) may be required to increase the diagnostic yield (to 50-60%). In Hong Kong, the government TB laboratory has switched from solid-based (Lowenstein-Jenson) to liquid-based medium, i.e. the MGIT culture system. Lowenstein-Jensen medium is used in addition to MGIT culture only for non-respiratory specimens. All MTB isolates are subjected to DST, which is a key component of the “DOTS-Plus” treatment strategy (more recently called Drug-resistance Programme) implemented in Hong Kong.

Molecular assays

Nucleic acid amplification tests (NAAT) are increasingly utilised to facilitate rapid diagnosis of TB and for the detection of drug resistance. Biohazards of these molecular assays are generally lower than live MTB culture. Besides respiratory samples, such tests can be applied to non-respiratory samples, such as lymph node aspirate, cerebrospinal fluid, urine, gastric aspirate, liver biopsy and bone marrow. It should be noted that the sensitivity of NAAT in smear-negative (culture positive) samples is only moderate, and the presence of polymerase chain reaction inhibitors, particularly in the extrapulmonary specimens, may lead to false-negative results. Thus a single negative result cannot be used to rule out the diagnosis of TB. NAATs cannot differentiate between dead or live tubercle bacilli, which limit their roles in assessing infectivity and monitoring treatment response.

The Xpert MTB/RIF, a newer generation molecular assay, was endorsed by WHO in 2010 for use in HIV/TB co-infected patients, and later in 2013 as the initial diagnostic test for patients with signs and symptoms of pulmonary TB.. For culture-positive TB, a single test shows a sensitivity of 72.5% and 98.2% in cases with negative and positive sputum smears respectively. Applying three tests may increase the sensitivity to 90.2% in smear-negative cases. Unlike microscopy, its sensitivity is not significantly reduced in patients with HIV co-infection. Another advantage is its short turnover time (can be less than 2 hours). For detection of rifampicin (RIF) resistance, the assay has pooled sensitivity and specificity of 95% and 98% respectively. [WHO policy statement 2011 http://whqlibdoc.who.int/publications/2011/9789241501545_eng.pdf]

For all suspected new tuberculosis cases and for patients planning to restart treatment with previous history of tuberculosis, the WHO-recommended Xpert MTB/RIF assay is performed if the sputum specimen is positive for AFB smear in Hong Kong. While further molecular assays (such as line probe assays) can be extended to detect bacillary resistance to isoniazid and fluoroquinolone, and perhaps second-line injectable drugs and pyrazinamide, conventional DST for phenotypic resistance is still necessary to confirm the results and to guide the use of second-line drugs when treating MDR-TB. Recently, whole-genome sequencing has been shown to be able to accurately characterise susceptibility to first-line drugs, and guide choice of therapy. However, robust software and database tools need to be developed for the full potential of whole genome sequencing in this area to be realised.

Management of TB in HIV-infected patients

Early diagnosis and prompt initiation of anti-tuberculous treatment is important to reduce mortality of HIV-associated TB. The updated WHO guidelines recommend universal combination antiretroviral therapy (ART) in all patients with active TB (regardless of CD4 count), and to initiate such therapy as soon as possible based on the CAMELIA, SAPIT and STRIDE trials, e.g. when patient tolerating anti-tuberculous drugs for 2 weeks, especially in patients with a CD4 count of less than 50/μL. Delaying ART to beyond 8 weeks is generally not recommended as it can be associated with increased deaths.[4] Voluntary HIV testing is therefore recommended for all new TB patients. Early referral of newly diagnosed HIV cases to an appropriate HIV service will facilitate timely initiation of highly active antiretroviral therapy (HAART). Recent evidences also indicate that starting HAART 1-4 weeks after initiation of anti-tuberculous treatment is associated with better overall survival in patients with advanced immunosuppression (e.g. CD4 count <50/μL), although there may be a higher incidence of immune reconstitution inflammatory syndrome (IRIS)[Chapter C17] and other ART-related adverse events (e.g. hepatotoxicities) and drug-drug interactions (DDI).[5][6] The only exception is probably in the case of TB meningitis when starting HAART too soon before adequate control of mycobacterial infection may sometimes precipitate intracranial IRIS, which might prove fatal.[7] In a local study of 260 co-infected patients, HAART within two months of TB treatment was associated with a favourable outcome (91% vs 67%). Algorithm 23 on the general approach to TB in HIV/AIDS is given at the end of the chapter.

