Efficacy and safety of daily drug regimens containing high-dose versus standard-dose rifampicin for treating pulmonary tuberculosis: Systematic review and meta-analysis Authors: Ashutosh N. Aggarwal, Ritesh Agarwal, Sahajal Dhooria, Perumal K. Bhavani, Kuruswamy T. Prasad, Inderpaul S. Sehgal, Valliappan Muthu Published: 2026 ABSTRACT Standard rifampicin doses used to treat pulmonary tuberculosis lie at the lower end of drug exposure-response curve. We evaluated if higher daily rifampicin doses result in greater efficacy or harm. We searched literature databases for randomized controlled trials comparing efficacy and/or safety of standard and high rifampicin doses (>=15 mg/kg/day) for treating pulmonary tuberculosis. Our co-primary efficacy endpoints were proportion of patients with positive mycobacterial culture at end of intensive phase and end of treatment. The primary safety endpoint was incidence of severe hepatotoxicity. Secondary efficacy endpoints were proportion of patients with culture positivity at 1, 3, and 4 months, death, treatment failure, recurrence after treatment, and loss to follow-up. Secondary safety endpoints were incidence of adverse event (any, serious, and drug-related serious), hepatotoxicity, and treatment interruption/discontinuation. We generated summary risk ratios (RR) using random effects models. We identified 5668 publications and selected 12 studies with 3230 participants. Nine studies had low risk of bias. Those on high-dose rifampicin had significantly lower culture positivity at end of intensive phase (summary RR 0.74, 95%CI 0.62-0.89), but not at end of treatment (summary RR 0.52, 95%CI 0.23-1.15), as well as significantly higher incidence of severe hepatotoxicity (summary RR 2.06, 95% CI 1.15-3.68) and treatment interruption/discontinuation (summary RR 1.89, 95%CI 1.09-3.28). Other secondary efficacy and safety outcomes were similar between the two treatment strategies. We conclude that higher daily rifampicin doses accelerate sputum conversion by the end of intensive phase of ATT, but severe hepatotoxicity remains a potential concern. INTRODUCTION Tuberculosis continues to be an important public health problem globally, with 10.8 million incident cases and 1.25 million deaths estimated worldwide in 2023.[1] Advances in pharmacotherapy have shortened the anti-tubercular treatment (ATT) duration to 6 months. However, the treatment success rates are still around 88% globally, with wide geographic variations, indicating a significant scope for improvement.[1] Rifampicin is a critical component of the current multi-drug short-course ATT because of its good bactericidal and sterilising activity, both intracellular and extracellular, and the ability to prevent recurrence after treatment. Unfortunately, the optimal rifampicin dose was never established rigorously, but largely chosen based on economic and safety considerations.[2] Indeed, the currently recommended daily rifampicin dose (10 mg/kg, range 8–12) is actually placed at the lower end of the drug exposure-response curve.[2] Subtherapeutic drug levels are not uncommon at these doses, which might lead to worse outcomes or acquisition of rifampicin resistance.[345] Hence, increasing rifampicin dose can further improve treatment outcomes, and potentially reduce the ATT duration.[6] Animal experiments indicate that greater rifampicin exposure accelerates disease clearance and reduces subsequent recurrence in murine tuberculosis models.[789] Other studies suggest that escalating rifampicin dose up to 35 mg/kg or even more could lead to an increase in peak drug concentration, enhancing its bactericidal activity.[10111213] However, increased risk of adverse effects, especially hepatotoxicity, remains a potential concern at such high doses.[14] Clinically, higher rifampicin doses have been safely and successfully used for patients exhibiting low plasma rifampicin levels while on ATT, and in patients with severe disease (including meningeal tuberculosis).[15] High-dose rifampicin has shown equivalent safety profile in patients with tuberculous meningitis, but does not reduce mortality.[1617] A recent study found that treatment with higher rifampicin doses may also augment host immunological response against pulmonary tuberculosis.[18] Therefore, consolidating the evidence base on this topic with reference to current standards of treating pulmonary tuberculosis is an unmet need.[19] We conducted this systematic review and meta-analysis to evaluate if higher daily oral rifampicin doses (≥15 mg/kg/day) administered to patients with pulmonary tuberculosis were associated with different therapeutic efficacies or incidence of adverse effects, when compared to standard-dose rifampicin. We also explore if progressively increasing daily rifampicin doses show any trend towards greater efficacy or harm. METHODS The approach for this systematic review and meta-analysis is outlined in a protocol registered with the PROSPERO International Prospective Register of Systematic Reviews (ID CRD42024510530). Approval from our Institutional Ethics Committee was not required as we extracted and summarised already published information. We report our findings in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.[20] Search strategy We initially searched PubMed and Embase databases and the Cochrane Central Register of Controlled Trials (CENTRAL) for publications indexed up to 15 February 2024 using the search terms (rifampicin OR rifampin) AND (dosage OR dose) AND (high OR higher OR increased OR greater) AND (tuberculosis OR TB OR mycobacterium). No geographic, linguistic, or date restrictions were applied. We further updated our search on 1 October 2024. Additionally, relevant publications were identified from the bibliographies of research articles and reviews, and from our personal files. Study selection We included parallel-group randomised controlled trials (RCTs) that compared the efficacy or safety of various rifampicin doses for treating rifampicin-susceptible active pulmonary tuberculosis in adult human subjects. We ensured that each included study had at least one group receiving oral rifampicin in higher doses from treatment initiation for at least the intensive phase of therapy, and at least another group receiving oral rifampicin in a standard dose throughout treatment. Tuberculosis was microbiologically confirmed for study participants based on sputum smear, molecular testing, or culture analysis. Both high-dose and standard-dose treatment regimens contained at least isoniazid and rifampicin, administered for a minimum of 4 months (or 16 weeks), with at least ethambutol or pyrazinamide in the initial intensive phase. We targeted publications reporting on daily ATT at least in the intensive phase as intermittent regimens throughout do not reflect current standards of tuberculosis care. Daily ATT regimens were considered as those in which patients received drugs for at least 5 days in a week. We defined standard-dose rifampicin as a daily dose of rifampicin of 10 mg/kg based on current guidance for tuberculosis treatment, and high-dose rifampicin as a daily rifampicin dosage of 15 mg/kg or greater. For fixed rifampicin dosing regimens, we considered daily rifampicin doses of up to 600 mg, and 900 mg or greater, as standard and high doses, respectively, and converted it to a per-unit-weight dose for a 60 kg person for our analysis. Identification of studies fulfilling our search criteria was done independently by two reviewers (ANA and RA). Any disagreements were resolved through mutual discussion or by consulting a third adjudicator (SD) in the absence of consensus. After removing duplicate citations, we screened all study titles and abstracts identified from our literature search. Apart from studies not fulfilling our inclusion criteria, we also excluded conference presentations, review articles, editorials, case reports, letters to editor not describing original research, and clinical trial protocols. We then evaluated full-text versions of the remaining potentially relevant publications against our eligibility criteria. Where multiple publications described any outcome of interest among the overlapping or same study population, we chose the one with a larger sample size or one reporting greater details on that outcome as our primary dataset. We finally selected studies meeting all our inclusion criteria and describing one or more outcomes of interest (see below). Study endpoints Our co-primary efficacy endpoints were the proportion of patients with positive mycobacterial culture at end of intensive phase (2 months