Share this article on:

Pharmacology of second-line antituberculosis drugs and potential for interactions with antiretroviral agents

Coyne, Katherine Ma; Pozniak, Anton La; Lamorde, Mohammedb; Boffito, Martaa

doi: 10.1097/QAD.0b013e328326ca50
Editorial Review

aHIV/GUM Directorate, St Stephen's Centre, Chelsea & Westminster NHS Foundation Trust, London, UK

bResearch Department, Infectious Diseases Institute, Makerere University, Uganda.

Received 7 October, 2008

Revised 11 December, 2008

Accepted 12 December, 2008

Correspondence to Katherine M. Coyne, HIV/GUM Directorate, St Stephen's Centre, Chelsea & Westminster NHS Foundation Trust, 369 Fulham Road, London SW10 9NH, UK. Tel: +44 20 8746 8000; fax: +44 20 8846 6188; e-mail:

Back to Top | Article Outline


An estimated 33.2 million people were living with HIV worldwide in 2007 [1]. In 2006 there were an estimated 700 000 cases of tuberculosis (TB) among HIV-positive people, and 200 000 deaths [2]. HIV increases the risk of active TB and the mortality rate.

When HIV-positive individuals are diagnosed with TB they usually have significant immunosuppression and require HIV therapy as well as TB treatment. First-line TB drugs are isoniazid, rifampicin, pyrazinamide and ethambutol. Co-administration with antiretrovirals often results in drug–drug interactions which may compromise efficacy or result in toxicity. Adverse effects are very common and include gastrointestinal intolerance, rashes and hepatotoxicity, which may necessitate stopping drugs and reintroducing them gradually. There are overlapping toxicity profiles such as peripheral neuropathy with isoniazid and didanosine, and anaemia with isoniazid, rifampicin, pyrazinamide and zidovudine. Whenever possible the most efficacious TB regimen should be continued, with adjustment to the antiretroviral drugs if necessary. However, sometimes it is unsafe to continue the first-line TB drugs and alternatives need to be found.

Second-line TB drugs also need to be employed when Mycobacterium tuberculosis is resistant to first-line drugs. Multidrug-resistant TB (MDR-TB) is defined as a TB which is resistant to both rifampicin and isoniazid and is an emerging epidemic, with 489 000 cases annually, representing nearly 5% of the global TB burden [2]. Treatment of MDR-TB is usually continued for 24 months rather than 6–12 months for drug-sensitive TB, and second-line agents are usually much more expensive. This adds up to the cost of drugs for MDR-TB exceeding drug costs for sensitive TB by 100-fold to 300-fold [3,4]. The management of MDR-TB is especially challenging in resource-poor settings in which there is limited availability of resistance testing, expertise in complex treatment decisions, second-line agents and facilities for monitoring safety, efficacy and plasma concentrations of drugs.

Adequate management of MDR-TB is crucial not only for recovery of the individual but also to prevent the acquisition of more resistance mutations, and the spread of drug-resistant strains between individuals. Extensively drug-resistant TB (XDR-TB) is defined as MDR-TB also resistant to a fluoroquinolone and at least one second-line injectable agent (amikacin, kanamycin and/or capreomycin), and is a small but growing problem. XDR-TB has a high mortality rate, especially among those infected with HIV [5].

Second-line TB drugs include some of the oldest antimicrobial agents developed, such as para-aminosalicylic acid (PAS), ethionamide and thiacetazone. When older drugs were developed, current knowledge of metabolic pathways and modern methods for screening for interactions were not available. Other classes of drugs were derived from antimicrobials produced by certain strains of actinomycetes, and include the aminoglycosides (amikacin and kanamycin), polypeptides (capreomycin, viomycin, enviomycin) and cycloserine. Some broad-spectrum antibiotics have activity against M. tuberculosis, including fluoroquinolones. There is less evidence for macrolides, amoxicillin with clavulanic acid and the newer synthetic antibacterial linezolid. The use of many of these agents has been limited by intolerability, toxicity and uncertain efficacy (see Table 1). On the positive side, this means that resistance to these drugs is uncommon.

Table 1

Table 1

The review will summarize the limited knowledge about pharmacology of second-line anti-TB agents, the small number of drug interactions studies which have been performed and the difficulty in predicting drug interactions. Implications for clinical practice will be discussed (see Table 2).

Table 2

Table 2

Table 2

Table 2

Back to Top | Article Outline

Interaction potential of antiretroviral drugs

There are five classes of drugs currently licensed for the treatment of HIV. These are:

  1. Nucleoside reverse transcriptase inhibitors (NRTIs; abacavir, didanosine, emtricitabine, lamivudine, zidovudine) and a nucleotide reverse transcriptase inhibitor (tenofovir)
  2. Nonnucleoside reverse transcriptase inhibitors (NNRTIs; efavirenz, etravirine, nevirapine)
  3. Protease inhibitors (PIs; atazanavir, darunavir, fosamprenavir, indinavir, lopinavir, nelfinavir, ritonavir, saquinavir, tipranavir)
  4. Entry inhibitors (the fusion inhibitor enfurvirtide, and the CCR5 antagonist maraviroc)
  5. Integrase inhibitors (raltegravir)

In clinical practice, three or more drugs from at least two classes are usually employed concurrently to suppress viral replication.

