The Use of Therapeutic Drug Monitoring in Complex Antituberculous and Antiretroviral Drug Dosing in HIV/Tuberculosis-Coinfected Patients

Abstract

We report 2 cases coinfected with HIV and tuberculosis (HIV/TB), requiring drug dose adjustments guided by therapeutic drug monitoring (TDM) and/or serum drug concentrations.

Case 1:

Over the course of the 9-months of TB treatment, drugs that required increased doses due to low concentrations included efavirenz (800 mg), rifampin (900 mg), and isoniazid (450 mg). Higher drug doses were well tolerated until the end of treatment.

Case 2:

Over the 12-month course of TB therapy, drugs that required increased doses due to incomplete and/or delayed absorption were rifampin (1500 mg), moxifloxacin (800 mg), and ethambutol (1600 mg). Higher drug doses were well tolerated until the end of treatment. Due to delayed/incomplete drug absorption and weight gain during therapy, higher antituberculous doses may be required in patients coinfected with HIV/TB. A daily dose of efavirenz 800 mg was well tolerated in both patients (weight over 70 kg). Managing patients coinfected with HIV/TB is complex, and, therefore, TDM of drug concentrations can help guide clinical decision making.

Keywords

HIV/TB antituberculous drugs drug interactions efavirenz therapeutic drug monitoring

Introduction

The treatment of patients coinfected with HIV/tuberculosis (HIV/TB) can be challenging, given the potential of underlying patient comorbidities, severe immunosuppression, patient wasting, and drug-related factors such as adherence to therapy, polypharmacy, drug malabsorption, toxicity and resistance, and complex drug interactions.¹ In treating patients coinfected with HIV/TB, regimens without rifampin or with shorter rifampin treatment length are considered less effective than those that used rifampin for the full treatment period.² Because of drug interactions and additive toxicities, only a few antiretroviral (ARV) drugs can be safely coadministered with the rifamycin class. The Department of Health and Human Services Adult Antiretroviral Guidelines provide recommendations on combinations that are suitable for coadministration.³ For instance, rifampin can be given with efavirenz (EFV); however, there is debate regarding the correct dose of EFV to be administered when stratifying based on patient weight. The guidelines recommend to maintain the EFV dose at 600 mg daily and to monitor closely for virologic response, and therapeutic drug monitoring (TDM) may also be warranted. According to the guidelines, some clinicians prefer to increase the EFV dose to 800 mg daily in patients of more than 60 kg.

While guidelines are useful in providing information on empiric drug dosing and potential drug interactions, little guidance is available on how to manage patients clinically when TDM is performed. In our program, we routinely perform antituberculous TDM in all HIV/TB-coinfected patients and ARV TDM in select complex cases that involve drug interactions, drug malabsorption, nonadherence, failing therapy, and/or drug toxicity.^4
–6 Herein, we report the management of 2 patients with complex drug dosing secondary to delayed drug absorption, drug malabsorption, and drug interactions.

Case 1

A 36-year-old HIV-positive foreign-born male from sub-Saharan Africa was initiated on ARV therapy, consisting of a once-daily fixed-dose combination of tenofovir 300 mg, emtricitabine 200 mg, and EFV 600 mg. His baseline CD4 count was 90 cells/mm³ and HIV RNA was 760 000 copies/mL. Two weeks after ARV initiation, he presented with marked left cervical lymphadenitis. A cervical lymph node biopsy confirmed the presence of necrotizing granulomatous inflammation with evidence of acid-fast bacilli on staining. Culture results confirmed the presence of Mycobacterium tuberculosis complex. Sputum and chest radiograph findings were unremarkable. He was initiated on isoniazid 300 mg (4.3 mg/kg), rifampin 600 mg (8.6 mg/kg), pyrazinamide 1500 mg (21.6 mg/kg), ethambutol 1200 mg (17.2 mg/kg) and (patient weight 69.6 kg), all 5 times per week under directly observed therapy (DOT) (patient weight 69.6 kg). Two months later, Mycobacterium bovis was speciated and found to be sensitive to first-line agents, with the exception of pyrazinamide, to which this organism is inherently resistant. At this point, pyrazinamide and ethambutol were discontinued and isoniazid and rifampin were continued for a total of 9 months under DOT. The patient was perfectly adherent to antituberculous therapy.