or 8 weeks) and at end of treatment (6 months or 24–26 weeks). The primary safety endpoint was incidence of severe hepatotoxicity during treatment. Secondary efficacy endpoints were time to mycobacterial culture conversion (assessed at the end of intensive phase and at end of treatment), proportion of patients with mycobacterial culture positivity at 1, 3, and 4 months (or equivalent weeks), death during treatment, treatment failure, recurrence after treatment completion, and loss to follow-up. Secondary safety endpoints were incidence of any adverse event, serious adverse event, drug-related serious adverse event, any hepatotoxicity, and treatment interruption/discontinuation due to adverse effects. Data extraction We extracted the following information from the included studies: Location and period of study, trial design, participant inclusion and exclusion criteria, demographic and clinical characteristics of study subjects, human immunodeficiency virus (HIV) status, method of tuberculosis diagnosis, drugs and their dosing and duration in each treatment arm, any co-interventions (if applicable), number of subjects randomised and analysed in each rifampicin regimen, supervision of drug administration, and outcomes of interest in each treatment arm. Wherever available, we used results from Mycobacterial Growth Indicator Tube (MGIT) platform over Lowenstein–Jenson slopes for analysing mycobacterial culture data. Wherever studies stratified adverse events using tools like Common Terminology Criteria for Adverse Events (CTCAE), or Division of AIDS (DAIDS) grading, we considered serious adverse events as those that were severe or potentially life-threatening or resulted in death (i.e., grade 3 or higher).[2122] If studies reported adverse effects as number of events, rather than number of participants experiencing events, we extracted data only for outcomes that were likely to occur a single time (e.g., treatment interruption, mortality). Study quality and publication bias Risk of bias appraisal was done independently for each included study using the Cochrane’s Risk of Bias Tool version 2.0.[23] This tool checks if adequate measures were undertaken to reduce the risk of bias arising from five specific domains: (a) randomisation process, (b) deviations from intended interventions, (c) missing outcome data, (d) outcome measurement, and (e) selective reporting. For each domain, risk of bias was graded as being ‘low risk’, ‘some concerns’, or ‘high risk’. A study was considered as having an overall high risk of bias if any domain was rated as high risk. Risk of publication bias was assessed by visual appraisal of funnel plot asymmetry, and Begg and Mazumdar’s rank correlation test, only when at least ten studies were available for meta-analysis of any pre-specified outcome.[2425] Statistical analysis We calculated risk ratios (RRs) and corresponding 95% confidence intervals (95% CI) for each study describing the number of subjects experiencing binary outcomes in the high-dose and standard-dose treatment arms. We extracted hazard ratio (HR) and corresponding 95% CI for time-to-event data. Wherever feasible, we assessed drug safety from patients administered any drug dose in the intention-to-treat (ITT) study population and evaluated drug efficacy from the modified intention-to-treat (mITT) study population. If results were available only for the per-protocol analysis, we assigned a corresponding favourable outcome to subjects not being counted (e.g., those unable to submit sputum specimens for culture at follow-up) before analysis. For studies reporting zero events in any one group, we applied a continuity correction by adding 0.5 to all values in 2 × 2 tables containing an empty cell. We performed meta-analyses using the statistical package R (version 4.4.2; R Foundation for Statistical Computing). When two or more clinical trials assessed the same outcome, their data were summarised using random effects inverse variance models according to the DerSimonian–Laird method.[26] If data from multiple treatment regimens were reported in any study, we attempted to pool information from all high-dose and all standard-dose regimens prior to our initial comparisons. We expressed between-study heterogeneity quantitatively through Higgins’ inconsistency index (I2) and considered it high for values >75%.[27] Wherever I2 value exceeded 50%, heterogeneity was further explored through a separate subgroup analysis if at least ten valid studies were available for the primary analysis. For subgroup analysis, we stratified studies based on the following: (a) risk of bias assessment (low, some concerns, or high), (b) number of participants receiving high-dose rifampicin (<100 or more), (c) studies using high-dose rifampicin throughout or only during intensive phase of ATT, and (d) studies including people living with HIV (PLHIV) or not. As a sensitivity analysis, we explored the robustness of our estimates for the primary outcomes by performing meta-analysis by another frequentist technique (Hartung-Knapp-Sidik-Jonkman method) and through Bayesian approaches with vague priors (assuming log of RR to have a normal distribution with a mean of zero and a standard deviation of 10) and semi-informative priors (assuming log of RR to have a normal distribution with a mean of zero and a standard deviation of one, with half-normal distribution of heterogeneity parameter).[282930] We also calculated the summary estimates after excluding studies with some concerns or high risk of bias. Since different high-dose rifampicin regimens could have resulted in different outcomes, the analyses were also stratified according to the daily rifampicin dose administered to the high rifampicin dose groups (15, 20–25, and ≥30 mg/kg). Forest plots were constructed to graphically display various meta-analyses results. Two-tailed statistical significance was based on individual 95% CIs excluding the null value. We assumed a largely similar baseline risk across studies for each of the outcome measures being assessed, and calculated the number needed to treat for an additional beneficial outcome (NNTB), or for an additional harmful outcome (NNTH), as the reciprocal of corresponding summary risk difference (RD) values.[3132] We also performed a meta-regression analysis to evaluate the impact of various high rifampicin doses on our primary outcomes. We also summarised the overall certainty of evidence for our three primary outcomes using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach.[33] RESULTS We identified 5668 citations from our literature search and finally selected 12 RCTs for data synthesis [Figure 1].[343536373839404142434445] Reasons for excluding other publications after full-text review are summarised in Table S1 of online supplement. All studies selected for this review were published during the last 9 years and include datasets from the RIFATOX,[3436] HIGHRIF,[35] MAMS-TB,[37] HIRIF,[38] RIFAVIRENZ,[39] OneRIF,[40] RIFASHORT,[41] SAEFRIF,[42] TRIDORE,[43] HICON-R,[44] and Hi-DoRi-3[45] clinical trials. Overall, 3230 subjects were randomised to receive ATT in these studies, 1310 with standard-dose rifampicin regimen, and 1920 with high rifampicin doses ranging between 15 and 35 mg/kg/day. Study selection flowchart Table S1Studies excluded after full-text review, and reasons for exclusionPublication detailsReason for exclusionVerbist L. Pharmacological study of rifampicin after repeated high dosage during intermittent combined therapy. I. Variation of the rifampicin serum levels (947 determinations). Respiration 1971;28 Suppl:7-16.Intermittent rifampicin dosingAnastasatu C, Bungeteanu G, Sibila S. The intermittent chemotherapy of tuberculosis with rifampicin regimens on ambulatory basis. Scand J Respir Dis Suppl 1973;84:136-9.Intermittent rifampicin dosingNitti V. Controlled clinical evaluation of three intermittent regimens employing low-dosage schedules of rifampicin in original treatment of pulmonary tuberculosis. Preliminary data on effectiveness and side effects. Scand J Respir Dis Suppl 1973;84:180-5.Daily rifampicin dose 600 mgZierski M. A trial of intermittent rifampicin and ethambutol in retreatment regimens. Scand J Respir Dis Suppl 1973;84:132-5.Daily rifampicin dose 600 mgEule H, Werner E, Winsel K, Iwainsky H. Intermittent chemotherapy of pulmonary tuberculosis using rifampicin and isoniazid for primary treatment: the influence of various factors on the frequency of side effects. Tubercle 1974;55:81-9.Intermittent rifampicin dosingHong Kong Tuberculosis Treatment Services, British Medical Research Council. A