Pharmacokinetic interactions with antiretroviral drugs may occur during absorption, distribution, metabolism and elimination. Gastric pH, activity of transmembrane transporters such as enterocyte P-glycoprotein (P-gp) and drug metabolism by enterocyte cytochrome P450 (CYP450) enzymes may all influence the bioavailability of antiretroviral drugs. Once absorbed, antiretroviral drugs bind variably to plasma proteins (principally albumin and alpha-1 acid glycoprotein) and the extent of binding may be altered by co-administered drugs.

Hepatic CYP450 enzymes are responsible for the metabolism of a wide variety of drugs. NNRTIs and PIs are extensively metabolized by CYP450 and clinically significant drug interactions with these drugs are common [6]. As well as being substrates for CYP450, efavirenz and nevirapine induce these enzymes (i.e. CYP2B6 and 3A4) and may increase clearance of other drugs. Ritonavir is a potent inhibitor of CYP3A4 used routinely to increase plasma concentrations of other PIs. Ritonavir-boosted PIs may therefore lead to higher concentrations of other drugs metabolized by CYP3A4 or other isoforms inhibited by ritonavir (i.e. CYP2D6). In contrast, NRTIs are not metabolized by CYP450. They undergo intracellular phosphorylation to the active drug and have a low potential for significant drug–drug interactions [7].

Evidence is emerging of the importance of phase II reactions in drug interactions involving antiretrovirals. Ritonavir, for example, is an inducer of glucuronidation and has been clearly shown to be responsible for significant decreases in concentrations of drugs that are eliminated via this reaction, including etravirine [8] and lamotrigine [9]. Furthermore, the effect of ritonavir on glucuronidation is dose-dependent.

Distribution of drugs is also mediated by plasma membrane influx and efflux transporters including P-gp, multidrug resistance protein (MRP), organic anion transport protein (OATP1 and OATP2). These transporters are present on many cell types including intestinal epithelium, hepatocytes and kidney. They control absorption, intracellular distribution, metabolism and elimination of drug through biliary canals and active tubular secretion.

Back to Top | Article Outline


Aminoglycosides including streptomycin, kanamycin and amikacin are bacteriocidal, show considerable in-vitro activity against M. tuberculosis [10] and are clinically efficacious. The potential for pharmacokinetic drug interactions is low, although additive toxicities may occur. Aminoglycosides achieve high concentrations in bone, pleural, ascetic, synovial and peritoneal fluid and show poor cerebrospinal fluid (CSF) penetration (except amikacin in children with meningitis [11]). They are excreted unchanged in urine by glomerular filtration. All cause renal dysfunction, although streptomycin has been reported to be less nephrotoxic than other aminoglycosides [12]. Whenever possible co-administration with other drugs associated with renal dysfunction should be avoided, including the NRTI, tenofovir. Aminoglycosides are ototoxic [13] and eighth cranial nerve function should be tested at baseline and during treatment.

Back to Top | Article Outline


Streptomycin was the first effective anti-TB agent. It is given intramuscularly at a usual dose of 1–2 g daily. Therapeutic drug monitoring should be employed to maintain C max of 15–40 μg/ml and C trough of less than 5 μg/ml (C trough lower than 1 μg/ml in adults over 50 years old or with renal impairment). The half-life of streptomycin is 2–3 h and it is 30% protein bound.

Back to Top | Article Outline


Kanamycin is given intravenously or intramuscularly at a dose of 15 mg/kg per min in two or three divided doses, with a dose reduction in renal impairment. It is also excreted rapidly by glomerular filtration with a serum half-life of about 4 h. Therapeutic drug monitoring may be useful, aiming for C max of 15–30 μg/ml and C trough of less than 10 μg/ml.

Back to Top | Article Outline


Amikacin is a semi-synthetic aminoglycoside derived from kanamycin and administered parenterally at a dose of 15 mg/kg per day in two or three divided doses, reduced in renal impairment. Therapeutic drug monitoring of both C max and C trough is recommended. Amikacin is excreted unchanged in urine with plasma half-life of about 2 h. Plasma protein binding is estimated at 0–11%.

Back to Top | Article Outline


Capreomycin, viomycin and enviomycin

Capreomycin, viomycin and enviomycin are not absorbed from the gastrointestinal tract and are given intramuscularly. CSF penetration is poor. They are mostly excreted unchanged by the kidney. About 50% of a dose of capreomycin is excreted by glomerular filtration within 12 h. Like aminoglycosides, the major problems are ototoxicity and nephrotoxicity. Renal function should be closely monitored, especially when co-administered with tenofovir.

Back to Top | Article Outline


Fluoroquinolones are broad-spectrum antibiotics and are tuberculocidal. Their use in TB treatment has been the subject of a recent Cochrane review [14]. Their mechanism of action is inhibition of bacterial topoisomerase IV and DNA gyrase enzymes required for bacterial DNA replication, transcription, repair and recombination. Oral bioavailability is high: ciprofloxacin 70% [15], ofloxacin 98% [16], levofloxacin 99% [17], gatifloxacin 96%, sparfloxacin 90% and moxifloxacin 90%. There is no substantial loss from first-pass metabolism. The absorption of all fluoroquinolones is reduced by buffered drugs including older formulations of didanosine, and they should be taken 2 h before or 6 h after any buffered drugs. They can also be administered intravenously. Binding to plasma proteins is moderate 20–50% [17,18], and not high enough to expect significant interactions with protein binding of other drugs. Levofloxacin, gatifloxacin and moxifloxacin may cause prolongation of the QT interval and should be used with caution with other agents that do the same, including PIs [19], efavirenz [20], clarithromycin, erythromycin and some antidepressants. If these drugs are used together electrocardiographic monitoring is strongly recommended.

