Reacción5.1 Drug-susceptible TB
5.1.1 Guidelines update 2010 to 2022
Refer to WHO consolidated guidelines on tuberculosis: module 4: treatment: drug-susceptible tuberculosis treatment: web annexes. Web Annex 4
https://iris.who.int/handle/10665/353398
5.2 Drug-resistant TB
5.2.1 Guidelines update 2011 to 2022
Refer to WHO consolidated guidelines on tuberculosis: module 4: treatment: drug-resistant tuberculosis treatment: web annexes, 2022 update. Web Annexes 3 and 4 https://iris.who.int/handle/10665/365284
5.2.2 Guideline update 2025
PICO 1
Question: Should a 6-month regimen using bedaquiline, delamanid, and linezolid with or without the addition of levofloxacin or clofazimine or both (BDLLfx/C) vs. the currently recommended standard of care, i.e. either the 9-month regimen (for patients without fluoroquinolone resistance) or an individualized long regimen (for patients with fluoroquinolone resistance) be used in patients with pulmonary RR-TB (with or without fluoroquinolone resistance)?
CI: confidence interval; RR: risk ratio
Explanations
a. In the BEAT TB trial, those in the intervention group who had established fluoroquinolone (FQ) resistance (22%) did not receive levofloxacin and those with established FQ sensitivity (60%) did not receive clofazimine. In the control group, those who had established FQ resistance (22%) received an individualized regimen with bedaquiline, linezolid, delamanid and other drugs and those with established FQ sensitivity (60%) continued the standard of care (South African RR TB regimen). Those in whom FQ sensitivity was not established continued with the BEAT TB regimen in the intervention group (17%) or continued standard of care in the control group (16%). A subgroup analysis according to FQ sensitivity and resistance showed a difference in the risk difference between groups, but with overlapping 95% confidence intervals and the test for interaction was not statistically significant. We did not downgrade for indirectness due to uncertainty in the subgroup effect.
b. Sustained treatment success: The 95% CI for the absolute effect crosses three thresholds, from moderate harm to large benefit. We, therefore, downgraded the certainty by three levels due to extremely serious imprecision.
c. For the absolute effect we used the risk difference and 95% CI calculated from the trial data.
d. Failure and recurrence: The 95% CI for the absolute effect crosses three thresholds, from small benefit to moderate harm. We, therefore, downgraded the certainty by three levels due to extremely serious imprecision.
e. Death: The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We, therefore, downgraded the certainty by three levels due to extremely serious imprecision.
f. Lost to follow up/Amplification (acquisition) of drug-resistance: The 95% CI for the absolute effect crosses one threshold, from small benefit to trivial harm. We, therefore, downgraded the certainty by one level due to serious imprecision.
g. Adverse events: The 95% CI for the absolute effect crosses four thresholds, from large benefit to small harm. We, therefore, downgraded the certainty by three levels due to extremely serious imprecision.
PICO 1
PICO 2.1
Question: A 9-month regimen using bedaquiline, linezolid, moxifloxacin, and pyrazinamide (BLMZ) compared to currently recommended longer WHO regimens in patients with pulmonary RR-TB (without fluoroquinolone resistance)
CI: confidence interval; RR: risk ratio
Explanations
a. Standard of care for the treatment of rifampicin-resistant and fluoroquinolone-susceptible tuberculosis was composed according to latest World Health Organization guidelines, as they evolved during the trial. In this group 82% of participants were treated with the longer (18–24 month) all-oral regimen. After Dec 18, 2018 the first change in WHO Guidelines (ending use of injectables) was applied to the trial control regimen. A sensitivity analysis did not show important differences, with overlap in 95% confidence intervals, when excluding the 18% randomized during the period when injectables were prescribed. We therefore did not downgrade for indirectness.
b. Sustained treatment success: The 95% CI for the absolute effect crosses two thresholds, from small benefit to large benefit. We therefore downgraded the certainty by two levels due to serious imprecision.
c. For the absolute effect we used the risk difference and 95% CI calculated from the trial data.
d. Failure and recurrence: The 95% CI for the absolute effect crosses two thresholds, from small benefit to small harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
e. Death: The 95% CI for the absolute effect crosses three thresholds, from small benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
f. Lost to follow-up: The 95% CI for the absolute effect crosses two thresholds, from small benefit to large benefit. We therefore downgraded the certainty by two levels due to very serious imprecision.
g. Adverse events (Grade 3–5): The 95% CI for the absolute effect crosses three thresholds, from large benefit to trivial harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
h. Adverse events (serious adverse events): The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
i. Risk ratios could not be calculated because there was no acquired resistance in the control arm. The risk difference was 0% (95% CI, -2.9% to 2.9%).