Anti-tuberculous treatment and directly observed treatment (DOT)

The general principles of anti-tuberculosis therapy similarly apply in HIV-associated TB.[Further reading A] For drug-sensitive pulmonary TB, every effort should be made to use a standard regimen containing a rifamycin (RIF or rifabutin) throughout the whole course of therapy, despite the complexity of drug interactions between rifamycins and antiretrovirals. Directly observed therapy (DOT) is deemed the standard of care. Recent systematic reviews have indicated higher risks of acquired drug resistance, especially rifamycin monoresistance, and treatment failure in patients receiving intermittent dosing regimens compared with those given daily dosing regimens, particularly among the HIV-infected patients.[8] As such, WHO recommended AGAINST the use of thrice-weekly dosing in both the intensive and continuation phases of therapy in all patients with drug-susceptible pulmonary TB, and that daily dosing remains the recommended dosing frequency in its latest treatment guideline. As regards TB treatment duration, WHO recommended a 6-months standard treatment regimen over an extended treatment for 8 months or longer in patients with drug-susceptible pulmonary TB who are living with HIV and receiving ART during TB treatment in its latest guideline, on the rationale that PLWHA who are responding to HAART need not expect a more unfavourable outcome to a treatment episode than an uninfected person.

In Hong Kong, the standard regimen for drug-sensitive TB is one containing isoniazid (INH) and a rifamycin, given for a total duration of 9 months, or >4 months after culture conversion in HIV-infected patients. Rifabutin (RFB) should replace RIF if drug interaction with an antiretroviral is a concern (see below). Pyrazinamide (PZA) and ethambutal (EMB) or streptomycin (SM) are also used in the initial phase (the first 8 weeks). The commonly used dosages of the first line anti-tuberculous drugs and their side effect profiles are listed in Box 23.1 and Box 23.2 respectively. The local recommendation is in line with the findings from a meta-analysis of TB outcomes among HIV-infected patients, that treatment with a rifamycin-containing regimen for ≥9 months was associated with a decreased risk of TB relapse, compared to a 6-month regimen.[8] When treating CNS disease, and possibly for bone and joint diseases, treatment duration should be extended to 12 months. Modification of drug regimen and prolongation of treatment course may be necessary due to drug resistance, DDI, drug intolerance, or unsatisfactory response.

Box 23.1. Usual dosage of the first line anti-tuberculous drugs

Drug daily dose* tiw dose*
Isoniazid (INH)# 5 (300 mg) 10-15 (1100 mg)
Rifampicin (RIF)# 10 (600 mg) 10 (600 mg)
Rifabutin (RFB) 5 (300 mg) ——-
Pyrazinamide (PZA) 20-30 (2.0 g) 30-40 (2.5 g)
Ethambutol (EMB) 15-20 (1600 mg) 25-30 (2200 mg)
Streptomycin (SM)# 12-15 (1.0 g) 15 (1.0 g)
*Dosage of common anti-tuberculosis drugs in mg/kg, maximum dose in ( ) tiw, three times per week dosing regimen.
#Some authorities recommend higher dosages (per kg body weight) of isoniazid, rifampicin and streptomycin for young children.
See Box 23.5 for dose adjustment required when used with some anti-retroviral agents.