Back to Top | Article Outline


Ciprofloxacin has very little anti-TB activity and a more active fluoroquinolone should be used if available.

Back to Top | Article Outline

Ofloxacin, levofloxacin and gatifloxcin

Levofloxacin is the optical S-(−) isomer of ofloxacin and is twice as potent in vitro [21]. Ofloxacin is widely distributed in body fluids, and penetrates CSF better than levofloxacin. Ofloxacin, levofloxacin and gatifloxacin are primarily excreted unchanged in urine [17] by active tubular secretion. Dose reduction is required in renal failure and caution should be exercised when co-prescribing drugs associated with renal dysfunction including tenofovir. Only 4–8% of the dose is recovered from the faeces. Unlike ciprofloxacin, they do not seem to have any significant effect on CYP450 activity or phase II reactions [16,22].

Back to Top | Article Outline


Sparfloxacin exhibits excellent tissue penetration with concentrations in most extracellular fluids at least as high as in plasma, although lower in CSF [18]. Sparfloxacin does not appear to interact with theophylline, caffeine, warfarin or cimetidine [18], suggesting that its metabolism does not involve CYP2E1, 1A2, 2C19 or 3A4. It is metabolized in the liver by glucuronidation with an elimination half-life of about 20 h [18]. Ritonavir induces glucuronidation and reduces the concentration of drugs metabolized by glucuronidation [9]. It would therefore be expected that ritonavir reduces concentrations of sparfloxacin. Atazanavir inhibits glucuronidation and hypothetically causes increased sparfloxacin concentrations, but this effect may be counteracted by the common co-administration of ritonavir with atazanavir. Sparfloxacin is excreted in equal amounts in faeces and urine as unchanged drug and as the glucuronide metabolite.

Back to Top | Article Outline


Unlike other fluoroquinolones, 52% of an oral or intravenous dose of moxifloxacin is metabolized via sulphate conjugation and glucuronidation. The sulphate conjugate accounts for 38% of the dose, and is primarily excreted in faeces, whereas 14% is converted to a glucuronide conjugate which appears in urine [23]. The remaining drug is excreted unchanged in urine (20%) and faeces (25%). Doses do not need to be modified in renal failure. Moxifloxacin does not alter CYP3A4, 2D6, 2C9, 2C19 and 1A2 activity and should not alter antiretroviral concentrations.

Concomitant administration of rifampicin caused a 27% decrease in moxifloxacin area under the curve (AUC) and a marked increase in the concentration of the inactive sulphate metabolite, probably by the induction of sulphate conjugation [24]. The clinical significance of this is unclear and antiretrovirals are unlikely to alter sulphate conjugation. Polymorphisms of the MDR1 gene, which encodes P-gp, are associated with delayed absorption of moxifloxacin, although total moxifloxacin drug exposure was unchanged [24]. This effect may also occur with ritonavir, which is an inhibitor of P-gp expression [25]. Neither the magnitude nor the clinical importance of this interaction has been investigated, and further pharmacodynamic studies are warranted.

Back to Top | Article Outline


Ethionamide and protionamide

Ethionamide is administered orally with a bioavailability of over 90% [26,27]. It is started at 250 mg daily. The dose is titrated up to 15–20 mg/kg per day (maximum 1 g daily) if gastric irritation permits. It is usually given once daily but divided doses can lessen gastrointestinal symptoms. It reaches high concentrations throughout the body including in CSF, and 10–30% is protein bound [28].

Ethionamide is metabolized in the liver to seven metabolites, some of which are biologically active. Metabolism occurs by sulphoxidation, desulphuration and deamination, followed by methylation [26,27]. The involvement of CYP450 enzymes means that drug interactions are likely. Less than 1% is excreted unchanged by the kidneys.

Therapeutic drug monitoring is recommended in hepatic impairment. A high incidence of hepatotoxicity has been reported when ethionamide has been used with other hepatotoxic drugs such as rifampicin [29,30]. Caution should therefore be exercised if co-prescribing antiretrovirals which have been strongly associated with abnormal liver function, including the NNRTIs efavirenz and nevirapine [31] and the PIs darunavir [32] and tipranavir [33].

Ethionamide can cause depression, anxiety and psychosis. It may therefore not be advisable to start efavirenz at the same time, since these side effects are commonly experienced in the first few weeks of efavirenz therapy [34]. A psychotic reaction has been reported with the combination of ethionamide and excess alcohol [35]. Patients should therefore be counselled to avoid excess alcohol.

Protionamide is closely related to ethionamide structurally, with almost identical mechanism of action and pharmacokinetics.

Back to Top | Article Outline


Cycloserine has a narrow therapeutic window and its use is limited by frequent neuropsychiatric side effects. Plasma concentrations should be monitored in those taking more than 500 mg daily, those with renal dysfunction and if toxicity occurs. Concentrations should be below 30 μg/ml. Alcohol increases the risk of convulsions so caution should be exercised with Norvir (ritonavir) or Kaletra (lopinavir/ritonavir) liquid.