PICO 2.1
PICO 2.2
Question: A 9-month regimen using bedaquiline, clofazimine, linezolid, levofloxacin, and pyrazinamide (BLLfxCZ) compared to currently recommended longer WHO regimens in patients with pulmonary RR-TB (without fluoroquinolone resistance)
CI: confidence interval; RR: risk ratio
Explanations
a. Standard of care for the treatment of rifampicin-resistant and fluoroquinolone-susceptible tuberculosis was composed according to latest World Health Organization guidelines, as they evolved during the trial. In this group 82% of participants were treated with the longer (18–24 month) all-oral regimen. After Dec 18, 2018 the first change in WHO Guidelines (ending use of injectables) was applied to the trial control regimen. A sensitivity analysis did not show important differences, with overlap in 95% confidence intervals, when excluding the 18% randomized during the period when injectables were prescribed. We therefore did not downgrade for indirectness.
b. Sustained treatment success, Lost to follow-up: The 95% CI for the absolute effect crosses two thresholds, from small benefit to large benefit. We therefore downgraded the certainty by two levels due to very serious imprecision.
c. For the absolute effect we used the risk difference and 95% CI calculated from the trial data.
d. Failure and recurrence: The 95% CI for the absolute effect crosses three thresholds, from trivial benefit to large harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
e. Death: The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
f. Adverse events (Grade 3–5): The 95% CI for the absolute effect crosses three thresholds, from large benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
g. Adverse events (serious adverse events): The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
h. Risk ratios could not be calculated because there was no acquired resistance in the control arm. The risk difference was 1.6% (95% CI, -1.3% to 5.6%).
i. Amplification (acquisition) of drug-resistance: The 95% CI for the absolute effect crosses one threshold, from trivial benefit to small harm. We therefore downgraded the certainty by one level due to serious imprecision.
PICO 2.2
Assessment
Summary of judgements
Type of recommendation
Conclusions
PICO 2.3
Question: A 9-month regimen using bedaquiline, delamanid, linezolid, levofloxacin, and pyrazinamide (BDLLfxZ) compared to currently recommended longer WHO regimens in patients with pulmonary RR-TB (without fluoroquinolone resistance)
CI: confidence interval; RR: risk ratio
Explanations
a. Standard of care for the treatment of rifampicin-resistant and fluoroquinolone-susceptible tuberculosis was composed according to latest World Health Organization guidelines, as they evolved during the trial. In this group 82% of participants were treated with the longer (18–24 month) all-oral regimen. After Dec 18, 2018 the first change in WHO Guidelines (ending use of injectables) was applied to the trial control regimen. A sensitivity analysis did not show important differences, with overlap in 95% confidence intervals, when excluding the 18% randomized during the period when injectables were prescribed. We therefore did not downgrade for indirectness.
b. Sustained treatment success: The 95% CI for the absolute effect crosses three thresholds, from small harm to large benefit. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
c. For the absolute effect we used the risk difference and 95% CI calculated from the trial data.
d. Failure and recurrence: The 95% CI for the absolute effect crosses two thresholds, from trivial benefit to moderate harm. We therefore downgraded the certainty by two levels due to very serious imprecision.
e. Death: The 95% CI for the absolute effect crosses three thresholds, from large harm to small benefit. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
f. Lost to follow-up: The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to small harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
g. Adverse events (Grade 3–5): The 95% CI for the absolute effect crosses three thresholds, from large benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
h. Adverse events (serious adverse events): The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
i. Risk ratios could not be calculated because there was no acquired resistance in the control arm. The risk difference was 0% (95% CI, -2.9% to 2.9%).