Box 23.2. Adverse reactions to the first line anti-tuberculosis drugs

Isoniazid (INH) Hepatitis, cutaneous hypersensitivity, peripheral neuropathy
Rifampicin (RIF) Hepatitis, cutaneous reactions, gastrointestinal reactions, thrombocytopenic purpura, febrile reactions, ‘flu’ syndrome
Rifabutin (RFB) Skin discoloration, uveitis, arthralgia, leucopenia
Pyrazinamide (PZA) Anorexia, nausea, flushing, hepatitis, arthralgia, hyperuricaemia, cutaneous hypersensitivity
Ethambutol (EMB) Retrobulbar neuritis, arthralgia

Treatment of drug resistant TB

Patients with drug-resistant TB should be managed with caution. The DOTS-plus strategy, with the use of multiple agents as guided by DST results, is strongly recommended. A single drug should NOT be added to a failing regimen to avoid progressive acquisition of drug resistance (the addition phenomenon). For mono-drug resistant TB, the suggested treatment regimens in Hong Kong are listed in Box 23.3. Of note, WHO has recommended the use of rifampicin (RIF), EMB, PZA and levofloxacin for 6 months for treatment of RIF-susceptible, INH-resistant pulmonary tuberculosis in both HIV negative patients and HIV-infected patients on HAART in its latest guideline published in 2018.[Further reading B] However, the recommendation was conditional, with very low certainty in the estimates of effects. It is essential that resistance to RIF be excluded by WHO-recommended genotypic or phenotypic methods. Preferably, resistance to fluoroquinolones, and if possible to PZA, is similarly be excluded prior to treatment in order to help avert the acquisition of additional drug resistance. For RR- and MDR-TB, the treatment regimen should comprise at least 4-5 drugs that are likely effective for the initial 6 months, followed by 3-4 drugs to complete the course. In recent years, two new drugs (bedaquiline and delamanid) and two repurposed drugs (linezolid and clofazimine) have been added to the armamentarium for the treatment of MDR-TB. WHO has regrouped the second line anti-TB drugs into three categories and ranked based on the latest evidence about the balance of effectiveness to safety.[9][Box 23.4]. Kanamycin and capreomycin are no longer recommended by WHO, given increased risk of treatment failure and relapse associated with their use in longer MDR-TB regimens in a recent individual data meta-analysis.[10] Confounding by indication and non-comparability between cases and control, however, cannot be excluded in retrospective observational analysis of treatment outcome data. In fact in the same meta-analysis, use of kanamycin and capreomycin were associated, respectively, with very good treatment success rates of 2192 of 2523 (86.9%) and 821 of 938 (87.5%) in drug-susceptible cases. There was also dispute on whether the relatively toxic drug, linezolid, and the expensive new drug, bedaquiline, with very short and incompletely established safety record, are needed in the treatment of MDR-TB in the absence of fluoroquinolone resistance. A final WHO guideline on treatment of RR- and MDR-TB is not available yet at the time of writing of this chapter. Readers are advised to refer to the final WHO guideline for detailed recommendations on the treatment of RR- and MDR-TB when the latter is released. In any case, expert consultation should be sought and reference should be made to local data, because of the complexity of treatment regimen, and frequent DDI or drug-related adverse effects in HIV-infected patients. Using an agent that the patient has not been exposed to, and a bactericidal rather than a bacteriostatic agent is preferred. Daily regimens should be used, except perhaps for the injectable agents (if used). High dose isoniazid may also be considered in the presence of low-level isoniazid resistance.

The total duration of therapy for MDR-TB has not been clearly established. Most experts recommend a total duration of at least 18 months, or 18 months after culture conversion. WHO has suggested a total treatment duration of 18-20 months for most patients, and the duration may be modified according to the patient’s response to therapy (conditional recommendation, very low certainty in the estimates of effect).[9] Local experiences suggest that, with a combination treatment regimen which includes a fluoroquinolone to which the bacilli are susceptible, the total duration of therapy may be shortened to 12-15 months, or 12 months after sputum culture conversion. A recent study among MDR-TB patients naive to second-line drugs also showed that using a combination regimen containing high-dose gatifloxacin and isoniazid can shorten the treatment duration to 9-12 months.