One month after starting the treatment for M bovis lymphadenitis, routine TDM was performed on antituberculous drugs and EFV (the patient weight increased to 71 kg). Results, interpretation, and reference ranges are shown in Table 1.^5
–7 At this time, the patient had a CD4 count of 210 cells/mm³ and an HIV RNA of 49 copies/mL and the lymphadenitis almost resolved, with no complaints of cough, fever, or night sweats.

Table 1.

Dosing of Antimycobacterial and Efavirenz Based on Therapeutic Drug Monitoring.

Month	Drug	Dosage, mg	Dose, mg/kg	2-Hour postdose, µg/mL	6-Hour postdose, µg/mL	10-Hour postdose, µg/mL	Therapeutic Range,^5 –7 µg/mL	Interpretation/Recommendation
Case 1
2	RIF	600	8.5	6.37			8-24	Low; ↑ to 900 mg
2	INH	300	4.2	3.6			3-6	Normal
2	EFV	600	8.5			1.1	1-4	Low–normal; ↑ to 800 mg; 24-hour trough (mg/L)
4	EFV	800	10.3			1.5	1-4	Normal; 24-hour trough (mg/L)
4	RIF	900	11.6	8.71			8-24	Normal
4	INH	300	3.9	2.04			3-6	Low; ↑ to 450 mg
Case 2
1	INH	300	4.1	2.53	1.79		3-6	Low; ↑ to 450 mg
1	RIF	600	8.3	6.21	2.91		8-24	Low
1	INH	450	6.2	Trace	2.68		3-6	Low and delayed; D/C—hepatotoxicity
1	RIF	600	8.3	0	5.8		8-24	Low and delayed; ↑ to 900 mg
3	RIF	900	12.4	Trace	4.01		8-24	Low and delayed; ↑ to 1200 mg
4	RIF	1200	15.9	Trace	4.76		8-24	Low and delayed; ↑ to 1500 mg
5	RIF	1500	19.9	0	2.75		8-24	Low
6 and 7	ETB	1000	11.5	0.82	0.71	0.39	2-6	Low; ↑ to 1600 mg
7	EFV	800	9.2			4.84	1-4	Slightly high; 10.5-hour postdose (mg/L); dose maintained
7	RIF	1500	17.3		19.71	8.55	8-24	Delayed, but adequate
6 and 7	MOXI	600	6.9	2.01	1.86	1.49	3-5	Low and delayed; ↑ to 800 mg
8	RIF	1500	17.3	Trace	17.64		8-24	Delayed, but adequate
8	MOXI	800	9.2	0.27	3.2		3-5	Delayed, but adequate
8	ETB	1600	18.5	0.44	2.64		2-6	Delayed, but adequate

Abbreviations: EFV, efavirenz; ETB, ethambutol; INH, isoniazid; MOXI, moxifloxacin; RIF, rifampin; D/C, discontinued.

In response to the subtherapeutic rifampin serum drug concentration (SDC), rifampin was increased to 900 mg (12.9 mg/kg) 5 times per week under DOT. Given the borderline low EFV trough concentration with the 600-mg daily dose, the dose was increased to 800 mg once daily. Approximately 1 month after the dosing changes to both drugs, TDM was repeated and the results are shown in Table 1. At this time, his HIV RNA was <40 copies/mL, and the patient had also gained 8 kg since the initiation of antituberculous drugs. As a result of the low isoniazid SDC, the dose was increased to 450 mg (5.8 mg/kg) 5 times/week and rifampin 900 mg was maintained. The patient continued to tolerate the higher drug doses well through to the end of treatment and the laboratory findings, including liver enzymes, were stable. The patient’s lymphadenitis resolved and he remained clinically stable 12 months post-TB treatment. His CD4 count had increased to 420 cells/mm³ and HIV RNA was <40 copies/mL.