controlled clinical trial of small daily doses of rifampicin in the prevention of adverse reactions to the drug in a once-weekly regimen of chemotherapy in Hong Kong: second report:-the results at 12 months. Tubercle 1974;55:193-210.Intermittent rifampicin dosingHong Kong Tuberculosis Treatment Services, Brompton Hospital, British Medical Research Council. A controlled clinical trial of daily and intermittent regimens of rifampicin plus ethambutol in the retreatment of patients with pulmonary tuberculosis in Hong Kong. Tubercle 1974;55:1-27.Standard daily rifampicin doseCooperative Tuberculosis Chemotherapy Study in Poland. A comparative study of daily followed by twice or once weekly regimens of ethambutol and rifampicin in retreatment of patients with pulmonary tuberculosis. The results at 1 year. Tubercle 1975;56:1-26.Daily rifampicin dose 600 mgSingapore Tuberculosis Service, British Medical Research Council. Controlled trial of intermittent regimens of rifampicin plus isoniazid for pulmonary tuberculosis in Singapore. Lancet 1975;2:1105-9.Intermittent rifampicin dosingKreis B, Pretet S, Birenbaum J, Guibout P, Hazeman JJ, Orin E, et al. Two three-month treatment regimens for pulmonary tuberculosis. Bull Int Union Tuberc 1976;51:71-5.Anti-tubercular therapy for <4 monthsZierski M, Bek E, Bergson H, Kucharska A, Szelagowicz B. Retreatment of chronic pulmonary tuberculosis with regimens including high and low doses of rifampicin in the intermittent phase recent and late results - a controlled comparison study. Bull Int Union Tuberc 1976;51:121-6.Daily rifampicin dose 600 mgSingapore Tuberculosis Service, British Medical Research Council. Controlled trial of intermittent regimens of rifampin plus isoniazid for pulmonary tuberculosis in Singapore. The results up to 30 months. Am Rev Respir Dis 1977;116:807-20.Intermittent rifampicin dosingLong MW, Snider DE, Jr., Farer LS. U.S. Public Health Service Cooperative trial of three rifampin-isoniazid regimens in treatment of pulmonary tuberculosis. Am Rev Respir Dis 1979;119:879-94.Daily rifampicin dose 750 mgAnastasatu C, Bercea O, Corlan E. Pilot study on short-course (3+3 months) chemotherapy starting with quadruple (HRSZ) intermittent (2/7) regimens in new smear-positive cases of pulmonary tuberculosis. Bull Int Union Tuberc 1984;59:23-6.Intermittent rifampicin dosingTsang SH, Lum D, Chan SL. Hong Kong Chest Service/British Medical Research Council study of a fully supervised programme of chemotherapy for pulmonary tuberculosis given once weekly in the continuation phase in the rural areas of Hong Kong. Bull Int Union Tuberc 1984;59:204.Standard daily rifampicin doseHong YP, Kim SC, Chang SC, Kim SJ, Jin BW, Park CD. Comparison of a daily and three intermittent retreatment regimens for pulmonary tuberculosis administered under programme conditions. Tubercle 1988;69:241-53.Intermittent rifampicin dosingCohn DL, Catlin BJ, Peterson KL, Judson FN, Sbarbaro JA. A 62-dose, 6-month therapy for pulmonary and extrapulmonary tuberculosis. A twice-weekly, directly observed, and cost-effective regimen. Ann Intern Med 1990;112:407-15.Cross-sectional study designTe Water Naude JM, Donald PR, Hussey GD, Kibel MA, Louw A, Perkins DR, et al. Twice weekly vs. daily chemotherapy for childhood tuberculosis. Pediatr Infect Dis J 2000;19:405-10.Standard daily rifampicin doseJindani A, Nunn AJ, Enarson DA. Two 8-month regimens of chemotherapy for treatment of newly diagnosed pulmonary tuberculosis: international multicentre randomised trial. Lancet 2004;364:1244-51.Standard daily rifampicin doseOkwera A, Johnson JL, Luzze H, Nsubuga P, Kayanja H, Cohn DL, et al. Comparison of intermittent ethambutol with rifampicin-based regimens in HIV-infected adults with PTB, Kampala. Int J Tuberc Lung Dis 2006;10:39-44.Standard daily rifampicin doseRuslami R, Nijland H, Aarnoutse R, Alisjahbana B, Soeroto AY, Ewalds S, et al. Evaluation of high- versus standard- dose rifampin in Indonesian patients with pulmonary tuberculosis. Antimicrob Agents Chemother 2006;50:822-3.Daily rifampicin dose 600 mgDiacon AH, Patientia RF, Venter A, van Helden PD, Smith PJ, McIlleron H, et al. Early bactericidal activity of high-dose rifampin in patients with pulmonary tuberculosis evidenced by positive sputum smears. Antimicrob Agents Chemother 2007;51:2994-6.High-dose rifampicin for only five daysRuslami R, Nijland HM, Alisjahbana B, Parwati I, van Crevel R, Aarnoutse RE. Pharmacokinetics and tolerability of a higher rifampin dose versus the standard dose in pulmonary tuberculosis patients. Antimicrob Agents Chemother 2007;51:2546-51.Daily rifampicin dose 600 mgLin CB, Tang M, Hsu AH, Miu WC, Lee YS, Lee JJ. Pilot study of twice-weekly therapy for pulmonary tuberculosis in Taiwan. J Formos Med Assoc 2011;110:438-45.Intermittent rifampicin dosingIto K, Fukuuchi K, Kaguraoka S, Watanabe H, Yoshiyama T, Urakawa M, et al. Importance of monitoring the drug dosage and treatment duration for the management of tuberculosis patients at public health centers. Kekkaku 2014;89:771-6.Standard daily rifampicin doseZheng X, Zhong F, Zhang X. Doctors’ compliance with national guidelines and clinical pathway on the treatment of tuberculosis inpatients in Hubei, China. J Eval Clin Pract 2014;20:288-93.Cross-sectional study designBoeree MJ, Diacon AH, Dawson R, Narunsky K, du Bois J, Venter A, et al. A dose-ranging trial to optimize the dose of rifampin in the treatment of tuberculosis. Am J Respir Crit Care Med 2015;191:1058-65.High-dose rifampicin for only 14 daysAlvarez-Uria G, Midde M, Naik PK. Mortality in HIV-infected patients with tuberculosis treated with streptomycin and a two-week intensified regimen: data from an HIV cohort study using inverse probability of treatment weighting. PeerJ 2016;4:e2053.High-dose rifampicin for only 14 daysPeloquin CA, Velásquez GE, Lecca L, Calderón RI, Coit J, Milstein M, et al. Pharmacokinetic evidence from the HIRIF trial to support increased doses of rifampin for tuberculosis. Antimicrob Agents Chemother 2017;61:e00038-17.Inadequate information on outcomes of interestSvensson EM, Svensson RJ, Te Brake LHM, Boeree MJ, Heinrich N, Konsten S, et al. The potential for treatment shortening with higher rifampicin doses: Relating drug exposure to treatment response in patients with pulmonary tuberculosis. Clin Infect Dis 2018;67:34-41.Modeling studySvensson RJ, Aarnoutse RE, Diacon AH, Dawson R, Gillespie SH, Boeree MJ, et al. A population pharmacokinetic model incorporating saturable pharmacokinetics and autoinduction for high rifampicin doses. Clin Pharmacol Ther 2018;103:674-83.Modeling studySvensson RJ, Svensson EM, Aarnoutse RE, Diacon AH, Dawson R, Gillespie SH, et al. Greater early bactericidal activity at higher rifampicin doses revealed by modeling and clinical trial simulations. J Infect Dis 2018;218:991-9.Modeling studyAljayyoussi G, Donnellan S, Ward SA, Biagini GA. Intracellular PD modelling (PD(i)) for the prediction of clinical activity of increased rifampicin dosing. Pharmaceutics 2019;11:278.Modeling studyAruldhas BW, Hoglund RM, Ranjalkar J, Tarning J, Mathew SK, Verghese VP, et al. Optimization of dosing regimens of isoniazid and rifampicin in children with tuberculosis in India. Br J Clin Pharmacol 2019;85:644-54.Modeling studyLitjens CHC, Aarnoutse RE, van Ewijk-Beneken Kolmer EWJ, Svensson EM, Colbers A, Burger DM, et al. Protein binding of rifampicin is not saturated when using high-dose rifampicin. J Antimicrob Chemother 2019;74:986-90.Inadequate information on outcomes of interestRadtke KK, Dooley KE, Dodd PJ, Garcia-Prats AJ, McKenna L, Hesseling AC, et al. Alternative dosing guidelines to improve outcomes in childhood tuberculosis: a mathematical modelling study. Lancet Child Adolesc Health 2019;3:636-45.Modeling studySeijger C, Hoefsloot W, Bergsma-de Guchteneire I, Te Brake L, van Ingen J, Kuipers S, et al. High-dose rifampicin in tuberculosis: Experiences from a Dutch tuberculosis centre. PLoS One 2019;14:e0213718.Cross-sectional study designSvensson RJ, Niward K, Davies Forsman L, Bruchfeld J, Paues J, Eliasson E, et al. Individualised dosing algorithm and personalised treatment of high-dose rifampicin for tuberculosis. Br J Clin Pharmacol 2019;85:2341-50.Cross-sectional study designDu J, Zhang L, Ma Y, Chen XY, Ge QP, Tian XZ, et al. Treatment and recurrence on re-treatment tuberculosis patients: A randomized clinical trial and 7-year perspective cohort study in China. Eur J Clin Microbiol Infect Dis 2020;39:93-101.Standard daily rifampicin doseSun Y, Cao B, Rao W, Li X, Yu B, Zhang T, et al. Comparison of efficacy and safety of two