Because of their potential toxicity, second line anti-TB drugs should be used after balancing benefits and risks, and careful clinical and laboratory monitoring are required. Development of newer anti-tuberculous drugs is urgently required; some of the candidates are already put into clinical trials (e.g. Pretomanid (PA-824)).

Box 23.3. Suggested treatment regimens for mono-drug resistant TB

Drug resistance Suggested treatment regimen
SM resistance ‘standard regimen’: RIF/RFB+INH+PZA+EMB (2 months), then RIF/RFB+INH (7 months)
INH resistance RIF/RFB+PZA+EMB+SM (2 months), then RIF/RFB+PZA+EMB (10 months)
RIF resistance* As guided by DST if reliable DST for other drugs are available; treat as MDRTB if reliable DST for other drugs are not available
PZA resistance# RIF/RFB+INH+EMB (2 months), then RIF/RFB+INH (7 months)
*investigation for MDR-TB is recommended; #DST may be unreliable; M. bovis is intrinsically resistant to PZA

Box 23.4. Regrouping of drugs recommended for use in MDR-TB by WHO [9]

Group Drug
A Levofloxacin or Moxifloxacin
B Clofazimine
Cycloserine or Terizidone
C Ethambutol
Imipenem-cilastatin or Meropenem
Amikacin or Streptomycin
Ethionamide or Prothionamide
P-aminosalicylic acid
*Evidence on the safety and effectiveness of Bedaquiline beyond 6 months was insufficient for review; extended Bedaquiline use in individual patients will need to follow ‘off-label’ use best practices.
#Optimal duration of use of Linezolid is not established. Use for at least 6 months was shown to be highly effective, although toxicity may limit its use.
(Source: Rapid Communication: Key changes to treatment of multidrug- and rifampicin-resistant tuberculosis, WHO 2018 Aug)

Concomitant antiretroviral therapy

Rifamycins, especially RIF, induce the CYP3A component of cytochrome P450, can increase the clearance of many protease inhibitors (PI) and some non-nucleoside reverse transcriptase inhibitors (NNRTIs) that are metabolised via the same pathway. This may reduce plasma concentrations of the antiretrovirals to subtherapeutic levels, resulting in the emergence of drug-resistant HIV. Conversely PIs and NNRTIs may either act as enzyme inhibitors or inducers, affecting the level of rifamycins. RIF can also decrease serum levels of integrase inhibitors (INSTI), e.g raltegravir (RAL) and dolutegravir (DTG) and entry inhibitors, e.g. maraviroc (MRV). Compared to RIF, RFB is a less potent enzyme inducer, but has similar efficacy against MTB.

Because efavirenz (EFV) reduces the concentration of co-administered RFB, RIF is the rifamycin of choice for patients taking EFV-based ART. If a first line PI-based regimen is used, RFB should be considered and given at an adjusted dose; monitoring for antimycobacterial activity and therapeutic drug levels may be required.[Box 23.5] Alternatives include a nevirapine (NVP)-based and possibly a RAL-based regimen, with RFB substitution. There is insufficient data on RFB use in patients receiving dual PI or PI/NNRTI combinations. Potential adverse effects of RFB (e.g. uveitis, skin discoloration, neutropaenia) should be monitored, with therapeutic drug monitoring (TDM)[Chapter C13] for both RFB and antiretroviral levels considered when necessary.

Of note, first line antiretrovirals have been gradually been replaced by INSTI such as RAL in recent years. As RIF decreases the trough concentrations of RAL 400 mg twice daily by about 60%, doubling the dose of RAL to 800 mg twice daily in adults taking RIF for TB has been recommended. The trough concentrations were still reduced when compared to RAL 400 mg twice daily without RIF, however. While awaiting further efficacy data of double-dose RAL, use of RFB may be preferred in HIV-TB co-infected persons who are on RAL-based antiretrovirals. Similarly, levels of DTG are reduced by concomitant RIF; its dosage should be doubled to 50 mg twice daily.