Cycloserine is readily and almost completely absorbed from the gastrointestinal tract. Plasma protein binding is less than 20%. It is widely distributed throughout body fluids including CSF [36]. It is excreted largely unchanged by glomerular filtration with a plasma half-life of about 10 h [37]. Fifty percent of a dose appears in urine within 12 h and 70% by 72 h. The remainder is presumed to be metabolized but the pathways are unknown. Neuropsychiatric side effects include anxiety, confusion, depression, psychosis, suicidal ideation, aggression and paranoia. It is probably advisable not to start efavirenz concurrently.

Back to Top | Article Outline

Para-aminosalicylic acid

Para-aminosalicylic acid was introduced into clinical use in 1948 and was the second antibiotic found to be effective against tuberculosis, after streptomycin. It is bacteriostatic [38,39]. Its use has been limited worldwide, so most TB isolates remain susceptible. PAS does not penetrate CSF unless the meninges are inflamed, in which case CSF concentrations reach 10–50% of plasma concentrations [40]. It competitively blocks absorption of vitamin B12 and can induce a malabsorption syndrome [41].

Para-aminosalicylic acid is rapidly acetylated in the liver and then excreted by glomerular filtration [42]. It is 80% excreted into urine, with 50% excreted in the inactive acetylated form. Approximately 50–70% of PAS is protein bound and the plasma half-life is 45–60 min [41]. No interactions with antiretroviral agents have been described.

Back to Top | Article Outline


Thiacetazone is structurally related to sulphonamide antibiotics. It was shown to be bacteriostatic against M. tuberculosis in 1946 [43]. It is well absorbed orally and 95% protein bound. Its plasma half-life is 8–12 h but its metabolism is not understood. Twenty percent is excreted unchanged in urine.

Thiacetazone has been widely used in resource-poor settings, formulated in a fixed combination tablet with isoniazid. However, concerns about low potency and toxicity have limited its use. There have been reports of increased rates of serious skin reactions including Stevens–Johnson syndrome in HIV-positive individuals, some fatal, so its administration in HIV infection is not recommended [44,45].

Back to Top | Article Outline


Clofazimine is a substituted iminophenazine dye, originally developed as an anti-TB agent, but primarily used as an antileprosy drug. Its mechanisms of action are largely unknown but it binds preferentially to mycobacterial DNA and inhibits RNA transcription [46]. It also has anti-inflammatory properties. After repeated oral dosing it has a very long half-life of up to 70 days. It is highly lipophilic and tends to be deposited in fatty tissue and cells of the reticuloendothelial system. It does not penetrate CSF well [37]. It is largely eliminated unchanged in faeces, both as unabsorbed drug and via biliary excretion. A small amount is excreted in urine as unchanged drug and metabolites [47]. Clofazimine is a weak inhibitor of CYP3A4. Clofazimine may delay absorption of rifampicin and prolong the time to C max [48]. Interactions with dapsone, oestrogen and vitamin A have also been reported [47]. Interactions with antiretrovirals are difficult to predict but the potential exists for clofazime to increase plasma concentrations of drugs metabolized by CYP3A4 including protease inhibitors and etravirine.

Back to Top | Article Outline

Amoxicillin with clavulanic acid

β-Lactam antibiotics have had a very limited role in the treatment of TB because mycobacteria produce β-lactamase. Amoxicillin with clavulanic acid has been used at high doses in multidrug regimens against TB with some success [49]. Amoxicillin is commonly used in individuals with HIV infection and interactions with antiretrovirals are unlikely.

Back to Top | Article Outline



Clarithromycin is a semi-synthetic macrolide antibiotic used to treat M. avium as well as second-line against M. tuberculosis. When administered orally it undergoes first-pass metabolism so that the bioavailability is 55%. It can also be given intravenously.

Clarithromycin is metabolized to the active 14-hydroxyclarithromycin and both reach high concentrations in tissues, in part because of intracellular uptake [50]. Plasma protein binding is 80%. Clarithromycin is extensively metabolized in the liver, when it is both a substrate for and an inhibitor of CYP3A. Metabolites are mostly excreted in urine with a smaller amount in faeces via bile.

Interactions between clarithromycin and several antiretrovirals have been studied. Simultaneous co-administration of clarithromycin reduces zidovudine concentrations, but no reduction is seen if clarithromycin and zidovudine are given at least 2 h apart [51–53]. Care should be exercised if there are other factors reducing zidovudine concentrations such as malabsorption or interactions with other drugs.

Clarithromycin and ritonavir are both inhibitors of CYP3A enzymes. Clarithromycin had little impact on plasma concentrations of ritonavir. However, co-administration of ritonavir with clarithromycin resulted in complete inhibition of the formation of the 14-hydroxy metabolite of clarithromycin and an increase in clarithromycin AUC of 77% [54]. This is unlikely to be clinically significant in patients with normal renal function, but in those with renal impairment taking ritonavir, the dose of clarithromycin should be reduced by 50% with creatinine clearance 30–60 ml/min and by 75% with creatinine clearance less than 30 ml/min, and the daily dose should not exceed 1 g.

Simultaneous administration of clarithromycin and didanosine resulted in no statistically significant change in didanosine pharmacokinetics [55].

Efavirenz induces CYP3A4, reduces clarithromycin concentrations [56] and increases the concentration of the 14-hydroxy metabolite, associated with an increased frequency of skin rashes. Nevirapine also reduces clarithromycin concentrations [56]. Azithromycin should therefore be considered instead of clarithromycin in patients taking efavirenz or nevirapine.