fixed-dose combination regimens for tuberculosis in the Chinese population. Infezioni in Medicina 2020;28:565-71.Standard daily rifampicin doseVelayutham B, Jawahar MS, Nair D, Navaneethapandian P, Ponnuraja C, Chandrasekaran K, et al. 4-month moxifloxacin containing regimens in the treatment of patients with sputum-positive pulmonary tuberculosis in South India - a randomised clinical trial. Trop Med Int Health 2020;25:483-95.Standard daily rifampicin doseYu B, Zhang T, Zhang P, Sun Y, Cao B, Rao W, et al. Comparison of clinical effect of fixed-dose combination for tuberculosis treatment in Hebei and Jilin Province, China. Acta Medica Mediterranea 2020;36:779-85.Cross-sectional study designGarcia-Prats AJ, Svensson EM, Winckler J, Draper HR, Fairlie L, van der Laan LE, et al. Pharmacokinetics and safety of high-dose rifampicin in children with TB: the Opti-Rif trial. J Antimicrob Chemother 2021;76:3237-46.High-dose rifampicin for only one daySaid BN, Heysell SK, Yimer G, Aarnoutse RE, Kibiki GS, Mpagama S, et al. Pharmacodynamic biomarkers for quantifying the mycobacterial effect of high doses of rifampin in patients with rifampin-susceptible pulmonary tuberculosis. Int J Mycobacteriol 2021;10:457-62.Study population overlapping with another selected studySanni S, Wachinou AP, Merle CSC, Bekou KW, Esse M, Gossa S, et al. Hepatic safety of high-dose rifampicin for tuberculosis treatment in TB/HIV co-infected patients: a randomized clinical trial. Arch Pharm Pract 2021;12:66-72.Inadequate information on outcomes of interestTe Brake LHM, de Jager V, Narunsky K, Vanker N, Svensson EM, Phillips PPJ, et al. Increased bactericidal activity but dose-limiting intolerability at 50 mg.kg(-1) rifampicin. Eur Respir J 2021;58:2000955.High-dose rifampicin for only seven daysTrentalange A, Borgogno E, Motta I, Antonucci M, Pirriatore V, Costa C, et al. Rifampicin and isoniazid maximal concentrations are below efficacy-associated thresholds in the majority of patients: Time to increase the doses? Int J Antimicrob Agents 2021;57:106297.Cross-sectional study designZhang N, Savic RM, Boeree MJ, Peloquin CA, Weiner M, Heinrich N, et al. Optimising pyrazinamide for the treatment of tuberculosis. Eur Respir J 2021;58:2002013.Study population overlapping with another selected studyKathamuthu GR, Bhavani PK, Singh M, Saini JK, Aggarwal A, Ansari MSS, et al. High-Dose rifampicin mediated systemic alterations of cytokines, chemokines, growth factors, microbial translocation markers, and acute-phase proteins in pulmonary tuberculosis. Front Pharmacol 2022;13:896551.Inadequate information on outcomes of interestDawson R, Diacon AH, Narunsky K, De Jager VR, Stinson KW, Zhang X, et al. Phase I single ascending dose and food effect study in healthy adults and phase I/IIa multiple ascending dose study in patients with pulmonary tuberculosis to assess pharmacokinetics, bactericidal activity, tolerability, and safety of OPC-167832. Antimicrob Agents Chemother 2023;67:e0147722.Standard daily rifampicin doseKengo A, Gausi K, Nabisere R, Musaazi J, Buzibye A, Omali D, et al. Unexpectedly low drug exposures among Ugandan patients with TB and HIV receiving high-dose rifampicin. Antimicrob Agents Chemother 2023;67:e0043123.Inadequate information on outcomes of interestKengo A, Nabisere R, Gausi K, Musaazi J, Buzibye A, Omali D, et al. Dolutegravir pharmacokinetics in Ugandan patients with TB and HIV receiving standard- versus high-dose rifampicin. Antimicrob Agents Chemother 2023;67:e0043023.Inadequate information on outcomes of interestMartinecz A, Boeree MJ, Diacon AH, Dawson R, Hemez C, Aarnoutse RE, et al. High rifampicin peak plasma concentrations accelerate the slow phase of bacterial decline in tuberculosis patients: Evidence for heteroresistance. PLoS Comput Biol 2023;19:e1011000.High-dose rifampicin for only seven daysPaton NI, Cousins C, Suresh C, Burhan E, Chew KL, Dalay VB, et al. Treatment strategy for rifampin-susceptible tuberculosis. N Engl J Med 2023;388:873-87.92% subjects in high-dose rifampicin arms received treatment for ≤12 weeksPerumal R, Naidoo K, Naidoo A, Letsoalo MP, Esmail A, Joubert I, et al. The impact of enteral feeding and therapeutic monitoring of rifampicin with dose escalation in critically ill patients with tuberculosis. Int J Infect Dis 2023;126:174-80.Cross-sectional study designStemkens R, Jager V, Dawson R, Diacon AH, Narunsky K, Padayachee SD, et al. Drug interaction potential of high-dose rifampicin in patients with pulmonary tuberculosis. Antimicrob Agents Chemother 2023;67:e0068323.High-dose rifampicin for only 15 daysSusanto BO, Svensson EM, Te Brake L, Aarnoutse RE, Boeree MJ, Simonsson USH, et al. Model-based analysis of bactericidal activity and a new dosing strategy for optimised-dose rifampicin. Int J Antimicrob Agents 2023;61:106813.High-dose rifampicin for only seven days Studies excluded after full-text review, and reasons for exclusion Study characteristics and interventions All included studies were phase 2 or 3 RCTs conducted across Africa, Asia, and South America [Table 1], with three (25.0%) operating concurrently in multiple countries.[343741] Most study participants were men, ranging between 63% and 90% across the 12 datasets [Table 1]. One study enrolled only subjects with prior history of tuberculosis treatment,[43] while another reported 7.0% such study participants.[40] Two (16.7%) studies were conducted exclusively on PLHIV,[3942] while six (50.0%) studies excluded all such individuals [Table 1].[343640414445] Tuberculosis was diagnosed using smear microscopy, culture, and/or molecular testing on respiratory specimens, except for one (9.1%) study on PLHIV that additionally included urinary lipoarabinomannan as a diagnostic criterion.[42] Sputum smear grading was reported in six (50.0%) publications, and most subjects in these studies showed 2+ smear positivity or higher [Table 2].[353638414445] Chest radiographic severity was assessed in only four (33.3%) studies [Table 2]. Cavitary disease was present in <50% subjects in three studies,[383944] and in 86.8% subjects in another study.[41] In two studies, most subjects had advanced lung disease or involvement of more than two lung zones.[4144] Two (16.7%) studies were placebo-controlled,[3538] while the remaining were open-label. Table 1Study characteristicsAuthorStudy designCountries where study conductedInclusion criteriaExclusion criteriaPeriod of enrolmentMethod of TB diagnosisAge (years) and genderWeight (kg) and BMI (kg/m2)Subjects living with HIVJindani[34]Phase 2B open label RCTBolivia, Nepal, UgandaAdults (18–65 years) with TB with <1 month of prior ATT intakeCritically ill; extra-pulmonary TB; drug-resistant TB; alcoholism; pregnancy; HIV seropositivity; psychiatric or hematologic or hepatic or neurologic or renal diseaseFeb 2011 to May 2013Sputum smearMedian 30; 68.3% malesMedian 50; NSNoneAarnoutse[35]Phase 2B placebo controlled RCTTanzaniaAdults (18–65 years) with TBATT in previous 3 years; clinical liver disease or hepatic dysfunction; azotaemia; current or planned ART; Karnofsky score <40; pregnancy or lactation; rifampicin resistanceNSSputum smearMedian 34 (27–40); 90.0% malesMedian 55.5 (52.0–59.3); NS15 (10%)Atwine36Phase 2 open label RCTUgandaHIV negative adults (>18 years) with TB with <1 month of prior ATT intakeCritically ill; extra-pulmonary TB; alcoholism; psychiatric illness; blood disorders; diabetes; epilepsy; peripheral neuritis; pregnancy; severe anaemia; significant hepatic or renal dysfunction; rifampicin resistanceFeb 2011 to May 2013Sputum smearMean 36.2±11.6; 80.0% malesMean 51.2±7.8; Mean 19.0±2.4NoneBoeree[37]Phase 2B open label RCTTanzania, South AfricaAdults (≥18 years) with previously untreated rifampicin-susceptible TB; weight 35–90 kgPoor general condition; pregnancy or lactation; HIV seropositive on ART or with CD4 count <200; known intolerance to study drugs; significant metabolic or gastrointestinal or hepatic or neurologic or psychiatric or cardiac or endocrine disorder or malignancy; TB in last 5 years; use of CYP450 inducers or inhibitorsMay 2013 to Mar 2014Sputum smear, GeneXpertMedian 33 (26–40); 70.7% malesMedian 53 (49–58); NS24 (6.6%)Velásquez[38]Phase 2 placebo controlled RCTPeruTreatment naïve adults (18–60 years) with drug-susceptible TBWeight <30 kg; extrapulmonary TB; significant haemoptysis; pregnancy or lactation; severe anaemia or leukopenia; significant hepatic dysfunction or hepatitis B or C seropositivity; uncontrolled diabetes; Karnofsky score <50Sep 2013 to Feb 2015Sputum smear, Hain GenoType MTBDRplusMedian 25 (21–37); 63.3% malesNS; Median 21.2 (19.6–23.7)5 (2.9%)Atwine[39]Phase 2 open label RCTUgandaHIV seropositive ART naïve adults (≥18 years) with rifampicin-susceptible TB; fit for ambulatory therapy; weight >35 kgDrug-resistant TB; opportunistic infections; Karnofsky score <80; severe anaemia; active hepatitis; psychiatric illness; pregnancyMar 2014 to Aug 2016Sputum GeneXpertNS; 73.2% malesNS; NSAllMaug[40]Phase 3 open label RCTBangladeshAdults (≥15 years) with TBDrug-resistant TB; need for hospitalisation because of poor clinical condition; clinically active liver disease; known hepatitis B or C or HIV infection; pregnancyNov 2014 to Sep 2015Sputum smearMedian 44 (28–55); 73.2% malesMedian 42 (36–46); Median 16.4 (15.1–17.8)NoneJindani[41]Phase 3 open label RCTBotswana, Guinea, Nepal, Pakistan, Peru, UgandaAdults (≥18 years) with drug susceptible TB and <1 week of prior ATT intakeHIV seropositivity; weight <35 kg; diabetes; significant liver, renal or psychiatric disease; severe anaemia; critically ill; pregnancy or lactationJan 2017 to Dec 2020Sputum GeneXpertMedian 29; 85.5% malesMedian 52; Median 18.6NoneSekaggya-Wiltshire[42]Phase 2B open label RCTUgandaHIV seropositive adults (>18 years) with newly diagnosed TBRifampicin resistance; pregnancy; hepatic or renal dysfunctionApr 2019 to Feb 2021Sputum GeneXpert, culture; Urinary LAMMedian 36 (30–43); 63.3% malesNS; Median 19.3 (17.7–21.7)AllSouleymane[43]Phase 3 open label RCTNigerAdults and children with recurrent TBRifampicin resistance; hepatic dysfunction; pregnancy or lactation; HIV co-infected patients requiring protease inhibitorMar 2021 to Feb 2022Sputum smear, GeneXpertMedian 37 (30–48); 87.4% malesNS; Median 18.1 (16.3–19.7)3 (2.4%)Bhavani[44]Phase 2B open label RCTIndiaAdults (≥18 years) with drug susceptible TBPsychiatric illness or drug abuse; diabetics receiving insulin; baseline hepatic dysfunction; alcohol dependence; HIV seropositivity; baseline isoniazid resistanceJan 2018 to Nov 2019Sputum smear, GeneXpertMean 35.0±12.3; 70.0% malesNS; Mean 18.3±3.1NoneKwak[45]Phase 3 open label RCTSouth KoreaAdults (≥19 years) with drug susceptible TBHIV seropositivity; already receiving ATT for >7 days; chronic liver disease; malignancy needing chemotherapy; contraindication for study drugs; pregnancy or planned pregnancy or lactationNov 2020 to Jan 2022Sputum/bronchoscopic GeneXpert, cultureMedian 60; 62.1% malesNS; NSNoneART=Anti-retroviral treatment, ATT=Anti-tubercular treatment, BMI=Body mass index, HIV=Human immunodeficiency virus, LAM=Lipoarabinomannan, NS=Not specified, RCT=Randomized controlled trial, TB=Tuberculosis Study characteristics ART=Anti-retroviral treatment, ATT=Anti-tubercular treatment, BMI=Body mass index, HIV=Human immunodeficiency virus, LAM=Lipoarabinomannan, NS=Not specified, RCT=Randomized controlled trial, TB=Tuberculosis Table 2Baseline microbiologic and radiologic profiles of study participantsStudySputum microscopySputum GeneXpertSputum cultureChest radiographyJindani[34]300/300 positive No information on gradingNot reportedLJ cultures done, but baseline data not reportedNot reportedAarnoutse[35]150/150 positive 1+ 15, 2+ 51, 3+ 55, 4+ 16Not reported143/150 MGIT culture positive, 138/150 LJ culture positiveNot reportedAtwine[36]100/100 positive Scant 4, 1+ 9, 2+ 25, 3+ 62Not reported95/100 MGIT culture positive, 98/100 LJ culture positive, 100/100 positive on eitherNot reportedBoeree[37]365/365 positive No information on grading365/365 positiveMGIT and LJ cultures done, but baseline data not reportedNot reportedVelásquez[38]180/180 positive 2+ 87, 3+ 93Not reportedMGIT and LJ cultures done, but baseline data not reportedCavitary disease 66/178Atwine[39]81/96 positive No information on grading98/98 positive86/97 MGIT culture positive, 77/97 LJ culture positiveCavitary disease 42/97Maug[40]701/701 positive No information on gradingDone, but baseline data not reportedLJ cultures done, but baseline data not reportedNot reportedJindani[41]625/672 positive 1+ 251, 2+ 150, 3+ 224672/672 positiveLJ/Ogawa cultures done, but baseline data not reportedCavitary disease 583/672 Minimal disease 20/662 Moderate disease 297/662 Advanced disease 345/662Sekaggya-Wiltshire[42]Not assessedDone, but baseline data not reportedMGIT and LJ cultures done, but baseline data not reportedNot reportedSouleymane[43]127/127 positive No information on gradingNot reportedCultures done, but baseline data not reportedNot reportedBhavani[44]333/333 positive Scanty/1+ 75, 2+/3+ 258333/333 positive331/333 MGIT/LJ culture positiveCavitary disease 162/333 ≤2 lung zones 142/333 >2 lung zones 191/333Kwak[45]16/58 positive Trace 2, 1+ 6, 2+ 2, 3+ 1, 4+ 5Done, but baseline data not reported47/58 MGIT culture positiveNot reportedLJ=Lowenstein-Jensen, MGIT=Mycobacterial Growth Indicator Tube Baseline microbiologic and radiologic profiles of study participants LJ=Lowenstein-Jensen, MGIT=Mycobacterial Growth Indicator Tube The included studies investigated several different treatment schedules and rifampicin dosages [Table 3]. All patients received daily (7 days a week) ATT throughout, except in one study where drugs were administered thrice-weekly during the continuation phase.[38] Drug intake was supervised in eight (66.7%) studies.[3435363840414344] Drugs were self-administered in one (8.3%) study,[37] and three (25.0%) studies did not provide any information.[394245] All protocols allowed drug dose adjustment based on patient weight-bands, except in one (8.3%) study where drug dosing was fixed.[35] One study evaluated shorter ATT regimens (4 months) for its experimental arms [Table 3].[41] Another study had a flexible intensive phase, and participants were switched from intensive to continuation phase of ATT at point of sputum culture conversion.[45] High-dose rifampicin was used throughout the treatment duration in three (25.0%) studies.[404145] One study also concurrently investigated another investigational anti-tubercular drug candidate (SQ109) in some subjects receiving standard-dose or high-dose rifampicin.[37] Another study administered both high-dose rifampicin and high-dose isoniazid to subjects in their experimental arm.[43] One study also simultaneously randomised PLHIV to dolutegravir- or efavirenz-based anti-retroviral therapy.[42] Although six (50.0%) studies collected pharmacokinetic data, none correlated rifampicin blood levels with therapeutic outcomes.[353738394244] Table 3Anti-tubercular treatment regimens and rifampicin dosages in the included studiesStudyTreatment arms with high-dose rifampicinTreatment arms with standard-dose rifampicin Drug treatment scheduleaRifampicin doseNo. of subjectsbDrug treatment scheduleaRifampicin doseNo. of subjectsbJindani[34](8w)R’HZE/(8w)R’H/(10w)RH (8w)R’HZE/(8w)R’H/(10w)RH15 mg/kg 20 mg/kg100, 80 100, 91(8w)RHZE/(18w)RH10 mg/kg100, 92Aarnoutse[35](2m)R’HZE/(4m)RH (2m)R’HZE/(4m)RH900 mg 1200 mg50, 48 50, 482RHZE/4RH600 mg50, 49Atwine[36](2m)R’HZE/(2m)R’H/(2m)RH (2m)R’HZE/(2m)R’H/(2m)RH15 mg/kg 20 mg/kg32, 32 34, 342RHZE/4RH10 mg/kg34, 34Boeree[37](12w)R’HZX/(14w)RH (12w)R’HZM/(14w)RH (12w)R’HZE/(14w)RH20 mg/kg 20 mg/kg 35 mg/kg57, 56 63, 63 63, 63(8w)RHZE/(18w)RH (12w)RHZX/(4w)RH10 mg/kg 10 mg/kg123, 123 59, 58Velásquez[38]c(8w)R’HZE/(18w)RH (8w)R’HZE/(18w)RH15 mg/kg 20 mg/kg60, 58 60, 60(8w)RHZE/(18w)RH10 mg/kg60, 56Atwine[39](8w)R’HZE/(16w)RH20 mg/kg65, 64(8w)RHZE/(16w)RH10 mg/kg33, 33Maug[40](2m)R’HZE/(4m)R’H20 mg/kg353, 347(2m)RHZE/(4m)RH10 mg/kg348, 343Jindani[41](2m)R’HZE/(2m)R’H (2m)R’HZE/(2m)R’H1200 mg 1800 mg223, 212 225, 212(2m)RHZE/(4m)RH10 mg/kg224, 212Sekaggya-Wiltshire[42](8w)R’HZE/(16w)RH35 mg/kg61, 61(8w)RHZE/(16w)RH10 mg/kg67, 67Souleymane[43](6m)R’H’ZE30 mg/kg62, 62(6m)RHZE10 mg/kg65, 65Bhavani[44](2m)R’HZE/(4m)RHE (2m)R’HZE/(4m)RHE25 mg/kg 35 mg/kg114, 112 110, 106(2m)RHZE/(4m)RHE10 mg/kg109, 105Kwak[45](C)R’HZ/(12w)R’H30 mg/kg28, 26(2m)RHZE/(4m)RHE10 mg/kg38, 32aC Intensive phase completed after culture negativity, E Ethambutol, H Isoniazid, H’ Triple-dose isoniazid, m Months, M Moxifloxacin, R Rifampicin, R’ High-dose rifampicin, w Weeks, X New investigational drug, Z Pyrazinamide, bFigures derived from both intention-to-treat, and followed by modified intention-to-treat, study populations, cDrugs were administered daily in intensive phase and thrice weekly in continuation phase in this study; drugs were administered daily throughout in all other studies Anti-tubercular treatment regimens and rifampicin dosages in the included studies aC Intensive phase completed after culture negativity, E Ethambutol, H Isoniazid, H’ Triple-dose isoniazid, m Months, M Moxifloxacin, R Rifampicin, R’ High-dose rifampicin, w Weeks, X New investigational drug, Z Pyrazinamide, bFigures derived from both intention-to-treat, and followed by modified intention-to-treat, study populations, cDrugs were administered daily in intensive phase and thrice weekly in continuation phase in this study; drugs were administered daily throughout in all other studies All except three (25.0%) studies were considered as showing low risk of bias across all domains [Figure S1 of online supplement]. These three studies had some concerns over selection bias in the randomisation process domain as they did not report on concealment of allocation of their treatment arms.