As the overall RAL and DTG concentrations are not significantly affected by RFB, and trough RAL and DTG concentrations are diminished by concomitant RFB relatively modestly by 20% and 30%, standard-dose RAL (400 mg twice daily) and DTG (50 mg once daily) with RFB can be used.

For patients who are already on RIF and about to be switched to RFB so that a PI could be given, a washout period of 2 weeks is recommended after discontinuation of RIF. In the meantime, RFB in full dose (i.e. 300 mg per day) is given until the PI is started. Dosage of both RFB and antiretrovirals may need to be adjusted subsequently. Conversely if a PI is to be discontinued, RIF may be started at half dose after 2-3 days of washout and increased to full dose after a week. Updated information on DDI between anti-tuberculous and antiretroviral agents and their respective dosages are available elsewhere (http://www.cdc.gov/tb/TB_HIV_Drugs/default.htm or http://aidsinfo.nih.gov/guidelines/html/1/adult-and-adolescent-arv-guidelines/0/).

Box 23.5. Dosage adjustment of rifabutin (RFB) when used in combination with antiretrovirals (ART)

  EFV NVP ETR (without PI) ATV RTV-boosted PI* RAL DTG MVC
RFB dose
(↑ or ↓)
450-600 mg daily or 600mg 3x/wk (↑) 300 mg daily (–)# 300 mg daily (–) 150 mg daily or 300 mg 3x/wk (↑) 150 mg daily or 300 mg 3x/wk (↑) 300 mg daily (–) 300 mg daily (–) 300 mg daily (–)
ART dose (–) (–) (–) (–) (–) (–) (–) (–)
*Monitor for antimycobacterial activity and consider therapeutic drug monitoring (e.g. ATV/r, LPV/r, FPV/r).
#Use with caution
(EFV, efavirenz; NVP, nevirapine; ETR, etravirine; ATV, atazanavir; LPV, lopinavir; FPV, fos-amprenavir; r, ritonavir; RAL, raltegravir; DTG, dolutegravir; MVC, maraviroc) (no change, –)

Monitoring of treatment

Reactions to anti-tuberculous drugs tend to be more common in HIV-infected patients. Caution should be exercised over potential DDI and additive drug toxicities, especially with the second-line anti-tuberculous drugs. Close monitoring is essential (e.g. for hepatitic symptom, skin rash). Pre-treatment and serial monitoring of liver function and regular visual assessment (with EMB) are recommended. Pyridoxine supplementation is recommended to prevent INH-related peripheral neuropathy. In the event of drug sensitivity or toxicity, ART interruption may result in resistance that is difficult to manage. Similarly, re-initiation of TB treatment following interruption is complicated and may also result in resistance. Expert consultation is required. As regards monitoring of response to treatment, sputum culture conversion to negative state usually occurs within ≤2 months in majority (~85%) of drug susceptible TB cases. Persistent culture positivity ≥4 months may indicate non-compliance and/or development of drug resistance. Drug adherence can be checked by urine testing for isoniazid metabolite, and serum RIF level assay. Repeated imaging may sometimes be indicated to document resolution of disease, especially for extrapulmonary sites which are inaccessible for repeated bacteriological examinations.

Immune reconstitution inflammatory syndrome (IRIS) [Chapter C17]

