FeedbackRecommendation 1.
New patients with pulmonary TB should receive a regimen containing 6 months of rifampicin: 2HRZE/4HR (strong recommendation, high certainty of evidence)
Remarks
A: The recommendation also applies to extrapulmonary TB - except TB of the central nervous system, bone or joint for which some expert groups suggest longer therapy.
B: WHO recommends that national TB control programmes provide supervision and support for all TB patients in order to ensure completion of the full course of therapy.
C: WHO recommends drug resistance surveys (or surveillance) for monitoring the impact of the treatment programme, as well as for designing standard regimens.
Source of recommendation
This recommendation was first put forward in 2010 and was considered valid in 2017 guidelines update (see mapping of recommendations in Annex). The recommendation is copied without modification into this consolidated document and appears exactly as in the 2010 guidelines.
Justification and evidence
A systematic review and meta-analysis included 21 472 participants in 312 arms of 57 randomized controlled trials conducted in various regions of the world since 1965 [9]. In three of the 57 trials, patients were randomly assigned to either a 2-month rifampicin or a 6-month rifampicin arm; rates of failure, relapse and acquired drug resistance were compared "head-to-head" across the two study arms. In a multivariate regression analysis, each arm of the 57 trials was treated as a separate cohort and results were adjusted for potentially confounding patient and treatment factors.
The three studies with head-to-head comparisons showed that the risk of relapse after a 6-month rifampicin regimen was significantly lower than that after a 2-month rifampicin regimen. If a country were to change from a 2-month to a 6-month rifampicin regimen, the benefit would be an estimated 112 relapses averted per 1000 TB patients.
Regression analysis suggests that changing to a 6-month regimen would significantly reduce failure and acquired drug resistance rates, in addition to relapse rates.⁷ This analysis found that regimens with 5-7 months of rifampicin have 0.43 times the failure rate, and 0.32 times the relapse rate of regimens with 1-2 months of rifampicin. Among the failures and relapses from regimens with 5-7 months of rifampicin, the rate of acquired drug resistance is 0.28 times that of the regimens with 1-2 months of rifampicin.
Patients with isoniazid resistance would realize major benefits if the 2-month rifampicin regimen were replaced with a 6-month regimen. Among patients with isoniazid mono-resistance at the start of treatment, 38% relapsed after treatment with 2-month rifampicin regimens, which is significantly higher than the 5.5% relapse rate after treatment with 6-month rifampicin regimens. Thus, changing to the 6-month rifampicin regimen would avert 325 relapses per 1000 patients who start treatment with isoniazid resistance.
Even for patients with pan-susceptible TB, the proportion who relapsed after the 2-month rifampicin regimen was 8.2%, which was significantly higher than the 3.1% for the 6-month rifampicin regimen.
When the first course of therapy is considered along with retreatment for patients who fail or relapse, it is estimated that the 6-month rifampicin regimen would avert between 3 and 12 deaths per 1000 compared with the 2-month rifampicin regimen across 7 countries modelled with a range of drug resistance among new patients. In addition, 0.6-4.4 failures and relapses with drug resistance other than MDR-TB would be averted per 1000 TB patients, but an additional 0.6-1.3 MDR-TB cases would be generated.
Among patients who failed or relapsed after their first course of treatment containing 6 months of rifampicin, regression analysis found a reduction in overall acquired drug resistance; however, the pattern of acquired drug resistance was different from that in patients who received the 2-month rifampicin regimen. The risk of acquiring drug resistance other than MDR-TB is higher with the 2-month rifampicin regimen, but the risk of acquiring MDR-TB is higher with the 6-month rifampicin regimen. Among failures, the proportion with MDR-TB is predicted to be 4-56% after initial treatment with the 2-month rifampicin regimen but 50-94% after initial treatment with the regimen containing 6 months of rifampicin.
Subgroup considerations
The interactions of rifampicin with antiretroviral therapy (ART) are of concern. Switching to the 6-month rifampicin regimen means that these drug interactions must be taken into account for the full 6 months rather than for just the first 2 months of therapy. However, the 6-month rifampicin regimen has marked benefits for persons living with HIV, and the drug interactions can be managed [10].