Case 2

A 46-year-old HIV-positive foreign-born male from sub-Saharan Africa presented with a 14-kg weight loss, anorexia, decreased energy, and night sweats. He was ARV naive and had a CD4 count of 97 cells/mm³ and an HIV RNA of 105 543 copies/mL. He was initiated on a once daily fixed-dose combination of tenofovir 300 mg, emtricitabine 200 mg, and EFV 600 mg. A chest radiograph was significant for right upper lobe fibrotic changes suggestive of prior granulomatous infection. His sputum was smear negative but culture positive for M tuberculosis complex. Therefore, for treatment of pulmonary TB (patient weight 72.5 kg), daily antituberculous therapy was started, which consisted of isoniazid 300 mg (4.1 mg/kg), rifampin 600 mg (8.3 mg/kg), pyrazinamide 1800 mg (24.8 mg/kg), ethambutol 1400 mg (19.3 mg/kg), and moxifloxacin 600 mg (8.2 mg/kg). A 5-drug regimen was used due to concerns about possible drug-resistant TB. Ten days post ARV initiation, the EFV dose was also increased to 800 mg daily in anticipation of a potential interaction with rifampin. Subsequent susceptibility confirmed M tuberculosis that was sensitive to first-line antituberculous agents.

On day 4 of TB therapy, 2- and 6-hour postdose SDCs revealed subtherapeutic concentrations for isoniazid and rifampin (Table 1). Two days later, significant increases were observed in serum transaminases (alanine transaminase [ALT] and aspartate transaminase [AST] approximately 2-5 times the upper limit of normal), and TB medications were held for almost 1 week. The medications were reintroduced at the following doses: isoniazid increased to 450 mg (6.2 mg/kg) and ethambutol decreased to 1200 mg (16.5 mg/kg), but rifampin and moxifloxacin were maintained at the same doses. Pyrazinamide was discontinued. Repeat SDCs revealed delayed absorption and subtherapeutic concentrations for both isoniazid and rifampin (Table 1). Unfortunately, after a few days of restarting the medications, again significant elevations in serum transaminases were observed (ALT and AST 10-20 times the upper limit of normal) and therefore isoniazid was discontinued.

Once again, TB therapy was restarted, with rifampin 750 mg (10.3 mg/kg), ethambutol 1000 mg (13.7 mg/kg), and moxifloxacin 600 mg (8.2 mg/kg) 5 times weekly under DOT. Over the next few months, the dose of rifampin was gradually titrated to 1500 mg (17.3 mg/kg) due to the consistently low 6-hour SDCs (patient weight 86.7 kg). After this increase to 1500 mg, the 6-hour postdose rifampin concentration was finally therapeutic at 19.71 µg/mL. Delayed absorption and low SDCs were also observed with ethambutol and moxifloxacin, resulting in dosage increase of 1600 mg (18.5 mg/kg) and 800 mg (9.2 mg/kg), respectively. During the time of suspected antituberculous drug malabsorption and drug titration, EFV plasma concentrations were also assessed. A 10.5-hour mid-dosing interval sample was slightly supratherapeutic at 4.84 mg/L (1-4 mg/L); however, the 800-mg dose was maintained as the drug was well tolerated.⁵

During the remainder of TB therapy (a total of 12 months), the patient tolerated the medications well and was perfectly adherent to therapy. At 2 months into TB treatment, the chest radiograph remained stable when compared to baseline and sputum culture had converted to negative. The CD4 count increased to 305 cells/mm³ and the HIV RNA was suppressed at 9 months of TB therapy, which is the last available data because after completing therapy, the patient lost to follow-up. As a result, no final clinical assessment was completed. Prior to this, he had improved clinically with weight gain and resolution of night sweats and fatigue.

Discussion

These case reports provide a practical clinical description of how TDM can guide the dosing of antituberculous and ARV drugs in HIV/TB coinfection. In our previously described cohort of patients coinfected with HIV/TB who had completed antituberculous TDM, 20 (74%) of the 27 patients had a low SDC for at least 1 drug and underwent subsequent dose modification.⁸ These 2 cases were selected as they provide an illustration of low SDCs and/or delayed absorption of antituberculous drugs. As shown in case 1, rifampin 900 mg daily and isoniazid 450 mg daily were required to achieve therapeutic SDCs. In case 2, rifampin 1500 mg daily, moxifloxacin 800 mg daily, and ethambutol 1600 mg daily were required to reach therapeutic but delayed peak SDCs. Of note, rifampin required much higher dose adjustments above standard dosing to achieve a therapeutic concentration. Further, while our TB program does not routinely perform ARV TDM in all patients coinfected with HIV/TB, both reported cases underwent EFV TDM to determine the appropriate dose within the clinical context, further illustrating the complexities of dosing in this population.