Paradoxical worsening of disease after an initial period of improvement with antituberculous treatment may occur when a patient is started on ART. There may be recurrence of fever, worsening of chest infiltrates on radiograph, increasing pleural effusions, enlarging peripheral and/or mediastinal lymph nodes, and expanding CNS lesions, but typically, evidence of active infection or disease dissemination is absent (e.g. culture generally remains negative). Such reactions are believed to be associated with partial restoration of immune function and exacerbation of inflammation after antiretroviral therapy, thus commonly termed as immune reconstitution inflammatory syndrome (IRIS).[Chapter C17] IRIS may occur in up to 36% of TB patients receiving HAART, and may appear as soon as 2 weeks. In contrast, paradoxical reactions occur in only about 7% of non-HIV patients with anti-tuberculous treatment alone. Sometimes, antiretroviral-related IRIS may unmask a previously undiagnosed TB infection. Risk factors for IRIS include a very low baseline CD4 count (e.g. <50 μL), a short interval between starting anti-TB treatment and ART, disseminated TB and a high HIV-1 viral load. In the absence of a specific test, diagnosis of IRIS requires careful exclusion of on-going uncontrolled infection, non-adherence, treatment failure, drug-resistance, drug fever, and other opportunistic infections and malignancies such as lymphoma. Rapid drop in HIV viral load and exaggerated increase in CD4 cell count shortly after ART initiation are important diagnostic clues. For most cases of IRIS, the reactions are mild to moderate and symptomatic treatment is adequate. Discontinuation of ART is not required. Occasionally, severe inflammation can result in excessive tissue damage or even fatality (e.g. intracranial lesions). Patients with TB meningitis have been reported to have a mortality rate of up to 30 % following development of IRIS in some studies. Corticosteroids may be considered in such circumstances (e.g. prednisolone 1 mg/kg/d for initial 1-2 weeks, then taper off within a few weeks).

Regarding prevention of TB-IRIS, prednisone initiated alongside ART in selected patients with CD4 less than 100/μL has been shown to reduce the risk of paradoxical TB-IRIS by 30% in a recent randomised-controlled trial.[11] Prednisolone use was not associated with an excess risk of severe infections, cancers, or adverse events. Based on this result, preventive treatment with prednisolone may be considered for HIV-infected TB patients with a CD4 nadir less than 100/μL, who have had hepatitis B and Kaposi’s sarcoma excluded, and who are symptomatically improving on TB treatment prior to ART.

Diagnosis and treatment of latent TB infection

Diagnosis and treatment of latent TB infection (LTBI) are indicated in HIV-infected patients because of their excessive risks of reactivation. Treatment of LTBI decreases the risk of active TB disease by 62%.[12] Conventionally LTBI is diagnosed by a positive tuberculin skin test (TST), as defined by an induration size ≥5mm in the HIV-infected, after ruling out active disease. Newer blood tests based on assay of IFN-γ released (IGRA) by T cells in response to M. tuberculosis-specific antigens are not influenced by previous BCG vaccination or exposure to environmental mycobacteria and have several operational advantages (e.g. no return visit required, reading of TST result is operator independent, no boosting effect with repeated tuberculin testing). IGRAs has gained increasing acceptance in high income areas, particularly for screening of BCG-vaccinated subjects. Its predictive value for incident active TB is higher than TST in high income countries with low or intermediate burden of disease.

In Hong Kong, the Scientific Committee on AIDS and STI recommended screening for TB in all HIV-infected subjects by performing annual TST.[HK Guidelines 23A] IGRA is an acceptable alternative to TST. At a low CD4 count <100/μL, dual testing by TST and IGRA is advised to increase case finding, as most incident TB cases occur at this CD4 range. A positive result with either test is an indication for treatment. There is no evidence for protective efficacy of preventive treatment among HIV-infected subjects with a negative TST, but preventive treatment may still be considered for anergic HIV-infected persons with significant recent exposure to an infectious source of TB. Of note, LTBI testing by either TST or IGRA is not a strict requirement for initiating preventive treatment in PLWHA in resource-constrained areas with high ongoing risk of TB transmission, as recommended in the WHO guidelines in 2018.

