XDR-TB is much more difficult to treat than other MDR-TB and extremely difficult to treat in HIV-infected patients33,34. While reports of HIV-infected patients being promptly diagnosed with XDR-TB and placed on adequate regimen are non-existent to date, a few reports of cohorts of HIV-negative patients have been shown to have cure rates that exceed 50%33,35.
1 - Consider a longer duration of use for the injectable agent (12 months or possibly the whole treatment). If the patient’s strain is resistant to all injectable agents, use one the patient has never used before1 .
2 - Use a third-generation fluoroquinolone such as moxifloxacin. The potential benefit of moxifloxacin should be weighed against the increased risk of QT prolongation when combined with bedaquiline.
3 - Use all Group 4 agents that have not been used extensively in a previous regimen or any that are likely to be effective.
4 - Use two or more agents from Group 5. Add bedaquiline. Consider high-dose H if lowlevel resistance is documented or no katG mutation is detected.
5 - Use any likely effective Group 1 drugs.
6 - Consider adjuvant surgery if there is localized disease.
7 - Consider compassionate use of new agents (Appendix 11).
Extension of therapy to 24 months is the suggested minimum length of treatment for XDR-TB.
Box 10.2 - A case of XDR-TB and example regimen
A patient is receiving Km-Ofx-Eto-Cs-Z and remains smear-positive and culture-positive after 8 months of treatment. In addition the patient is not improving clinically. The DST performed on a sputum collected 2 months ago reveals resistance to H, R, Z, E, S, Km, Cm and Ofx. This patient has XDR-TB.
The regimen should be designed based on the principles described in Section 10.1. Bdq should be considered. A higher generation FQ may have some effect.
The recommended regimen to be considered in this patient would be:
While the reproducibility and reliability of DST to injectables is good, there is little data on clinical relevance of the test. Options with XDR-TB are very limited and some strains may be affected in vivo by an injectable agent even though they are testing resistant in vitro.