Inadequate absorption of first-line antituberculous therapy, including both isoniazid and/or rifampin in HIV/TB coinfection, has been observed.^1,8

–11 In both HIV-infected and noninfected TB patients, low SDCs have been associated with treatment failure, relapse, and acquired drug resistance, and drug dose adjustments have been related to clinical improvements.^10,12

–15 Higher rates of acquired resistance to rifampin have been associated with HIV infection, and pharmacokinetic data suggest that reduced antituberculous SDCs observed with intermittent rifamycin dosing may be a contributing factor.^12,13 Due to these findings, TDM is supported in HIV/TB treatment guidelines in certain situations such as the possibility of reduced drug absorption, the potential for complex or difficult to predict drug interactions, non-EFV-based regimens, failure to respond to treatment, or if less well-studied treatment is used.^2
–4 However, the relationship between inadequate SDCs and TB treatment outcomes requires further study.

Given the frequency of reduced antituberculous SDCs reported in the literature, the standard rifampin 600 mg dose and isoniazid 300 mg dose may be suboptimal and weight-based dosing of 10 and 5 mg/kg/d, respectively, should be considered.⁸ Based on TDM, it has been suggested that the median doses required to achieve therapeutic concentrations in an HIV/TB cohort were isoniazid 600 mg/d (range, 300-1500 mg/d), rifampin 1050 mg/d (range, 600-1200 mg/d), and rifabutin 300 mg (range, 150-450 mg 3 times/week).¹ Although limited literature exists regarding the safety of higher antituberculous dosing, no patients experienced adverse effects secondary to these higher drug doses in this study.

In HIV/TB coinfection, rifampin is central to the treatment of active TB and has been associated with improved clinical outcomes and shorter duration of treatment. Studies have demonstrated that antituberculous regimens without rifampin had higher rates of treatment failure and relapse.² Concurrent management of HIV/TB is challenging due to interactions related to rifampin, which is a potent inducer of the hepatic cytochrome P (CYP) 450 enzymes. Rifampin significantly interacts with most ARV drugs including all protease inhibitors (PIs), nonnucleoside reverse transcriptase inhibitors, maraviroc, and integrase inhibitors. In managing the potential interactions with rifampin, the most compatible ARV drug options (when used with 2 nucleoside reverse transcriptase inhibitors) include EFV (600 or 800 mg once daily for patients weighing >60 kg), raltegravir (doubling the dose to 800 mg twice daily), or dolutegravir (50 mg twice daily).^2,3 Coadministration of PIs is not recommended due to hepatotoxicity and significant decreases in plasma concentrations of these agents when coadministered with rifampin.^2,3,16,17

Rifabutin, a less potent enzyme inducer, can be considered for use with PIs; however, decreased rifabutin dosing is recommended (ie, 300 mg thrice weekly or 150 mg daily) along with rifabutin TDM, if available.^2,3

As EFV remains a commonly used first-line ARV drug, its coadministration with rifampin in patients coinfected with HIV/TB is also common. Pharmacokinetic data from HIV-infected patients showed a 22% decrease in area under the curve and 24% decrease in the peak concentration of EFV with concomitant rifampin use.^18,19 Based on this, some experts recommend increasing the EFV dose to 800 mg daily in patients who weigh more than 50 or 60 kg.^2,3,18,20 However, a wide degree of interindividual variability in EFV plasma concentrations has been observed, thus making EFV weight-based dosing somewhat unreliable.^21,22 Efavirenz is mainly metabolized through hepatic CYP450 CYP3A4 and 2B6, and recent studies have shown that some ethnicities may be predisposed to slow-metabolizing genetic polymorphisms of CYP2B6 and may not require the dose increase.^2,23 Furthermore, EFV 800 mg daily has not been consistently shown to result in superior virologic suppression and increased concentrations have been associated with central nervous system toxicities.^23,24 Thus, when coadministered with rifampin, EFV dose adjustments should likely be individualized based on clinical and virologic response and TDM when available, rather than on weight alone.⁷

