Early MDR-TB detection and the prompt initiation of an effective treatment regimen are important factors for successful outcomes. The benefits of rapid DSTs for patient and public health include better prognosis, prevention of further drug resistance, and a reduction in the spread of drug-resistant strains.

The results of DSTs are essential for designing the appropriate treatment regimen. However, DSTs for drugs such as ethambutol, streptomycin, and those in group 4 and 5 do not have high reproducibility or reliability. Therefore, the DST results should be interpreted along with a detailed clinical history. If a patient has used a drug as part of a failing regimen for over a month, the strain should be considered "probably resistant," even if the results of the DST indicated susceptibility.

Fluoroquinolones and injectable drugs are two groups of second-line core drugs with potent bactericidal activity. Other second-line drugs are accompanying drugs that are responsible for protecting the core drugs against resistance. To build a MDR-TB treatment regimen, begin with two core drugs (a fluoroquinolone and an injectable drug) and then add oral second-line drugs in the following order: ethionamide or protionamide, cycloserine, and p-aminosalicylic acid (PAS)142829. If the preceding drugs are not sufficient to make a regimen, group 1 and then group 5 drugs could also be added.

The recent WHO guidelines have changed the priority of first-line drugs14. Although first-line drugs were previously given the highest priority for inclusion in treatment regimens (if the strain showed susceptibility to the drugs)3031, the meta-analysis have shown only a slight benefit or no benefit for successful outcomes15. Similarly, group 5 drugs showed no association with successful outcomes. Based on such clinical evidence, first-line drugs and group 5 drugs may be included in a regimen but should not to be considered as active drugs.

The optimal number of drugs required to cure MDR-TB is not known. The meta-analysis showed that the use of at least four active drugs in the intensive phase and three active drugs in the continuation phase was likely to be effective15. There was no evidence to support the use of more than four second-line drugs in patients with extensive disease. Increasing the number of second-line drugs is permissible if the effectiveness of some drugs is uncertain. As opposed to this recommendation, several retrospective cohort studies showed that an aggressive regimen of at least 5 likely effective drugs during the intensive phase was associated with a reduced risk of treatment failure, death or relapse 323334.

Based on clinical evidence from the meta-analysis, the intensive phase of treatment should consist of at least pyrazinamide, a fluoroquinolone, an injectable drug, protionamide (or ethionamide), and cycloserine (or PAS if cycloserine cannot be used).

There have been controversies concerning the optimal duration for injectable drug use and total treatment. The American Thoracic Society guidelines from 2003 recommended a maximum cumulative streptomycin dose of 120 g (due to its cumulative toxic effects) but did not mention any other injectable drugs35. The WHO had suggested the time following sputum conversion in the recommendation of treatment duration. In 1997, the WHO recommended a minimum intensive phase of 3 months or until after culture conversion36. The 2006 and 2008 WHO guidelines recommended an intensive phase of at least 6 months or at least 4 months after culture conversion3031. However, in the 2011 and 2014 guidelines, the time for the intensive phase was increased to a minimum of 8 months and bacteriological conversion was not considered1314. This recommendation was based on clinical evidence from the meta-analysis, which did not determine the optimal time following culture conversion15. Therefore, the injectable drugs are recommended to use for at least eight months, but the duration can be modified based upon severity of disease, prior therapy, drug resistance patterns, response to therapy, and timing of sputum conversion.

Currently, the WHO recommends a total treatment duration of 20 months for patients who had no previous MDR-TB treatment. Patients who have had previous treatment for MDR-TB may need longer treatment. The duration may be modified depending on bacteriological status and other indicators of treatment progress.

A non-randomized study conducted in Bangladesh reported very promising results with a 9-month regimen37, and a randomized controlled trial is ongoing to test a shorter regimen38. However, until sufficient evidence is available, a shorter treatment regimen is not recommended as a standard. In addition, treatment with new drugs, such as bedaquiline39 and delamanid40, may allow for shorter MDR-TB therapy; however, optimization studies of these drugs are needed.

One of the major changes in the revised WHO guidelines is how to include ethambutol and pyrazinamide in the MDR regimen. In the previous guidelines, ethambutol and pyraziamide were recommended as the first option in the drug selection (as part of four active drugs) if susceptibility was shown3031. However, in the meta-analysis, pyrazinamide showed only a slightly added benefit and ethambutol was associated with a marginal but statistically significant reduction in the likelihood of cure15. Therefore, based on clinical evidence, ethambutol and pyrazinamide may be included in MDR-TB treatment regimens but should not be considered active drugs.