[354345] Primary outcomes Eight RCTs provided data for sputum culture positivity rates after 2 months (8 weeks) of ATT from 1334 and 795 subjects receiving high-dose and standard-dose rifampicin, respectively, in their mITT study population.[3435373839414244] Those on high-dose rifampicin had significantly lower culture positivity than those receiving standard-dose rifampicin (summary RR: 0.74, 95% CI: 0.62–0.89), with little heterogeneity between studies (I2 = 15.5%). Summary risk ratio estimates were largely similar for frequentist and Bayesian meta-analysis [Figure 2]. Results from all RCTs favoured high-dose rifampicin, except for one study, although such differences were significant only in two recent studies [Figure 3].[4144] Summary estimates remained similar even after excluding a single study with some concern of bias [Table S2 of online supplement].[35] Based on summary RD values, 18 patients need to be treated with high-dose rifampicin to reduce one positive sputum culture [Table 4]. Based on the GRADE criteria, the overall quality of evidence was rated as high [Table 5], providing high confidence that high-dose rifampicin improved sputum conversion at 2 months of ATT in this population. Summary risk ratio estimates for the primary outcomes using frequentist and Bayesian meta-analytic approaches Residual mycobacterial culture positivity during anti-tubercular treatment among study subjects receiving high-dose or standard-dose rifampicin Table S2Summary outcomes stratified by risk of bias of included studiesOverallStudies with low risk of biasStudies with some concern of bias No. of studiesSummary risk ratioI2, %No. of studiesSummary risk ratioI2, %No. of studiesSummary risk ratioI2, %Primary efficacy outcomes - Culture positivity at 2 months80.74 (0.62 to 0.89)15.570.70 (0.56 to 0.87)20.010.88 (0.62 to 1.25)- - Culture positivity at 6 months50.52 (0.23 to 1.15)16.550.52 (0.23 to 1.15)16.50––Primary safety outcome - Severe hepatotoxicity during treatment72.06 (1.15 to 3.68)0.052.01 (1.09 to 3.71)0.022.61 (0.30 to 23.0)27.3Secondary efficacy outcomes - Culture positivity at 1 month30.82 (0.66 to 1.02)89.220.80 (0.54 to 1.1994.410.84 (0.74 to 0.96)– - Culture positivity at 3 months30.77 (0.49 to 1.22)51.620.63 (0.32 to 1.22)58.211.09 (0.64 to 1.86)– - Culture positivity at 4 months30.80 (0.49 to 1.29)0.030.80 (0.49 to 1.29)0.00–– - Death during treatment80.77 (0.36 to 1.67)18.460.58 (0.27 to 1.28)9.522.21 (0.45 to 10.9)0.0 - Treatment failure50.86 (0.44 to 1.69)1.040.76 (0.29 to 1.98)21.610.62 (0.12 to 3.10)– - Recurrence after treatment61.21 (0.39 to 3.74)52.951.01 (0.30 to 3.39)57.616.11 (0.31 to 122)– - Loss to follow-up50.90 (0.48 to 1.69)0.050.90 (0.48 to 1.69)0.00––Secondary safety outcomes - Any adverse event during treatment81.06 (0.94 to 1.20)55.561.07 (0.93 to 1.23)60.621.17 (0.69 to 1.99)62.7 - Serious adverse events during treatment101.26 (0.92 to 1.72)9.071.16 (0.85 to 1.58)0.032.13 (0.69 to 6.60)46.5 - Drug-related serious adverse events during treatment51.97 (0.86 to 4.50)0.031.99 (0.75 to 5.28)0.022.53 (0.25 to 25.9)52.0 - Any hepatotoxicity during treatment41.31 (0.76 to 2.27)67.431.13 (0.61 to 2.08)73.812.88 (0.97 to 8.58)– - Treatment interruption/discontinuation41.89 (1.09 to 3.28)0.041.89 (1.09 to 3.28)0.00––Figures in parentheses are 95% confidence intervals; I2 Higgins inconsistency index Summary outcomes stratified by risk of bias of included studies Figures in parentheses are 95% confidence intervals; I2 Higgins inconsistency index Table 4Summary outcomes and clinical implications while comparing high-dose versus standard-dose rifampicin regimensNo. of studiesSummary risk ratioI2, %Summary risk difference, %NNTB or NNTHPrimary efficacy outcomes- Culture positivity at 2 months80.74 (0.62 to 0.89)15.55.64 (1.74 to 9.53)NNTB 18 (NNTB 10 to NNTB 58)- Culture positivity at 6 months50.52 (0.23 to 1.15)16.53.02 (0.48 to 5.55)NNTB 33 (NNTB 18 to NNTB 207)Primary safety outcome- Severe hepatotoxicity during treatment72.06 (1.15 to 3.68)0.0-2.20 (-3.69 to -0.71)NNTH 45 (NNTH 27 to NNTH 141)Secondary efficacy outcomes- Culture positivity at 1 month30.82 (0.66 to 1.02)89.214.31 (-0.65 to 29.26)NNTB 7 (NNTB 3 to ∞ to NNTH 154)- Culture positivity at 3 months30.77 (0.49 to 1.22)51.65.09 (0.10 to 10.07)NNTB 20 (NNTB 10 to NNTB 1020)- Culture positivity at 4 months30.80 (0.49 to 1.29)0.00.10 (-1.79 to 1.99)NNTB 982 (NNTB 50 to ∞ to NNTH 56)- Death during treatment80.77 (0.36 to 1.67)18.40.51 (-0.52 to 1.53)NNTB 197 (NNTB 65 to ∞ to NNTH 193)- Treatment failure50.86 (0.44 to 1.69)1.00.14 (-1.88 to 2.15)NNTB 736 (NNTB 47 to ∞ to NNTH 53)- Recurrence after treatment61.21 (0.39 to 3.74)52.9-0.44 (-2.37 to 1.49)NNTH 228 (NNTH 42 to ∞ to NNTB 67)- Loss to follow-up50.90 (0.48 to 1.69)0.00.44 (-0.68 to 1.57)NNTB 226 (NNTB 64 to ∞ to NNTH 147)Secondary safety outcomes- Any adverse event during treatment81.06 (0.94 to 1.20)55.5-4.14 (-11.07 to 2.79)NNTH 24 (NNTH 9 to ∞ to NNTB 36)- Serious adverse events during treatment101.26 (0.92 to 1.72)9.0-1.63 (-3.77 to 0.51)NNTH 61 (NNTH 27 to ∞ to NNTB 195)- Drug-related serious adverse events during treatment51.97 (0.86 to 4.50)0.0-2.69 (-4.81 to -0.56)NNTH 37 (NNTH 21 to NNTH 178)- Any hepatotoxicity during treatment41.31 (0.76 to 2.27)67.4-7.92 (-20.50 to 4.67)NNTH 13 (NNTH 5 to ∞ to NNTB 21)- Treatment interruption/discontinuation41.89 (1.09 to 3.28)0.0-1.98 (-3.31 to -0.66)NNTH 50 (NNTH 30 to NNTH 152)Figures in parentheses are 95% confidence intervals. I2 Higgins inconsistency index, NNTB Number needed to treat for an additional beneficial outcome, NNTH Number needed to treat for an additional harmful outcome Summary outcomes and clinical implications while comparing high-dose versus standard-dose rifampicin regimens Figures in parentheses are 95% confidence intervals. I2 Higgins inconsistency index, NNTB Number needed to treat for an additional beneficial outcome, NNTH Number needed to treat for an additional harmful outcome Table 5Summary of findings for the primary outcomesOutcomeResidual culture positivity at end of intensive phaseResidual culture positivity at end of treatmentSevere hepatotoxicity during treatmentNumber of studies8 RCTs5 RCTs7 RCTsNumber of participants212910881808Events in high-dose rifampicin treatment arms231/1334 (17.3%)18/645 (2.8%)54/1160 (4.7%)Events in standard dose rifampicin treatment arms218/795 (27.4%)29/443 (6.5%)14/648 (2.2%)Summary risk ratio (95% CI)0.74 (0.62–0.89)0.52 (0.23–1.15)2.06 (1.15–3.68)Assumed risk with standard dose rifampicin274 per 100065 per 100022 per 1000Corresponding risk with high-dose rifampicin (95% CI)203 per 1000 (169 to 243)34 per 1000 (15 to 75)44 per 1000 (25 to 79)Serious risk of biasNot presentNot presentNot presentIndirectnessNoneNoneNoneInconsistencyNoneNoneNoneImprecisionNoneLikelyLikelyLikelihood of publication biasNot assessedNot assessedNot assessedCertainty of the evidenceHigh, ⊕⊕⊕⊕Moderate,* ⊕⊕⊕○Moderate,* ⊕⊕⊕○CI=Confidence interval, RCT=Randomised controlled trial. *Downgraded one level for imprecision, as summary estimates were based on few events and had wide and variable 95% confidence/credible intervals Summary of findings for the primary outcomes CI=Confidence interval, RCT=Randomised controlled trial. *Downgraded one level for imprecision, as summary estimates were based on few events and had wide and variable 95% confidence/credible intervals After 6 months of ATT, there was no significant difference in residual sputum culture positivity between the two groups (summary RR: 0.52, 95% CI: 0.23–1.15, I2 = 16.5%), based on data from four studies reporting on 645 and 443 patients administered high-dose and standard-dose rifampicin, respectively.[3637384244] A Bayesian approach with vague prior resulted in slighter lower summary risk ratio estimate with wider 95% credible intervals (95% CrI) [Figure 2]. Individually, all studies favoured high-dose rifampicin, although results were significant only for a single study [Figure 3].