Back to Top | Article Outline


Azithromycin is derived from erythromycin. It is given orally with rapid absorption and 40% bioavailability [57], or intravenously. It is widely distributed into tissues in which concentrations reach up to 100-fold higher than plasma concentrations, and fall more slowly [57]. Its long elimination half-life of 69 h [58] allows once daily, or even once weekly, dosing. It penetrates CSF poorly except when meninges are inflamed.

The metabolism of azithromycin is not completely understood but is via hepatic pathways other than CYP450 enzymes [59]. It is excreted mainly as unchanged drug in bile. It is a less potent inhibitor of CYP3A than clarithromycin and interactions are less likely.

Co-administration of azithromycin and efavirenz does not significantly affect the concentration of either drug [60].

Nelfinavir significantly increased the C max and AUC of azithromycin by over 100%, possibly via inhibition of P-gp [61]. Although the combination was tolerated, caution should be exercised when using these drugs together. In this study azithromycin caused a small decrease in nelfinavir concentrations, which is unlikely to be clinically significant. Ritonavir is a potent inhibitor of P-gp and may also increase azithromycin levels significantly, although this drug combination has not been studied.

Azithromycin had no significant impact on the C max and AUC of zidovudine, although it significantly increased the intracellular exposure to phosphorylated zidovudine by 110% [62]. The same study showed that azithromycin had no significant effect on didanosine pharmacokinetics, whereas azithromycin concentrations were not measured [62]. Azithromycin may therefore be safely co-administered with both zidovudine and didanosine.

Back to Top | Article Outline


Linezolid is a new synthetic antibacterial agent of the oxazolidinone class, which has recently been used in successful regimens against MDR-TB [63]. However, long-term toxicities are concerning. Linezolid can cause reversible myelosuppression [64,65] and should be used cautiously in patients with preexisting cytopaenias, including those with anaemia on zidovudine. Prolonged courses of linezolid have been associated with lactic acidosis and optic or peripheral neuropathy [66] and appear to inhibit mitochondrial protein synthesis [66,67]. It would therefore be advisable not to co-administer other drugs, which suppress mitochondrial activity such as didanosine, stavudine, and to a lesser extent zidovudine.

Linezolid is given orally with 100% bioavailability [68], or intravenously. Protein binding is 31% and there is good penetration into tissues including CSF [69]. It is primarily metabolized by oxidation to two inactive metabolites. Fifty percent of administered drug is excreted in urine as metabolites, and 35% appears in urine as unchanged drug [70]. Linezolid is not an inducer of CYP450 in rats, and in-vitro studies have shown that linezolid is not detectably metabolized by human CYP450 and it does not inhibit the activities of clinically significant human CYP isoforms (1A2, 2C9, 2C19, 2D6, 2E1, 3A4). Linezolid is a reversible inhibitor of monoamine oxidase A and B and should not be co-administered with selective serotonin reuptake inhibitors or tricyclic antidepressants. Pharmacokinetic drug–drug interactions with antiretrovirals are not anticipated.

Back to Top | Article Outline

Drugs in development

New agents are being developed with novel mechanisms of action, which are active against drug-resistant TB. They may transform the prognosis of MDR-TB, and priority should be given to studying their interactions with antiretrovirals in order to develop compatible regimens.

TMC207 is a novel diarylquinolone, which targets mycobacterial ATP synthase. It has shown in-vivo activity in preliminary clinical trials [71]. It is metabolized by CYP3A4 to an active N-monodesmethyl metabolite. Pharmacokinetic studies of TMC207 and ketoconazole (an inhibitor of CYP3A4) showed an increase in TMC207 AUC of 22%. It would therefore be predicted that TMC207 concentrations increase with the inhibition of CYP3A4 by ritonavir and decrease with enzyme induction by efavirenz or nevirapine.

Nitroimidazoles include PA-824 and OPC-67683. Both have good in-vitro activity against drug-resistant TB, no CYP450 interactions have been described and results of phase II trials are awaited.

SQ109 is a promising diamine drug currently in phase I clinical trials. It is metabolized by CYP2D6 and CYP2C19 so interactions with ritonavir are likely.

Nitrofuranylamides are a newly discovered class of drugs with in-vitro activity against drug-susceptible and drug-resistant strains of M. tuberculosis [72].

Back to Top | Article Outline


Second-line TB drugs include several different agents characterized by diverse metabolic pathways. Some of these are among the oldest antimicrobials introduced into clinical practice, others have been recently approved. Most may interact with one or more antiretroviral classes. Pharmacokinetic or pharmacodynamic interactions may cause toxicity and alter efficacy. Knowledge of the pharmacology of second-line TB drugs is fundamental to managing HIV-infected patients who are intolerant of first-line TB drugs or are infected by MDR-TB or XDR-TB.

MDR-TB and XDR-TB are emerging epidemics and their management is challenging. More needs to be done in resource-poor settings, in which the burden of HIV/TB co-infection is greatest. The burden of MDR-TB and XDR-TB can be reduced by better implementation of Directly Observed Therapy programmes to rapidly identify, trace and re-instate treatment in poorly adherent patients receiving first-line TB drugs. The World Health Organization Green Light Committee Initiative's effort to support national TB programmes and improve access to second-line TB treatment in developing countries is welcome. High-quality techniques need to be available to identify drug-resistant organisms. Regional surveillance programmes must collect resistance data and implement local guidelines for second-line treatment on the basis of drugs, which are likely to work. These second-line regimes should then be matched to antiretroviral combinations, which are available locally and have minimal interactions.