As with antituberculous TDM, the use of ARV TDM is supported in guidelines in certain situations including (but not limited to) clinically significant drug–drug interactions that may result in reduced efficacy or increased dose-related toxicities and pathophysiologic states that may alter drug absorption.^3,5 In our case reports, in addition to monitoring for clinical and virologic response to treatment, TDM assisted in the evaluation of the EFV dose. Both patients weighed between 70 and 90 kg throughout their TB treatment. In case 1, the patient had a low but still a therapeutic EFV trough plasma concentration of 1.1 mg/L on the 600-mg daily dose, while the dose increase to EFV 800 mg resulted in a slight increase in concentration to 1.5 mg/L. For case 2, EFV 800 mg demonstrated a slightly supratherapeutic plasma concentration of 4.84 mg/L at the mid-dosing interval. Due to geographical issues and the long delay in obtaining the TDM results, the dose was maintained in this patient since it was well tolerated. Neither patient experienced adverse effects related to the 800-mg dose. Thus, the EFV dose to be used should be based on the ability to monitor efficacy (HIV viral suppression) and the potential for central nervous system toxicity with increased EFV doses. Based on these case reports it is unlikely that the rifampin-induced decrease in EFV concentration in isolation significantly affects viral suppression, especially when compared to the larger contributors to poor viral suppression such as poor adherence to therapy.²

Our case reports also highlight the importance of the timing of antituberculous SDC measurement. Both malabsorption and delayed absorption can occur more commonly in the HIV population and 6-hour and even 10-hour postdose concentrations can distinguish between delayed absorption (therapeutic 6-hour concentration) and malabsorption (persistently low 6-hour concentration).⁶ In case 2, antituberculous drug sampling at 2-, 6-, and 10-hour postdose captured delayed peak concentrations. Other factors that can affect antituberculous drug dosing include potentially dramatic increases in weight gain with therapy, especially if wasting had occurred prior to initiation of TB or ARVs. Additional factors include the presence of gastrointestinal infection, a low CD4 count due to HIV enteropathy, or the presence or absence of food with drug administration.²⁵

Conclusion

Therapeutic drug monitoring is a helpful clinical tool when assessing absorption and titrating drug doses in patients coinfected with HIV/TB. These 2 cases reports illustrate the complexities of managing these concurrent infections and confirm the observation of low antituberculous drug concentrations, the presence of delayed absorption, tolerance of higher doses of antimycobacterials, and the need to individualize EFV dosing. Despite the complex issues, utilization of TDM can help guide clinical decision making.

Footnotes

Acknowledgments

The authors acknowledge the important clinical contributions of Avril Beckon, Sandy Cockburn, and Jeff Kapler to the preparation of this article.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Michelle Foisy received speaker honoraria and/or unrestricted educational grants from ViiV Healthcare, Gilead Sciences Canada Inc, Janssen Pharmaceuticals, Bristol-Myers Squibb, and Merck Canada.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Holland

Hamilton

Weintrob

. Therapeutic drug monitoring of antimycobacterial drugs in patients with both tuberculosis and advanced human immunodeficiency virus infection. Pharmacotherapy. 2009;29 (5):503–510. doi:10.1592/phco.29.5.503.

US Centers for Disease Control and Prevention. Managing drug interactions in HIV-related tuberculosis [online]; 6 2013. Web site. http://www.cdc.gov/tb/TB_HIV_Drugs/default.htm. Accessed October 30, 2014.

Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services; 5 2014. Web site. http://www.aidsinfo.nih.gov/contentfiles/lvguidelines/adultandadolescentgl.pdf. Accessed October 30, 2014.

Canadian Thoracic Society and Public Health Agency of Canada. Canadian Tuberculosis Standards. 7th ed. Ottawa: Canadian Thoracic Society and Public Health Agency of Canada; 2013.

la Porte

CJL

Back

Blaschke

. Updated guideline to perform therapeutic drug monitoring for antiretroviral agents. Rev Antiviral Ther. 2006;3:4–14.

Peloquin

. Therapeutic drug monitoring in the treatment of tuberculosis. Drugs. 2002;62 (15):2169–2183. doi:10.2165/00003495-200262150-00001.

Marzolini

Telenti

Decosterd

Greub

Biollaz

Buclin

. Efavirenz plasma levels can predict treatment failure and central nervous system side effects in HIV-1-infected patients. AIDS. 2001;15(1):71–75. doi:10.1097/00002030-200101050-00011.