In Hong Kong, the recommended treatment regimen for LTBI in HIV-infected persons is INH 300mg/day plus pyridoxine 10-50mg/day for a total of 9 months (or INH 900mg 2x/week plus pyridoxine for 9 months under DOT). Prolonged treatment with INH for >36 months has been found to be more effective than 6 months’ treatment among TST-positive HIV-coinfected patients in one but not another study.[13][14] Rifapentine 900 mg plus INH 900 mg weekly for 12 weeks has also been shown to be equally effective as 9 months of INH.[13] Of note, an ultra-short course TB preventive therapy consisting of once daily rifapentine 450-600 mg plus INH 300 mg for 1 month has been reported to be non-inferior in efficacy to INH given for 9 months and had fewer adverse events in a recent multicentre, randomised, open-label, phase 3 trial that enrolled HIV-infected individuals >13 y living in high TB-burden areas or who were TST/IGRA positive (ACTG 5279). Modification of local recommendation will await further efficacy data and field experiences of newer treatment regimens, especially when used concomitantly with antiretroviral therapy. For those patients who are contacts of MDR-TB, expert consultation should be sought to evaluate if treatment is warranted and with what regimen, given that evidence is relatively scarce under this scenario.

Effectiveness of the existing policy to treat LTBI identified by annual TST for TB control in PLWHA in Hong Kong has been demonstrated in a local study conducted in 2013.[15] The high risk of TB during the early period of ART supports early use of TST and/or IGRA to screen for LTBI. On the other hand, given the reduction in TB disease transmission risk in recent years locally, it may be time to re-visit if annual TST for LTBI is required for PLWHA who test negative on initial evaluation. Further data are needed to assess whether annual TST may be offered only to PLWHA with additional risk factors for TB or who have a risk for continual exposure to MTB.