Implementation considerations
To help minimize the acquisition of MDR-TB, it is critically important that national TB control programmes ensure adequate supervision of rifampicin. Implementing patient supervision for the 4-month continuation phase will require additional resources in areas where the continuation phase has been self-administered - an investment that may be offset by the savings from relapses (and therefore retreatments) averted. In 2008, 23 countries (including four that are considered high-burden) still used the 2-month rifampicin regimen for their new patients. These countries reported 706 905 new cases in 2007, or 13% of the global new TB notifications that year.
Monitoring and evaluation
This recommendation places high value on saving lives. Given both the high certainty of evidence for this benefit and the fact that the potential harm of acquired DR-TB can be mitigated by supervision of treatment. Periodic drug resistance surveys (or ongoing surveillance) in each country are essential for monitoring the impact of the regimen and the overall treatment programme.
Recommendation 2.
Wherever feasible, the optimal dosing frequency for new patients with pulmonary TB is daily throughout the course of therapy (strong recommendation, high certainty of evidence)
Source of recommendation
This recommendation was first put forward in 2010 and considered valid in the 2017 guidelines update (see mapping of recommendations in Annex). The recommendation is copied without modification into this consolidated document and appears exactly as in the 2010 guidelines.
Justification and evidence
A systematic review and meta-analysis included 21 472 participants in 312 arms of 57 randomized controlled trials conducted in various regions of the world since 1965 [9]. In a multivariate regression analysis, each arm of the 57 randomized controlled trials was treated as a separate cohort, and results were adjusted for potentially confounding patient and treatment factors. Only one study of 223 patients evaluated a rifampicin-containing regimen administered twice weekly throughout therapy; this study was not included in the meta-analyses.
No significant increase in failure, relapse or acquired drug resistance was found when daily dosing throughout therapy was compared with the following intermittent regimens in new TB patients overall, namely: daily then thrice weekly; daily then twice weekly; or thrice weekly throughout therapy.
However, the regression analysis showed that patients being treated thrice weekly throughout therapy had rates of acquired drug resistance that were 3.3 times higher than those in patients who received daily drug administration throughout treatment.
The meta-analysis revealed no difference in rates of failure, relapse or acquired drug resistance in pan-susceptible new patients being treated with these dosing schedules. However, the use of a three times weekly intensive phase schedule in patients with pre-treatment isoniazid resistance was associated in another meta-analysis with a significantly higher risk of failure and acquired drug resistance [11].
Implementation considerations
When based in a health facility, daily administration of therapy places a larger burden on TB programmes and patients than does intermittent therapy. Intermittent regimens require stronger programmes with higher-quality patient supervision, but all regimens should be provided with full patient supervision and support.
Studies of patients' preferences for dosing schedules were not systematically reviewed. The higher isoniazid dose used in intermittent therapy was not considered to have an increased incidence of adverse effects. The rifampicin dosage was unchanged when using intermittent therapy.
In an international, multicentre, randomized trial (Union Study A), Jindani, Nunn & Enarson found thrice weekly dosing resulted in significantly lower culture conversion rates at 2 months [12]. In developing recommendations, this endpoint was ranked by the GDG as important but not critical for decision-making and was not part of the systematic review.
For new patients without HIV infection, high- certainty of evidence demonstrated no significant difference between regimens that were administered daily throughout treatment, daily initially and then intermittently in the continuation phase, or thrice weekly throughout treatment.
Daily dosing is optimal because it probably achieves better adherence under programme conditions. While the definition of the term varies across countries, "daily" is considered to mean at least five times per week. In addition, meta-analyses showed the superiority of daily (compared with thrice weekly) intensive-phase dosing for patients with pre-treatment isoniazid resistance and for preventing acquired drug resistance in patients overall.
Recommendation 3.
In all patients with drug-susceptible pulmonary TB, the use of thrice-weekly dosing is not recommended in both the intensive and continuation phases of therapy and daily dosing remains the recommended dosing frequency (conditional recommendation, very low certainty of evidence).
Source of recommendation
This recommendation was first put forward in 2010 and then updated in the 2017 guidelines (see mapping of recommendations in Annex). It is copied without modification into this consolidated document and appears exactly as in the 2010 guidelines.
Justification and evidence
The use of intermittent dosing of TB medications has been adopted in some geographical settings in an effort to improve treatment adherence and to reduce the burden on the health-care system due to daily treatment support. However, it was unclear how this intermittent dosing might affect treatment outcomes. In addition to the evidence from a systematic review conducted in 2009 of treatment regimens with intermittent dosing schedules [9], this systematic review was updated with the most recent randomized controlled trials [13-18].