Ahmed

Cooper

Foisy

Der

Kunimotto

. Factors associated with reduced antituberculous serum drug concentrations in patients with HIV/TB coinfection. J Int Assoc Physicians AIDS Care. 2012;11 (5): 273–276. doi:10.1177/1545109712454454.

McIlleron

Rustomjee

Vaheda

. Reduced antituberculosis drug concentrations in HIV-infected patients who are men or have low weight: implications for international dosing guidelines. Antimicrob Agents Chemother. 2012;56 (6):3232–3238. doi:10.1128/AAC.05526-11.

10.

Tappero

Bradford

Agerton

. Serum concentrations of antimycobacterial drugs in patients with pulmonary tuberculosis in Botswana. Clin Infect Dis. 2005;41 (4):461–469. doi:10.1086/431984.

11.

Chideya

Winston

Peloquin

. Isoniazid, rifampin, ethambutol, and pyrazinamide pharmacokinetics and treatment outcomes among a predominantly HIV-Infected cohort of adults with tuberculosis from Botswana. Clin Infect Dis. 2009;48 (12):1685–1694. doi:10.1086/599040.

12.

Weiner

Burman

Vernon

. Low isoniazid concentration associated with outcome of tuberculosis treatment with once weekly isoniazid and rifapentine. Am J Respir Crit Care Med. 2003;167 (10):1341–1347.

13.

Weiner

Benator

Burman

. Association between acquired rifamycin resistance and the pharmacokinetics of rifabutin and isoniazid among patients with HIV and tuberculosis. Clin Infect Dis. 2005;40 (10):1481–1491. doi:10.1086/429321.

14.

Kimerling

Phillips

Patterson

Hall

Robinson

Dunlap

. Low serum antimycobacterial drug levels in non-HIV infected tuberculosis patients. Chest. 1998;113 (5):1178–1183.

15.

Mehta

Shantaveerapa

Byrd

Jr Morton

Fountain

Roy

. Utility of rifampin blood levels in the treatment and follow-up of active pulmonary tuberculosis in patients who were slow to respond to routine directly observed therapy. Chest. 2001;120 (5):1520–1524. doi:10.1378/chest.120.5.1520.

16.

Isentress (Raltegravir). Kirkland, QC: Merck Frosst Canada Ltd; 2012.

17.

Tivicay (Dolutegravir). Research Triangle Park, NC: ViiV Healthcare; 2013.

18.

López-Cortés

Ruiz-Valderas

Viciana

. Pharmacokinetic interactions between efavirenz and rifampicin in HIV-infected patients with tuberculosis. Clin Pharmacokinet. 2002;41 (9):681–690.

19.

Pfister

Labbe

Hammer

. Population pharmacokinetics and pharmacodynamics of efavirenz, nelfinavir, and indinavir: Adult AIDS Clinical Trial Group Study 398. Antimicrob Agents Chemother. 2003;47 (1):130–137. doi:10.1128/AAC.47.1.130-137.2003.

20.

Sustiva (Efavirenz). Montreal, QC: Bristol-Myers Squibb Canada; 2012.

21.

Ståhle

Moberg

Svensson

Sönnerborg

. Efavirenz plasma concentrations in HIV-infected patients: inter- and intraindividual variability and clinical effects. Ther Drug Monit. 2004;26 (3):267–270.

22.

Luetkemeyer

Rosenkranz

. Relationship between weight, efavirenz exposure, and virologic suppression in HIV-infected patients on rifampin-based tuberculosis treatment in the AIDS Clinical Trials Group A5221 STRIDE Study. Clin Inf Dis. 2013;57 (4):586–593. doi:10.1093/cid/cit246.

23.

Orrell

Cohen

Conradie

. Efavirenz and rifampicin in the South African context: is there a need to dose-increase efavirenz with concurrent rifampicin therapy? Antivir Ther. 2011;16 (4):527–534. doi:10.3851/IMP1780.

24.

Gutiérrez

Navarro

Padilla

. Prediction of neuro-psychiatric adverse events associated with long term efavirenz therapy, using plasma level monitoring. Clin Infect Dis. 2005;41 (11):1648–1653. doi: 10.1086/497835.

25.

Pozniak

Coyne

Miller

. British HIV association guidelines for the treatment of TB/HIV coinfection 2011. HIV Med. 2011;12 (9):517–524. doi:10.1111/j.1468-1293.2011.00954.x.