Back to Top | Article Outline


The paper was jointly written and revised by all four authors. K.C. drafted and edited the paper and incorporated the responses of the reviewers. A.P. reviewed and edited the paper and provided expert input into the management of HIV/TB co-infection and use of second-line anti-TB drugs. M.L. reviewed the paper and co-wrote the section on ‘Interaction Potential of Antiretroviral drugs’. M.B. co-wrote the section on ‘Interaction Potential of Antiretroviral drugs’, provided expert input in pharmacology and edited the final version.

Back to Top | Article Outline


1. UNAIDS. AIDS Epidemic Update 2007. Geneva: World Health Organisation; 2007.
2. World Health Organisation. Global Tuberculosis Control 2008. Surveillance, planning, financing. Geneva: World Health Organisation; 2008.
3. White VL, Moore-Gillon J. Resource implications of patients with multidrug resistant tuberculosis. Thorax 2000; 55:962–963.
4. Rajbhandary SS, Marks SM, Bock NN. Costs of patients hospitalized for multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2004; 8:1012–1016.
5. Ghandi NR, Moll A, Sturm AW, Pawinski R, Govender T, Lalloo U, et al. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet 2006; 368:1554–1556.
6. de Maat MM, Ekhart GC, Huitema AD, Koks CH, Mulder JW, Beijnen JH. Drug interactions between antiretroviral drugs and comedicated agents. Clin Pharmacokinet 2003; 42:223–282.
7. Barry M, Mulcahy F, Merry C, Gibbons S, Back D. Pharmacokinetics and potential interactions amongst antiretroviral agents used to treat patients with HIV infection. Clin Pharmacokinet 1999; 36:289–304.
8. Kakuda TN, Scholler-Gyure M, Peeters M, Vingerhoets J, Corbett C, Woodfall BJ, et al. Pharmacokinetics and pharmacodynamics of TMC125 in HIV-infected patients with nonnucleoside reverse transcriptase inhibitor and protease inhibitor resistance: TMC125-C223. 14th Conference on Retrovirus and Opportunistic Infections; Los Angeles, 2007.
9. van der Lee MJ, Dawood L, ter Hofstede HJ, de Graaf-Teulen MJ, van Ewijk-Beneken Kolmer EW, Caliskan-Yassen N, et al. Lopinavir/ritonavir reduces lamotrigine plasma concentrations in healthy subjects. Clin Pharmacol Ther 2006; 80:159–168.
10. Ho YI, Chan CY, Cheng AF. In-vitro activities of aminoglycoside-aminocyclitols against mycobacteria. J Antimicrob Chemother 1997; 40:27–32.
11. Gaillard J-L, Silly C, Le Masne A, Mahut B, Lacaille F, Cheron G, et al. Cerebrospinal fluid penetration of amikacin in children with community-acquired bacterial meningitis. Antimicrob Agents Chemother 1995; 39:253–255.
12. McCracken GH. Aminoglycoside toxicity in infants and children. Am J Med 1986; 80(Suppl 6B):172–178.
13. Brummett RE, Fox KE. Aminoglycoside-induced hearing loss in humans. Antimicrob Agents Chemother 1989; 33:797–800.
14. Ziganshina LE, Vizel AA, Squire SB. Fluoroquinolones for treating tuberculosis. In: The Cochrane Database of Systematic Reviews; Issue 3. Chichester: John Wiley; 2005.
15. Vance-Bryan K, Guay DR, Rotschafer JC. Clinical pharmacokinetics of ciprofloxacin. Clin Pharmacokinet 1990; 19:434–461.
16. Lamp KC, Bailey EM, Rybak MJ. Ofloxacin clinical pharmacokinetics. Clin Pharmacokinet 1992; 22:32–46.
17. Fish DN, Chow AT. The clinical pharmacokinetics of levofloxacin. Clin Pharmacokinet 1997; 32:101–119.
18. Shimada J, Nogita T, Ishibashi Y. Clinical pharmacokinetics of sparfloxacin. Clin Pharmacokinet 1993; 25:358–369.
19. Anson BD, Weaver JGR, Ackerman MJ, Akinsete O, Henry K, January CT, et al. Blockade of HERG channels by HIV protease inhibitors. Lancet 2005; 365:682–686.
20. Castillo R, Pedalino RP, El-Sherif N, Turitto G. Efavirenz-associated QT prolongation and torsade de pointes arrhythmia. Ann Pharmacother 2002; 36:1006–1008.
21. Davis R, Bryson HM. Levofloxacin: a review of its antibacterial activity, pharmacokinetics and therapeutic efficacy. Drugs 1994; 4:677–700.
22. Aminimanizani A, Beringer P, Jelliffe R. Comparative pharmacokinetics and pharmacodynamics of the newer fluoroquinolone antibacterials. Clin Pharmacokinet 2001; 40:169–187.
23. Stass H, Kubitza D. Pharmacokinetics and elimination of moxifloxacin after oral and intravenous administration in man. J Antimicrob Chemother 1999; 43(Suppl B):83–90.
24. Weiner M, Burman W, Luo CC, Peloquin CA, Engle M, Goldberg S, et al. Effects of rifampicin and multidrug resistance gene polymorphism on concentrations of moxifloxacin. Antimicrob Agents Chemother 2007; 51:2861–2866.