PICO 2.3
Assessment
Summary of judgements
Type of recommendation
Conclusions
PICO 2.4
Question: A 9-month regimen using delamanid, clofazimine, linezolid, levofloxacin, and pyrazinamide (DCLLfxZ) compared to currently recommended longer WHO regimens in patients with pulmonary RR-TB (without fluoroquinolone resistance)
CI: confidence interval; RR: risk ratio
Explanations
a. Standard of care for the treatment of rifampicin-resistant and fluoroquinolone-susceptible tuberculosis was composed according to latest World Health Organization guidelines, as they evolved during the trial. In this group 82% of participants were treated with the longer (18–24 month) all-oral regimen. After Dec 18, 2018 the first change in WHO Guidelines (ending use of injectables) was applied to the trial control regimen. A sensitivity analysis did not show important differences, with overlap in 95% confidence intervals, when excluding the 18% randomized during the period when injectables were prescribed. We therefore did not downgrade for indirectness.
b. Sustained treatment success: The 95% CI for the absolute effect crosses three thresholds, from large harm to large benefit. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
c. For the absolute effect we used the risk difference and 95% CI calculated from the trial data.
d. Failure and recurrence: The 95% CI for the absolute effect crosses three thresholds, from trivial harm to large harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
e. Death: The 95% CI for the absolute effect crosses three thresholds, from small benefit to large harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
f. Lost to follow-up: The 95% CI for the absolute effect crosses two thresholds, from moderate benefit to small harm. We therefore downgraded the certainty by two levels due to very serious imprecision.
g. Adverse events (Grade 3–5): The 95% CI for the absolute effect crosses three thresholds, from large benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
h. Adverse events (serious adverse events): The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
i. Amplification (acquisition) of drug-resistance: The 95% CI for the absolute effect crosses two thresholds, from trivial harm to moderate harm. We therefore downgraded the certainty by two levels due to very serious imprecision.
j. Risk ratios could not be calculated because there was no acquired resistance in the control arm. The risk difference was 4.0% (95% CI, 0.9% to 8.7%).
PICO 2.4
Assessment
Summary of judgements
Type of recommendation
Conclusions
PICO 2.5
Question: A 9-month regimen using delamanid, clofazimine, moxifloxacin, and pyrazinamide (DCMZ) compared to currently recommended longer WHO regimens in patients with pulmonary RR-TB (without fluoroquinolone resistance)
CI: confidence interval; RR: risk ratio
Explanations
a. Standard of care for the treatment of rifampicin-resistant and fluoroquinolone-susceptible tuberculosis was composed according to latest World Health Organization guidelines, as they evolved during the trial. In this group 82% of participants were treated with the longer (18–24 month) all-oral regimen. After Dec 18, 2018 the first change in WHO Guidelines (ending use of injectables) was applied to the trial control regimen. A sensitivity analysis did not show important differences, with overlap in 95% confidence intervals, when excluding the 18% randomized during the period when injectables were prescribed. We therefore did not downgrade for indirectness.
b. Sustained treatment success: The 95% CI for the absolute effect crosses three thresholds, from moderate harm to large benefit. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
c. For the absolute effect we used the risk difference and 95% CI calculated from the trial data.
d. Failure and recurrence: The 95% CI for the absolute effect crosses three thresholds, from trivial harm to large harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
e. Death: The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
f. Lost to follow-up: The 95% CI for the absolute effect crosses two thresholds, from large benefit to small benefit. We therefore downgraded the certainty by two levels due to very serious imprecision.
g. Adverse events (Grade 3–5): The 95% CI for the absolute effect crosses three thresholds, from large benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
h. Adverse events (serious adverse events): The 95% CI for the absolute effect crosses three thresholds, from moderate benefit to moderate harm. We therefore downgraded the certainty by three levels due to extremely serious imprecision.
i. Amplification (acquisition) of drug-resistance: The 95% CI for the absolute effect crosses one threshold, from small harm to moderate harm. We therefore downgraded the certainty by one level due to serious imprecision.
j. Risk ratios could not be calculated because there was no acquired resistance in the control arm. The risk difference was 6.7% (95% CI, 3.2% to 11.9%).
PICO 2.5
Assessment
Summary of judgements
Type of recommendation
Conclusions
PICO 8
Question: Should HCV treatment be co-administered with MDR-TB treatment in patients co-infected by MDR/RR-TB and HCV?