Pyrazinamide has poor bactericidal activity, but it has potent sterilizing activity that contributes to a shorter treatment duration. Given its promising potential, pyrazinamide is routinely added to MDR-TB treatment regimens if susceptibility is documented or unknown29. Although the WHO has recently recommended routine use of pyrazinamide even if the strain shows resistance14, this recommendation should be verified by further studies. A recent retrospective study showed that the WHO-recommended regimen in which pyrazinamide was not likely effective was associated with higher mortality rates41.

Pyrazinamide can be used for the entire treatment or at least for the intensive phase. Pyrazinamide is generally used with companion rifampicin for the treatment of susceptible TB. Given that rifampicin also has a potent sterilizing action, pyrazinamide is recommended to be used during the first 2 months of treatment only. While rifampicin is no longer active in MDR-TB treatment, pyrazinamide may continue working after the first few months of treatment42. In addition, many MDR-TB patients have chronically inflamed lungs, which theoretically produce the acidic environment where pyrazinamide is active14.

In contrast, ethambutol is not routinely added to MDR-TB treatment regimens. Although ethambutol can be added in the regimen, it is never considered an active drug, even if the strain shows susceptibility14.

Among the injectable drugs, there is currently no strong evidence to indicate which drugs are superior in terms of efficacy or adverse effects. The meta-analysis did not show any injectable to be superior to any other15. Therefore, the choice of injectable drug depends on the availability and resistance in each country. In Korea, capreomycin is not readily available and amikacin is usually injected intravenously; thus, kanamycin is the recommended first choice of injectable drugs16.

In contrast, streptomycin is not recommended because of high rates of resistance in patients with MDR-TB. However, streptomycin should be considered in cases where the strain is resistant to all other second-line injectable drugs given that there is little cross-resistance between streptomycin and the other injectable drugs14.

Fluoroquinolones, especially later-generation fluoroquinolones, have been shown to be significantly associated with cure15. Consequently, fluoroquinolones should always be used in MDR-TB treatment. Moxifloxacin or levofloxacin is preferred while ciprofloxacin and ofloxacin are not recommended given their weaker efficacy. The dosage recommendation for levofloxacin is 750-1,000 mg/day. Although moxifloxacin has better in vitro activity against TB compared to levofloxacin, a recent randomized trial conducted in Korea revealed that moxifloxacin and levofloxacin have comparable efficacy in terms of 3-month culture conversion rates and adverse events43. The WHO recommends, despite lack of evidence, the use of moxifloxacin even if levofloxacin (or ofloxacin) resistance is documented because there is no complete cross-resistance between these fluoroquinolones14.

In the 2008 WHO guidelines31, there was no preference among group 4 drugs. However, in the meta-analysis, ethionamide demonstrated a stronger association with successful outcome than cycloserine or PAS15. Furthermore, cycloserine demonstrated a stronger association than PAS; therefore, ethionamide or protionamide is preferred followed by cycloserine and, subsequently, PAS.

There is high cross-resistance between isoniazid and prothionamide if the inhA mutation is present44. If the inhA mutation is present, which can be detected by a LPA, protionamide can still be included in an MDR-TB treatment regimen; however, it may not be the best second-line drug14.

Group 5 drugs did not show an association with successful outcomes in the meta-analysis15; however, linezolid and high-dose isoniazid could not be analyzed due to the small number of cases. Therefore, group 5 drugs may be used if drugs from groups 2-4 are not likely to be effective; however, group 5 drugs are not included among active drugs. A recent meta-analysis showed that only linezolid was independently associated with favorable outcomes in the treatment of XDR-TB or fluoroquinolone-resistant MDR-TB45. There is renewed interest in the efficacy of clofazimine. A randomized controlled trial in China showed that clofazimine accelerated sputum culture conversion and improved treatment success rates in the treatment of MDR-TB46.

There are few evidences regarding the drug selection sequence in group 5 drugs. Many experts agreed that linezolid is the first option and clofazimine and meropenem/clavulanic acid may be more effective than clarithromycin or amoxicillin/clavulanic acid28. Recently, bedaquiline and delamanid have been listed in group 5 as well14.