25. Foisy MM, Yakiwchuk EM, Hughes CA. Induction effects of ritonavir: implications for drug interactions. Ann Pharmacother 2008; 42:1048–1059.
26. Jenner PJ, Smith SE. Plasma levels of ethionamide and prothionamide in a volunteer following intravenous and oral dosages. Lepr Rev 1987; 58:31–37.
27. Jenner PJ, Ellard GA, Gruer PJK, Aber VR. A comparison of the blood levels and urinary excretion of ethionamide and prothionamide in man. J Antimicrob Chemother 1984; 13:267–277.
28. Berning SE, Peloquin CA. Antimicrobial agents: ethionamide. In: Yu VL, Merigan TC, Barriere S, White NJ, editors. Antimicrobial chemotherapy and vaccines. Baltimore, MD: Williams and Wilkins; 1998. pp. 650–654.
29. Pattyn SR, Janssens L, Bourland J, Saylan T, Davies EM, Grillone S, et al. Hepatotoxicity of the combination of rifampin-ethionamide in the treatment of multibacillary leprosy. Int J Lepr 1984; 52:1–6.
30. Cartel J-L, Millan J, Guelpa-Lauras CC, Grosset JH. Hepatitis in leprosy patients treated by a daily combination of dapsone, rifampin, and a thioamide. Int J Lepr 1983; 51:461–465.
31. Brück S, Witte S, Brust J, Schuster D, Mosthaf F, Procaccianti M, et al. Hepatotoxicity in patients prescribed efavirenz or nevirapine. Eur J Med Res 2008; 13:343–348.
32. Vispo E. Warning on hepatotoxicity of darunavir. AIDS Rev 2008; 10:63.
33. Dong BJ, Cocohoba JM. Tipranavir: a protease inhibitor for HIV salvage therapy. Ann Pharmacother 2006; 40:1311–1321.
34. Arendt G, de Nocker D, von Giesen HJ, Nolting T. Neuropsychiatry side effects of efavirenz therapy. Expert Opin Drug Saf 2007; 6:147–154.
35. Lansdown FS, Beran M, Litwak T. Psychotoxic reaction during ethionamide therapy. Am Rev Respir Dis 1967; 95:1053–1055.
36. Berning SE, Peloquin CA. Antimicrobial agents: cycloserine. In: Yu VL, Merigan TC, Barriere S, White NJ, editors. Antimicrobial chemotherapy and vaccines. Baltimore, MD: Williams and Wilkins; 1998. pp. 638–642.
37. Peloquin CA. Antituberculosis drugs: pharmacokinetics. In: Heifets LB, editor. Drug susceptibility in the chemotherapy of mycobacterial infections. Boca Raton, FL: CRC Press; 1991. pp. 89–122.
38. Verbist L. Mode of action of antituberculous drugs (Part I). Medicon Int 1974; 3:11–23.
39. Verbist L. Mode of action of antituberculous drugs (Part II). Medicon Int 1974; 3:3–17.
40. Spector R, Lorenzo WV. The active transport of para-aminosalicyclic acid from the cerebrospinal fluid. J Pharmacol Exp Ther 1973; 185:642–648.
41. Berning SE, Peloquin CA. Antimicrobial agents: para-aminosalicylic acid. In: Yu VL, Merigan TC, Barriere S, White NJ, editors. Antimicrobial chemotherapy and vaccines. Baltimore, MD: Williams and Wilkins; 1998. pp. 663–668.
42. Way EL, Smith PK, Howie DL. The absorption, distribution, excretion and fate of para-aminosalicylic acid. J Pharmacol Exp Ther 1948; 93:368–382.
43. Domagk G, Behnish R, Mietasch F, Schmidt H. Uber eine neue gegen tuberkelbazillon in vitro wirksame berbindungs klasse. Naturwiss 1946; 33:315.
44. Chintu C, Luo C, Bhat G, Raviglione M, DuPont H, Zumla A. Cutaneous hypersensitivity reactions due to thiacetazone in the treatment of tuberculosis in Zambian children infected with HIV-1. Arch Dis Child 1993; 68:665–668.
45. Nunn P, Kibuga D, Gathua S, Brindle R, Imalingat A, Wasunna K. Cutaneous hypersensitivity reactions due to thiacetazone in HIV-1 seropositive patients treated for tuberculosis. Lancet 1991; 337:627–630.
46. Arbiser JL, Moschella SL. Clofazimine: a review of its medical uses and mechanisms of action. J Am Acad Dermatol 1995; 32:241–247.
47. Holdiness MR. Clinical pharmacokinetics of clofazimine: a review. Clin Pharmacokinet 1989; 16:74–85.
48. Mehta J, Gandhi IS, Sane SB, Wamburkar MN. Effect of clofazimine and dapsone on rifampicin (Lositril) pharmacokinetics in multibacillary and paucibacillary leprosy cases. Lepr Rev 1986; 57(Suppl 3):67–76.
49. Nadler JP, Berger J, Nord JA, Cofsky R, Saxena M. Amoxicillin-clavulanic acid for treating drug-resistanct Mycobacterium tuberculosis. Chest 1991; 99:1025–1026.
50. Rodvold KA. Clinical pharmacokinetics of clarithromycin. Clin Pharmacokinet 1999; 37:385–398.
51. Petty B, Polis M, Haneiwich S, Dellerson M, Craft JC, Chaisson R. Pharmacokinetic assessment of clarithromycin plus zidovudine in HIV patients [Abstract 24]. In: Programe and abstracts of the 32nd Interscience Conference on Antimicrobial Agents and Chemotherapy; 1192 Washington DC: American Society of Microbiology. p. 114.