Background: We designed an online survey for collection of data about choice of treatment strategies when treating patients with MDR-TB and HCV co-infection and their health outcomes. Data collected from investigators (i.e. cohorts) who treated at least one patient with MDR-TB and HCV treatment co-administration (i.e., intervention arm) and at least one patient with MDR-TB treatment with delay of HCV treatment (i.e., comparator arm) were included in a comparative analysis, in which the risk ratio was calculated to assess treatment effects for the outcomes of interest.
Summary of findings:
CI: confidence interval; RR: risk ratio
Explanations
a. Data collected from investigators (i.e. cohorts) who treated at least one patient with MDR-TB and HCV treatment co-administration (i.e., intervention arm) and at least one patient with MDR-TB treatment with delay of HCV treatment (i.e., comparator arm) were included in the comparative analysis.
b. The outcome data is based on records or recall by experts from an expert-evidence survey, for cohorts of patients receiving either co-administration of MDR-TB and HCV treatments, or delayed HCV treatment in their healthcare setting. The comparative analysis to calculate relative effects based on these cohorts does not control for any potential confounding. We downgraded the certainty of evidence due to very serious risk of bias.
c. The 95% CI includes both small but important benefit as well as large benefit.
d. Additionally, 1/97 patients receiving co-administration of treatments and 3/230 patients receiving delayed HCV treatment experienced TB recurrence.
e. The 95% CI includes both a large benefit and large harm. We therefore downgraded the certainty of evidence due to imprecision.
f. The 95% CI includes both large benefit and harm. We therefore downgraded the certainty of evidence due to imprecision.
g. There was insufficient data provided by the experts to allow a calculation of relative effects, and information about the time frame of amplification of drug resistance and occurrence of drug-drug interactions was not available.
h. The pooled proportions of the outcomes of interest were calculated in the groups of patients receiving MDR-TB and HCV treatment co-administration and, separately, in the groups receiving MDR-TB treatment with delay of HCV treatment. The single arm proportions were pooled by fitting binomial generalized linear mixed models with a logit link. Inferences should not be made by comparing the pooled proportions in the two different groups as they are a mix of comparative and non-comparative scenarios derived from non-randomized data, and are subject to bias and serious confounding. This analysis was performed for descriptive purposes.
i. The pooled proportions in the two groups are derived from the single cohorts of patients receiving co-administration of MDR-TB and HCV treatments or MDR-TB treatment only with delayed HCV treatment.
j. The 95% CIs around the pooled proportions are wide.
PICO 8
Assessment
Summary of judgements
Type of recommendation
Conclusions
PICO 2.1–2.3 (Summary EtD with full notes on Conclusions)
Assessment
Summary of judgements
Type of recommendation
Conclusions
PICO 2.4–2.5 (Summary EtD with full notes on Conclusions)
Assessment
Summary of judgements
Type of recommendation
Conclusions
Comparison between BLMZ, BLLfxCZ and BDLLfxZ to support prioritization
Question 1: Should a 9-month regimen using bedaquiline, linezolid, moxifloxacin, and pyrazinamide (9BLMZ) vs. currently recommended longer WHO regimens be used in patients with pulmonary RR-TB (without fluoroquinolone resistance)?
Question 2: Should a 9-month regimen using bedaquiline, clofazimine, linezolid, levofloxacin, and pyrazinamide (9BLLfxCZ) vs. currently recommended longer WHO regimens be used in patients with pulmonary RR-TB (without fluoroquinolone resistance)?
Question 3: Should a 9-month regimen using bedaquiline, delamanid, linezolid, levofloxacin, and pyrazinamide (9BDLLfxZ) vs. currently recommended longer WHO regimens be used in patients with pulmonary RR-TB (without fluoroquinolone resistance)?
Question 4: Should a 9-month regimen using delamanid, clofazimine, linezolid, levofloxacin, and pyrazinamide (9DCLLfxZ) vs. currently recommended longer WHO regimens be used in patients with pulmonary RR-TB (without fluoroquinolone resistance)?
Question 5: Should a 9-month regimen using delamanid, clofazimine, moxifloxacin, and pyrazinamide (9DCMZ) vs. currently recommended longer WHO regimens be used in patients with pulmonary RR-TB (without fluoroquinolone resistance)?
Summary of judgements
Review
5.3 TB care and support
Refer to WHO consolidated guidelines on tuberculosis: module 4: treatment: tuberculosis care and support: web annexes.
Web Annexes 1 and 2 https://iris.who.int/handle/10665/352904