Algorithm 23. Management of TB in HIV infection

Algorithm 23. Management of TB in HIV infection


  1. Getahun H, Gunneberg C, Granich R, Nunn P. HIV infection-associated tuberculosis: the epidemiology and the response. Clin Infect Dis 2010;50(Suppl 3):S201-7. link
  2. Balcha TT, Skogmar S, Sturegård E, Schön T, Winqvist N, Reepalu A, Jemal ZH, Tibesso G, Björk J, Björkman P. A Clinical Scoring Algorithm for Determination of the Risk of Tuberculosis in HIV-Infected Adults: A Cohort Study Performed at Ethiopian Health Centers. Open Forum Infect Dis 2014; 1(3):ofu095. link
  3. Cain KP, McCarthy KD, Heilig CM, Monkongdee P, Tasaneeyapan T, Kanara N, Kimerling ME, Chheng P, Thai S, Sar B, Phanuphak P, Teeratakulpisarn N, Phanuphak N, Nguyen HD, Hoang TQ, Le HT, Varma JK. An algorithm for tuberculosis screening and diagnosis in people with HIV. N Engl J Med 2010;362(8):707-16. link
  4. World Health Organization. Antiretroviral therapy for HIV infection in adults and adolescents: recommendations for a public health approach, 2010 revision. Geneva: WHO, 2010. link
  5. Havlir DV, Kendall MA, Ive P, Kumwenda J, Swindells S, Qasba SS, Luetkemeyer AF, Hogg E, Rooney JF, Wu X, Hosseinipour MC, Lalloo U, Veloso VG, Some FF, Kumarasamy N, Padayatchi N, Santos BR, Reid S, Hakim J, Mohapi L, Mugyenyi P, Sanchez J, Lama JR, Pape JW, Sanchez A, Asmelash A, Moko E, Sawe F, Andersen J, Sanne I; AIDS Clinical Trials Group Study A5221. Timing of antiretroviral therapy for HIV-1 infection and tuberculosis. N Engl J Med 2011;365(16):1482-91. link
  6. Abdool Karim SS, Naidoo K, Grobler A, Padayatchi N, Baxter C, Gray AL, Gengiah T, Gengiah S, Naidoo A, Jithoo N, Nair G, El-Sadr WM, Friedland G, Abdool Karim Q. Integration of antiretroviral therapy with tuberculosis treatment. N Engl J Med 2011;365(16):1492-501. link
  7. Török ME, Yen NT, Chau TT, Mai NT, Phu NH, Mai PP, Dung NT, Chau NV, Bang ND, Tien NA, Minh NH, Hien NQ, Thai PV, Dong DT, Anh do TT, Thoa NT, Hai NN, Lan NN, Lan NT, Quy HT, Dung NH, Hien TT, Chinh NT, Simmons CP, de Jong M, Wolbers M, Farrar JJ. Timing of initiation of antiretroviral therapy in human immunodeficiency virus (HIV)-associated tuberculous meningitis. Clin Infect Dis 2011;52(11):1374-83. link
  8. Ahmad Khan F, Minion J, Al-Motairi A, Benedetti A, Harries AD, Menzies D. An updated systematic review and meta-analysis on the treatment of active tuberculosis in patients with HIV infection. Clin Infect Dis 2012;55(8):1154-63. link
  9. World Health Organization. WHO treatment guidelines for multidrug- and rifampicin-resistant tuberculosis, 2018 update. [Pre-final text]. Geneva: World Health Organization, WHO/CDS/TB/2018.15 link
  10. Collaborative Group for the Meta-Analysis of Individual Patient Data in MDR-TB treatment-2017; Ahmad N, Ahuja SD, Akkerman OW, Alffenaar JC, Anderson LF, Baghaei P, Bang D, Barry PM, Bastos ML, Behera D et al. Treatment correlates of successful outcomes in pulmonary multidrug-resistant tuberculosis: an individual patient data meta-analysis. Lancet 2018;392:821-34. link
  11. Meintjes G, Stek C, Blumenthal L, Thienemann F, Schutz C, Buyze J, Ravinetto R, van Loen H, Nair A, Jackson A, Colebunders R, Maartens G, Wilkinson RJ, Lynen L; PredART Trial Team. Prednisone for the Prevention of Paradoxical Tuberculosis-Associated IRIS. N Engl J Med 2018;15;379(20):1915-1925. link
  12. Akolo C, Adetifa I, Shepperd S, Volmink J. Treatment of latent tuberculosis infection in HIV infected persons. Cochrane Database Syst Rev. 2010(1):CD000171. link
  13. Samandari T, Agizew TB, Nyirenda S, Tedla Z, Sibanda T, Shang N, Mosimaneotsile B, Motsamai OI, Bozeman L, Davis MK, Talbot EA, Moeti TL, Moffat HJ, Kilmarx PH, Castro KG, Wells CD. 6-month versus 36-month isoniazid preventive treatment for tuberculosis in adults with HIV infection in Botswana: a randomised, double-blind, placebo-controlled trial. Lancet 2011;377(9777):1588-98. link
  14. Sterling TR, Scott NA, Miro JM, Calvet G, La Rosa A, Infante R, Chen MP, Benator DA, Gordin F, Benson CA, Chaisson RE, Villarino ME; Tuberculosis Trials Consortium, the AIDS Clinical Trials Group for the PREVENT TB Trial (TBTC Study 26ACTG 5259). Three months of weekly rifapentine plus isoniazid for treatment of M. tuberculosis infection in HIV co-infected persons. AIDS 2016;30(10):1607-15. link
  15. Lin AW, Chan KC, Chan WK, Wong KH. Tuberculin sensitivity testing and treatment of latent tuberculosis remains effective for tuberculosis control in human immunodeficiency virus-infected patients in Hong Kong. Hong Kong Med J 2013;19(5):386-92. link

Further reading

  1. World Health Organization. Guidelines for treatment of drug-susceptible tuberculosis and patient care (2017 update). Geneva: WHO, 2017. WHO/HTM/TB/2017.05. link
  2. World Health Organization. WHO treatment guidelines for isoniazid-resistant tuberculosis: Supplement to the WHO treatment guidelines for drug-resistant tuberculosis. Geneva: World Health Organization 2018 WHO/CDS/TB/2018.7. link

Hong Kong Guidelines

  1. Scientific Committee on AIDS and STI, Centre for Health Protection, Hong Kong. Recommendations on the management of HIV and tuberculosis co-infection. Hong Kong: Department of Health, 2015. Available from APPENDIX II: X6 and link