Evidence showed that when thrice-weekly dosing throughout therapy was compared to daily dosing throughout therapy, patients who received thrice-weekly dosing had a higher risk of treatment failure, disease relapse and acquired drug resistance in both drug-susceptible disease and when the strain susceptibility was unknown. Consequently, thrice-weekly dosing in the intensive phase should never be used.
Likewise, when thrice-weekly dosing during the continuation phase only is compared to daily dosing throughout, there were higher rates of treatment failure and relapse in the patients that received thrice-weekly treatment during the continuation phase. In this case, acquired drug resistance rates did not differ. If thrice-weekly dosing during the continuation phase is used, it is essential to make sure that patients do not miss any dose of the medications and that treatment support is used.
In this review, the use of twice-weekly dosing in the continuation phase only was also reviewed. Twice-weekly dosing in the continuation phase only had higher rates of treatment failure, disease relapse and drug resistance than thrice-weekly dosing in the continuation phase only. As a result, twice-weekly dosing should never be used during any part of TB therapy.
Adherence to treatment was not adequately addressed in the reviewed studies to be included as an outcome. However, in most studies included in the systematic review, intermittent dosing used treatment support, while the use of treatment support during daily dosing was variable.
The GDG also considered that health equity would be adversely affected with intermittent dosing because more vulnerable populations would receive inferior treatment if intermittent dosing were used. This is because people living in more resource-constrained settings would be at greater risk of missing doses of medication, not only because of their difficulty in reaching a clinic but also because of the risk of medication stock-outs in clinics. Additionally, patients who are co-infected with HIV or have other comorbidities may not absorb TB medications well and therefore they may receive less medication than they are ingesting. In order for TB medication to be used as part of a treatment regimen, no doses may be missed with thrice-weekly intermittent dosing during the continuation phase because the rates of unfavourable outcomes may rise. Consequently, populations that are more vulnerable are at risk of missing doses of medication or of not absorbing the doses well, and intermittent dosing puts them in a situation where there is an increased risk of unfavourable outcomes.
Intermittent dosing may also create problems at national and international levels by resulting in requirements for different drug manufacturing and packaging and a reduced drug supply buffer, leading to an increased risk of TB medication stock-outs.
Given the findings in this review, all countries are encouraged to use daily dosing exclusively in both the intensive and the continuation phases of treatment. Although two separate evidence assessments were conducted on thrice-weekly dosing in the intensive phase and the continuation phase, both the formulated recommendations were conditional and there was very low certainty in the evidence. A combined recommendation for both intensive and continuation phases was formulated to make it more convenient for use by the end-users.
Subgroup considerations
This recommendation is the same for HIV-negative people and for people living with HIV.
The data used in this review examined only patients with drug-susceptible pulmonary TB who had no extenuating circumstances - such as adverse reactions which might require modification of the dosing schedule.
Children were not considered specifically in this review. However, there is no biologically plausible reason why this recommendation should not also apply to children. It is recommended that all children receive daily dosing of TB medications during the intensive and continuation phases of therapy for the same reason as adults. See WHO's 2014 guideline Guidance for national tuberculosis programmes on the management of tuberculosis in children [19] for recommendations on the daily dosing of children with DS-TB.
Implementation considerations
There are no new implementation considerations as the recommended daily treatment is already widespread practice. However, intermittent dosing is still used in some countries. In such exceptional cases, implementation of the recommendation to use exclusively daily dosing in the intensive and continuation phases of TB therapy is likely to have implications for medication procurement, practitioner training, change of programme practice and patient support.
Monitoring and evaluation
There are no new monitoring and evaluation recommendations as the standard of care (daily dosing of medications during the intensive and continuation phases of therapy) is being recommended.
Recommendation 4.
The use of fixed-dose combination tablets is recommended over separate drug formulations in treatment of patients with drug-susceptible TB (conditional recommendation, low certainty of evidence)
Source of recommendation
This recommendation was first put forward in the 2017 guidelines update (see mapping of recommendations in Annex). It is copied without modification into this consolidated document and appears exactly as in the 2017 guidelines.
Justification and evidence
The evidence presented to the GDG was based on a systematic review of randomized controlled trials done by Albanna et al. [20] and by a recent Cochrane review [21]. This evidence showed that the fixed-dose combination (FDC) tablets are non-inferior and equally effective as separate drug formulations in terms of treatment failure, death, treatment adherence and adverse events. There was a small increase in 2-month culture conversion with FDC treatment; however, there was no difference in culture conversion rates by the end of treatment. Patient satisfaction was higher among persons treated with FDCs. A slightly higher rate of disease relapse and acquired drug resistance among patients treated with FDCs compared with those treated with separate drug formulations was not statistically significant.