52. Vance E, Watson-Bitar M, Gustavson L, Kazanjian P. Pharmacokinetics of clarithromycin and zidovudine in patients with AIDS. Antimicrob Agents Chemother 1995; 39:1355–1360.
53. Polis MA, Piscitelli SC, Vogel S, Witebsky FG, Conville PS, Petty B, et al. Clarithromycin lowers plasma zidovudine levels in persons with human immunodeficiency virus infection. Antimicrob Agents Chemother 1997; 41:1709–1714.
54. Ouellet D, Hsu A, Granneman GR, Carlson G, Cavanaugh J, Guenther H, et al. Pharmacokinetic interaction between ritonavir and clarithromycin. Clin Pharmacol Ther 1998; 64:355–362.
55. Gillum JG, Bruzzese VL, Israel DS, Kaplowitz LG, Polk RE. Effect of clarithromycin on the pharmacokinetics of 2′,3′-dideoxyinosine in patients who are seropositive for human immunodeficiency virus. Clin Infect Dis 1996; 22:716–718.
56. Kuper JI, D'Aprile M. Drug–drug interactions of clinical significance in the treatment of patients with Mycobacterium avium complex disease. Clin Pharmacokinet 2000; 39:203–214.
57. Lalak NJ, Morris DL. Azithromycin clinical pharmacokinetics. Clin Pharmacokinet 1993; 25:370–374.
58. Luke DR, Foulds G, Cohen SF, Levy B. Safety, toleration, and pharmacokinetics of intravenous azithromycin. Antimicrob Agents Chemother 1996; 40:2577–2581.
59. Rapp RP. Pharmacokinetics and pharmacodynamics of intravenous and oral azithromycin: enhanced tissue activity and minimal drug interactions. Ann Pharmacother 1998; 32:785–793.
60. Benedek IH, Joshi A, Fiske WD, White SJ, Jobes JL, Joseph JL, et al. Pharmacokinetic (PK) interaction studies in healthy volunteers with efavirenz (EFV) and the macrolide antibiotics, azithromycin (AZM) and clarithromycin (CLR). Program Abstr 5th Conf Retrovir Oppor Infect Conf Retrovir Oppor Infect 5th 1998 Chic Ill. 1998 Feb 1–5; 5th: 144 [abstract no. 347].
61. Amsden GW, Nafziger AN, Foulds G, Cabelus LJ. A study of the pharmacokinetics of azithromycin and nelfinavir when coadministered in healthy volunteers. J Clin Pharmacol 2000; 40:1522–1527.
62. Amsden G, Flaherty J, Luke D. Lack of an effect of azithromycin on the disposition of zidovudine and dideoxyinosine in HIV-infected patients. J Clin Pharmacol 2001; 41:210–216.
63. Fortun J, Martin-Davila P, Navas E, Perez-Elias MJ, Cobo J, Tato M, et al. Linezolid for the treatment of multidrug resistant tuberculosis. J Antimicrob Chemother 2005; 56:180–185.
64. Green SL, Maddox JC, Huttenbach ED. Linezolid and reversible myelosuppression. JAMA 2001; 285:1291.
65. Bishop E, Melvani S, Howden BP, Charles PG, Grayson ML. Good clinical outcomes but high rates of adverse reactions during linezolid therapy for serious infections: a proposed protocol for monitoring therapy in complex patients. Antimicrob Agents Chemother 2006; 50:1599–1602.
66. De Vriese AS, Coster RV, Smet J, Seneca S, Lovering A, Van Haute LL. Linezolid-induced inhibition of mitochondrial protein synthesis. Clin Infect Dis 2006; 42:1111–1117.
67. Garrabou G, Soriano A, López S, Guallar JP, Giralt M, Villarroya F. Reversible inhibition of mitochondrial protein synthesis during linezolid-related hyperlactatemia. Antimicrob Agents Chemother 2007; 51:962–967.
68. Rubinstein E, Isturiz R, Standiford HC, Smith LG, Oliphant TH, Cammarata S, et al. Worldwide assessment of linezolid's clinical safety and tolerability: comparator-controlled phase III studies. Antimicrob Agents Chemother 2003; 47:1824–1831.
69. MacGowan AP. Pharmacokinetic and pharmacodynamic profile of linezolid in healthy volunteers and patients with Gram-positive infections. J Antimicrob Chemother 2003; 51(Suppl S2):ii17–ii25.
70. Stalker DJ, Jungbluth GL. Clinical pharmacokinetics of linezolid, a novel oxazolidinone antibacterial. Clin Pharmacokinet 2003; 42:1129–1140.
71. Rustomjee R, Diacon AH, Allen J, Venter A, Reddy C, Patientia RF, et al. Early bacterial activity and pharmacokinetics of the Diarylquinolone TMC 207 in pulmonary tuberculosis. Antimicrob Agents Chemother 2008; 52:2831–2835.
72. Hurdle JG, Lee RB, Budha NR. A microbiological assessment of novel nitrofuranylamides as antituberculous agents. J Antimicrob Chemother 2008; 62:1037–1045.

antiretroviral agents; antitubercular agents; drug interactions; multidrug-resistant tuberculosis; Mycobacterium tuberculosis

© 2009 Lippincott Williams & Wilkins, Inc.