[44] We calculated that 33 patients will need to receive high-dose rifampicin-based ATT to reduce one positive sputum culture at end of treatment [Table 4]. The overall quality of evidence was rated as moderate. The evidence was downgraded for imprecision due to few events in most studies, and wide and variable CIs/CrIs [Table 5]. Seven publications reported on severe hepatotoxicity during treatment with high-dose (1160 subjects) or standard-dose rifampicin (648 subjects).[34353941424344] Most studies (five, 71.4%) used the DAIDS grading to assign severity status.[3439414244] All studies, except one, documented a higher RR in the high-dose group, although none of these differences were statistically significant [Figure 4]. Overall, there was a significantly higher incidence of severe hepatotoxicity among those receiving high-dose rifampicin (summary RR: 2.06, 95% CI: 1.15–3.68) with negligible heterogeneity between studies (I2 = 0.0%). This translates to an additional severe hepatotoxicity episode for every 45 patients receiving high-dose rifampicin-based ATT [Table 4]. However, the summary risk estimates were not significant using either the Bayesian or Hartung-Knapp-Sidik-Jonkman procedures for meta-analysis [Figure 2]. Summary estimates remained similar even after excluding two studies with some concern of bias [Table S2 of online supplement].[3543] The overall quality of evidence was rated as moderate, suggesting a fair degree of confidence in the likely negative impact of high-dose rifampicin. The evidence was downgraded for imprecision due to few events in most studies, and wide and variable CIs/CrIs [Table 5]. Adverse events during treatment among study subjects receiving high-dose or standard-dose rifampicin Secondary efficacy outcomes Three studies each reported on residual sputum culture positivity at 1, 3, and 4 months of ATT [Figure 3]. At all three time-points, summary RR estimates were not statistically significant [Table 4]. One recent study significantly favoured high-dose rifampicin at 1 and 3 months of ATT,[44] while another significantly favoured high-dose rifampicin at 1 month of ATT [Figure 3].[35] There was also significant heterogeneity between studies (I2 = 89.2%) included for analysis at the end of 1 month treatment. Studies evaluating culture after 3 months ATT showed moderate heterogeneity (I2 = 51.6%), and the summary RD estimates suggested that 20 patients treated with high-dose rifampicin could lead to one less sputum culture positivity [Table 4]. Summary estimates remained similar even after excluding a single study with some concern of bias, with no improvement in heterogeneity [Table S2 of online supplement].[35] Three studies described the time to sputum culture conversion at end of 2 or 6 months of ATT.[353744] However, all compared two different doses of high-dose rifampicin against the standard dose, and it was not possible to pool this information to a single high-dose category for analysis. Another study mentioned no significant difference in overall time to culture conversion among subjects receiving 30 mg/kg/day or standard-dose rifampicin, but did not provide additional details.[45] Ten studies reported data on death during treatment with high-dose (1599 subjects) or standard-dose rifampicin (1143 subjects).[35363738404142434445] Six clinical trials studied tuberculosis recurrence during post-treatment follow-up among 901 and 717 subjects receiving high-dose and standard-dose rifampicin, respectively.[373840414445] Five publications assessed treatment failure in 616 and 535 study participants receiving high-dose and standard-dose rifampicin, respectively.[3638404245] Six studies provided information on loss to follow-up among 1040 and 743 subjects receiving high-dose and standard-dose rifampicin, respectively.[383940414245] On meta-analysis, none of these outcomes were significantly different between high-dose and standard-dose rifampicin groups [Figure 5]. The analysis on disease recurrence indicated moderate heterogeneity between studies (I2 = 52.9%); other comparisons showed low between-study heterogeneity [Table 4]. The summary risk of death during treatment was somewhat higher from two studies with some concern of bias [Table S2 of online supplement].[3543] Miscellaneous efficacy outcomes among study subjects receiving high-dose or standard-dose rifampicin Secondary safety outcomes Eight studies compared incidence of any adverse event during ATT among 1201 and 768 subjects receiving high-dose and standard-dose rifampicin, respectively.[3738394142434445] However, the pooled differences were not significant (summary RR: 1.06, 95% CI: 0.94–1.20) [Figure 4]. Summary estimates remained similar even after excluding two studies with some concern of bias [Table S2 of online supplement].[4345] Ten studies reported data on serious adverse events during treatment with high-dose (1789 subjects) or standard-dose rifampicin (1233 subjects), which was similar between the two groups [Figure 4].[34353738404142434445] There was low between-study heterogeneity (I2 = 9.0%). Summary estimates remained similar even after excluding the three studies with some concern of bias, with further improvement in heterogeneity [Table S2 of online supplement].[354345] There was no indication of potential publication bias for this analysis, as supported by a largely symmetric funnel plot (t = 1.54, P = 0.16) and non-significant result on rank correlation test (Kendall’s tau = 0.24, P = 0.38). Five studies (total 526 and 424 participants on high-dose and standard-dose rifampicin, respectively) also provided information on drug-related serious adverse events during treatment.[3537384243] There was negligible between-study heterogeneity and no significant difference between the two rifampicin doses [Figure 4]. The summary risk difference suggested that treatment of 37 patients with high-dose rifampicin would cause an additional drug-related serious adverse event [Table 4]. Hepatotoxicity (any grade) was also similar between the two doses based on data from 735 and 573 subjects receiving high-dose and standard-dose rifampicin, respectively, in four clinical trials [Figure 4].[34384043] Summary estimates remained similar for both analyses even after excluding studies with some concern of bias [Table S2 of online supplement].[3543] Treatment with high-dose rifampicin was associated with a significantly higher rate of ATT interruption or discontinuation among 969 and 571 study participants getting high-dose and standard-dose rifampicin, respectively, in four studies (summary RR: 1.89, 95% CI: 1.09–3.28).[37384144] This implies an additional such occurrence for every 50 patients getting ATT with high rifampicin dose [Table 4]. Effect of rifampicin dose The studies included in this review compared multiple higher rifampicin doses with standard-dose ATT across a range of outcomes. We arbitrarily categorised data from these RCTs into three daily rifampicin dose subsets: 15 mg/kg (N = 4),[34353638] 20–25 mg/kg (N = 9),[343536373839404144] and 30 mg/kg or greater (N = 6),[374142434445] for a subgroup analysis. While most earlier studies evaluated the lower end of this dosage spectrum, more recent work has focussed on relatively higher doses [Table 3]. Results from individual studies for each dose subset are detailed in Tables S3 and S4 of online supplement. Table S3Comparison of efficacy outcomes with various high-dose rifampicin regimensParameterHigh dose rifampicinStudyn/N High dosen/N Standard doseRelative risk (95% CI)Culture positivity at 1m15 mg/kg/dayAarnoutse 201744/4846/490.98 (0.87–1.09)TOTAL0.98 (0.87–1.09)20–25 mg/kg/dayAarnoutse 201732/4846/490.71 (0.57–0.88)Boeree 2017103/119156/1811.00 (0.92–1.10)Bhavani 202457/11281/1050.66 (0.53–0.81)TOTAL0.79 (0.58–1.06)≥30 mg/kg/dayBoeree 201750/63156/1810.92 (0.80–1.06)Bhavani 202452/10681/1050.64 (0.51–0.79)TOTAL0.78 (0.55–1.10)Culture positivity at 2m15 mg/kg/dayJindani 201614/8023/920.70 (0.39–1.27)Aarnoutse 201724/4825/490.98 (0.66–1.45)Velásquez 20185/583/561.61 (0.40–6.42)TOTAL0.91 (0.66–1.26)20–25 mg/kg/dayJindani 201615/9123/920.66 (0.37–1.18)Aarnoutse 201719/4825/490.78 (0.50–1.21)Boeree 201758/11999/1810.89 (0.71–1.12)Velásquez 20187/603/562.18 (0.59–8.01)Atwine 20208/646/330.69 (0.26–1.82)Jindani 202313/21226/2120.50 (0.26–0.95)Bhavani 20249/11219/1050.44 (0.21–0.94)TOTAL0.74 (0.58–0.95)≥30 mg/kg/dayBoeree 201724/6399/1810.70 (0.49–0.98)Jindani 202318/21226/2120.69 (0.39–1.22)Sekaggya-Wiltshire 20238/6117/670.52 (0.24–1.11)Bhavani 20249/10619/1050.47 (0.22–0.99)TOTAL0.64 (0.50–0.83)Culture positivity at 3m15 mg/kg/dayAarnoutse 