Patient treatment satisfaction with FDCs was considered the most important factor for making decisions on this recommendation.
Studies in these reviews did not evaluate bioavailability of the drugs in the FDCs, but previous studies did not indicate that the FDC formulations used had significant bioavailability issues [20]. As no pharmacokinetic studies were done on these FDC formulations, the bioavailability of drugs within the FDCs versus the separate drug formulations remains an important consideration that indicates the need to procure FDCs of demonstrated bioavailability [22-24]. This area requires further research.
FDCs may provide programme benefits by making the ordering of medication easier, simplifying supply chain management, reducing the occurrence of stock-outs, and facilitating drug delivery and prescription preparation. FDCs may also provide benefits - especially in settings with a large number of TB patients and a limited number of health-care workers - by reducing the need for additional health-care staff and training in the dosing and dispensing of medications, as well as by contributing to a lower pill burden for patients. Nevertheless, national TB programmes are advised to have a quantity of separate drug formulations available for certain treatment conditions. Having single drug formulations available would be beneficial to national TB programmes when designing MDR-TB regimens that include some first-line drugs (i.e. pyrazinamide, EMB, high-dose isoniazid), when providing preventive therapy, and in cases of adverse reactions to TB medications when drugs must be reintroduced one at a time.
The GDG acknowledged that greater patient satisfaction is an advantage of FDCs over separate drug formulations.
Subgroup considerations
The reduced pill burden as a result of using FDCs may be especially valuable in patients with co-morbidities (notably HIV infection) and paediatric patients (who may have some difficulty in swallowing large amounts of medications).
Patients with some specific medical conditions (e.g. intolerance to certain TB drugs, liver or renal function impairment) are likely to require individual medication dose adjustment which can be done only with separate drug formulations.
Implementation considerations
There are no specific implementation considerations as the use of FDC formulations is already widespread.
Monitoring and evaluation
There are no specific new recommendations for monitoring and evaluation as the use of both types of drug formulation is already widespread.
Recommendation 5.
In new pulmonary TB patients treated with the regimen containing rifampicin throughout treatment, if a positive sputum smear is found at completion of the intensive phase, the extension of the intensive phase is not recommended (strong recommendation, high certainty of evidence)
Source of recommendation
This recommendation was first put forward in 2010 and considered valid in the 2017 guidelines update (see mapping of recommendations in Annex). It is copied without modification into this consolidated document and appears exactly as in the 2010 guidelines.
Justification and evidence
The systematic review identified only one relevant study (with results published in 2012). A study still under way (at moment of review) in Bangladesh of a 6-month rifampicin-containing regimen randomized 3775 new smear-positive patients who remained positive at 2 months to either the 1-month extension arm (extension of the intensive phase by 1 month) or the no-extension arm [25].
Preliminary results at 1 year of follow-up showed that patients in the 1-month extension arm had a significantly lower relapse rate (relative risk 0.37, 95% CI 0.21, 0.66) than patients in the no-extension arm. A smaller decrease in failure in the 1-month extension arm was not statistically significant. Given the preliminary nature of the results and the passive follow-up of patients, the evidence from the Bangladesh study was graded with moderate certainty.
In 1000 TB patients with a 7% risk of relapse, the Bangladesh study predicts that extending the treatment of 183 patients who are smear-positive at 2 months would avert 16 of the 70 expected relapses. However, to achieve this 23% reduction in relapses, 158 patients per 1000 would be incorrectly predicted to relapse; consequently their treatment would be extended unnecessarily.
While extending rifampicin beyond 6 months reduces the risk of relapse, there is insufficient evidence to determine which patients are most likely to benefit. Historically, when the new patient regimen included only 2 months of rifampicin, the extension of the intensive phase meant an extra month of supervised rifampicin. This extra month is less important now as the current recommended regimen is 6 months of supervised rifampicin. Given these considerations, together with preliminary results from one moderate-certainty study that showed only modest benefit, a conditional recommendation was made not to extend treatment on the basis of a positive smear at 2 months.
⁷ The difference in failure and acquired drug resistance was not statistically significant in these three randomized controlled trials.