201718/4814/491.31 (0.74–2.33)TOTAL1.31 (0.74–2.33)20–25 mg/kg/dayAarnoutse 201712/4814/490.88 (0.45–1.69)Boeree 201725/11944/1810.86 (0.56–1.33)Bhavani 20245/11212/1050.39 (0.14–1.07)TOTAL0.79 (0.55–1.13)≥30 mg/kg/dayBoeree 201711/6344/1810.72 (0.40–1.30)Bhavani 20245/10612/1050.41 (0.15–1.13)TOTAL0.62 (0.37–1.04)Culture positivity at 4m15 mg/kg/dayAtwine 20173/323/341.06 (0.23–4.89)TOTAL1.06 (0.23–4.89)20–25 mg/kg/dayAtwine 20174/343/341.33 (0.32–5.51)Boeree 201710/11926/1810.59 (0.29–1.17)Jindani 20234/2123/2121.33 (0.30–5.89)TOTAL0.76 (0.43–1.34)≥30 mg/kg/dayBoeree 20178/6326/1810.88 (0.42–1.85)Jindani 20233/2123/2121.00 (0.20–4.90)TOTAL0.90 (0.46–1.77)Culture positivity at 6m15 mg/kg/dayVelásquez 20181/583/560.32 (0.03–3.00)TOTAL0.32 (0.03–3.00)20–25 mg/kg/dayBoeree 20178/11920/1810.61 (0.28–1.34)Velásquez 20181/603/560.31 (0.03–2.90)Bhavani 20240/1124/1050.10 (0.01–1.91)TOTAL0.51 (0.25–1.05)≥30 mg/kg/dayBoeree 20177/6320/1811.01 (0.45–2.26)Sekaggya-Wiltshire 20231/612/670.55 (0.05–5.91)Bhavani 20240/1064/1050.11 (0.01–2.02)TOTAL0.76 (0.31–1.86)Time to culture conversion at 2m*20–25 mg/kg/dayBoeree 2017N=63N=1231.47 (0.90–2.40)Bhavani 2024N=112N=1051.76 (1.30–2.37)TotalN=175N=2281.68 (1.30–2.16)≥30 mg/kg/dayBoeree 2017N=63N=1231.73 (1.07–2.82)Bhavani 2024N=106N=1051.65 (1.22–2.25)TotalN=169N=2281.67 (1.29–2.17)Time to culture conversion at 6m*15 mg/kg/dayAarnoutse 2017N=48N=490.91 (0.55–1.52)TotalN=48N=490.91 (0.55–1.52)20–25 mg/kg/dayAarnoutse 2017N=48N=491.34 (0.82–2.18)Bhavani 2024N=112N=1051.81 (1.37–2.41)TotalN=160N=1541.67 (1.28–2.17)≥30 mg/kg/dayBhavani 2024N=106N=1051.74 (1.30–2.32)TotalN=106N=1051.74 (1.30–2.32)Death during treatment15 mg/kg/dayAarnoutse 20171/481/491.02 (0.07–15.86)TOTAL1.02 (0.07–15.86)20–25 mg/kg/dayAarnoutse 20171/481/491.02 (0.07–15.86)Atwine 20202/641/331.03 (0.10–10.96)Maug 20205/34711/3430.45 (0.16–1.28)Jindani 20234/2123/2121.33 (0.30–5.89)Bhavani 20241/1122/1050.47 (0.04–5.09)TOTAL0.68 (0.32–1.41)≥30 mg/kg/dayBoeree 20171/630/1818.53 (0.35–206.78)Jindani 20230/2123/2120.14 (0.01–2.75)Sekaggya-Wiltshire 20234/610/678.92 (0.48–165.28)Souleymane 20234/621/654.19 (0.48–36.49)Bhavani 20240/1062/1050.20 (0.01–4.08)TOTAL1.67 (0.29–9.56)Recurrence after treatment15 mg/kg/dayVelásquez 20180/582/560.19 (0.01–3.94)TOTAL0.19 (0.01–3.94)20–25 mg/kg/dayBoeree 20173/1193/1811.52 (0.31–7.41)Velásquez 20180/602/560.19 (0.01–3.81)Maug 20203/3471/3432.97 (0.31–28.37)Jindani 20239/2122/2124.50 (0.98–20.58)Bhavani 20240/1124/1050.10 (0.01–1.91)TOTAL1.23 (0.34–4.46)≥30 mg/kg/dayBoeree 20170/633/1810.41 (0.02–7.76)Jindani 20239/2122/2124.50 (0.98–20.58)Bhavani 20241/1064/1050.25 (0.03–2.18)Kwak 20242/260/326.11 (0.31–121.95)TOTAL1.38 (0.26–7.25)Treatment failure15 mg/kg/dayVelásquez 20181/583/560.32 (0.03–3.00)TOTAL0.32 (0.03–3.00)20–25 mg/kg/dayVelásquez 20181/603/560.31 (0.03–2.90)Atwine 20200/641/330.17 (0.01–4.17)Maug 202012/3478/3431.48 (0.61–3.58)TOTAL0.76 (0.21–2.74)≥30 mg/kg/daySekaggya-Wiltshire 20231/612/670.55 (0.05–5.91)Kwak 20242/264/320.62 (0.12–3.10)TOTAL0.59 (0.16–2.26)Loss to follow up15 mg/kg/dayVelásquez 20181/582/560.48 (0.05–5.18)TOTAL0.48 (0.05–5.18)20-25 mg/kg/dayVelásquez 20181/602/560.47 (0.04–5.01)Maug 20208/3478/3430.99 (0.38–2.60)Jindani 20230/2122/2120.20 (0.01–4.14)TOTAL0.79 (0.33–1.86)≥30 mg/kg/dayJindani 20231/2122/2120.50 (0.05–5.47)Sekaggya-Wiltshire 20236/616/671.10 (0.37–3.23)TOTAL0.96 (0.36–2.57)*Results presented as hazard ratio Comparison of efficacy outcomes with various high-dose rifampicin regimens *Results presented as hazard ratio Table S4Comparison of adverse effects with various high-dose rifampicin regimensParameterHigh dose rifampicinStudyn/N High dosen/N Standard doseRelative risk (95% CI)Any adverse event15 mg/kg/dayVelásquez 201831/6026/601.19 (0.82–1.74)TOTAL1.19 (0.82–1.74)20–25 mg/kg/dayBoeree 201791/120141/1820.98 (0.86–1.11)Velásquez 201823/6026/600.88 (0.57–1.36)Atwine 202012/656/331.02 (0.42–2.46)Jindani 2023109/223120/2240.91 (0.76–1.09)Bhavani 202480/11461/1091.25 (1.02–1.54)TOTAL1.01 (0.89–1.15)≥30 mg/kg/dayBoeree 201753/63141/1821.09 (0.95–1.24)Jindani 2023115/225120/2240.95 (0.80–1.14)Sekaggya-Wiltshire 202336/6139/671.01 (0.76–1.36)Souleymane 202318/6211/651.72 (0.88–3.33)Bhavani 202496/11061/1091.56 (1.30–1.87)Kwak 202425/2835/380.97 (0.83–1.14)TOTAL1.13 (0.95–1.34)Serious adverse event15 mg/kg/dayJindani 20162/1001/1002.00 (0.18–21.71)Aarnoutse 20174/502/502.00 (0.38–10.43)Velásquez 20181/602/600.50 (0.05–5.37)TOTAL1.42 (0.44–4.62)20–25 mg/kg/dayJindani 20164/1001/1004.00 (0.46–35.16)Aarnoutse 20175/502/502.50 (0.51–12.29)Boeree 201716/12020/1821.21 (0.66–2.25)Velásquez 20185/602/602.50 (0.50–12.39)Maug 20209/35315/3480.59 (0.26–1.33)Jindani 202310/2239/2241.12 (0.46–2.69)Bhavani 202411/11410/1091.05 (0.47–2.38)TOTAL1.13 (0.79–1.61)≥30 mg/kg/dayBoeree 20179/6320/1821.30 (0.62–2.70)Jindani 202310/2259/2241.11 (0.46–2.67)Sekaggya-Wiltshire 202310/616/671.83 (0.71–4.74)Souleymane 202311/622/655.77 (1.33–24.98)Bhavani 202413/11010/1091.29 (0.59–2.81)Kwak 20243/283/381.36 (0.30–6.23)TOTAL1.46 (1.00–2.15)Drug-related serious adverse event15 mg/kg/dayAarnoutse 20171/502/500.50 (0.05–5.34)Velásquez 20181/601/601.00 (0.06–15.62)TOTAL0.67 (0.11–4.04)20–25 mg/kg/dayAarnoutse 20173/502/501.50 (0.26–8.60)Boeree 20174/1201/1826.07 (0.69–53.62)Velásquez 20182/601/602.00 (0.19–21.47)TOTAL2.43 (0.75–7.92)≥30 mg/kg/dayBoeree 20173/631/1828.67 (0.92–81.81)Sekaggya-Wiltshire 20235/614/671.37 (0.39–4.88)Souleymane 20235/620/6511.52 (0.65–204.14)TOTAL3.49 (0.83–14.70)Hepatotoxicity15 mg/kg/dayJindani 201661/10032/1001.91 (1.38–2.64)Velásquez 201814/6016/600.88 (0.47–1.63)TOTAL1.35 (0.63–2.89)20–25 mg/kg/dayJindani 201654/10032/1001.69 (1.20–2.37)Velásquez 201814/6016/600.88 (0.47–1.63)Maug 20208/35311/3480.72 (0.29–1.76)TOTAL1.12 (0.64–1.97)≥30 mg/kg/daySouleymane 202311/624/652.88 (0.97–8.58)TOTAL2.88 (0.97–8.58)Severe hepatotoxicity15 mg/kg/dayJindani 20162/1001/1002.00 (0.18–21.71)Aarnoutse 20170/501/500.33 (0.01–7.99)TOTAL1.05 (0.16–7.06)20–25 mg/kg/dayJindani 20164/1001/1004.00 (0.46–35.16)Aarnoutse 20172/501/502.00 (0.19–21.36)Atwine 20204/652/331.02 (0.20–5.26)Jindani 20237/2233/2242.34 (0.61–8.95)Bhavani 20247/1144/1091.67 (0.50–5.56)TOTAL1.87 (0.92–3.79)≥30 mg/kg/dayJindani 20237/2253/2242.32 (0.61–8.87)Sekaggya-Wiltshire 20235/613/671.83 (0.46–7.34)Souleymane 20234/620/659.43 (0.52–171.57)Bhavani 202412/1104/1092.97 (0.99–8.93)TOTAL2.58 (1.28–5.22)Treatment interruption or discontinuation15 mg/kg/dayVelásquez 201811/608/601.38 (0.60–3.18)TOTAL1.38 (0.60–3.18)20–25 mg/kg/dayBoeree 20173/1202/1822.28 (0.39–13.41)Velásquez 201810/608/601.25 (0.53–2.95)Jindani 20232/2231/2242.01 (0.18–22.00)Bhavani 20247/1144/1091.67 (0.50–5.56)TOTAL1.51 (0.81–2.82)≥30 mg/kg/dayBoeree 20173/632/1824.33 (0.74–25.34)Jindani 20237/2251/2246.97 (0.86–56.18)Bhavani 202412/1104/1092.97 (0.99–8.93)TOTAL3.66 (1.56–8.57) Comparison of adverse effects with various high-dose rifampicin regimens In general, the culture positivity rates at 1–4 months of treatment were largely similar between standard-dose and 15 mg/kg/day rifampicin [Figure S2 of online supplement]. Culture positivity at 2 months was significantly lower with both 20–25 mg/kg/day (RR: 0.74, 95% CI: 0.58–0.95; NNTB 18, 95% CI: 10–72) and ≥ 30 mg/kg/day (RR: 0.64, 95% CI: 0.50–0.83; NNTB 12, 95% CI: 7–37) rifampicin doses. There appeared to a dose-dependent gradient for greater culture conversion rates at 1–3 months of ATT with progressively higher rifampicin doses [Figure S2 of online supplement]. Other efficacy outcomes (mortality, recurrence after ATT, treatment failure, and loss to follow-up) were broadly similar for all three high-dose rifampicin subsets, with no discernible gradient across dose categories [Figure S3 of online supplement]. The risk for drug-related serious adverse events, severe hepatotoxicity, and treatment interruption appeared to progressively worsen with increasing rifampicin doses [Figure S4 of online supplement]. Severe hepatoxicity was significantly higher only at ≥30 mg/kg/day rifampicin doses (RR: 2.58, 95% CI: 1.28–5.22; NNTH 31, 95% CI: 17–136). Similarly, treatment interruption was significantly greater with daily rifampicin doses ≥30 mg/kg (RR: 3.66, 95% CI: 1.56–8.57; NNTH 31, 95% CI: 19–89), but not with other high-dose rifampicin subsets. Meta-regression analysis did not show any statistically significant change in the primary outcomes with progressively increasing rifampicin doses [Figure S5 of online supplement]. DISCUSSION We identified 12 RCTs published since 2016 that evaluated several high-dose rifampicin-based ATT regimens given daily for pulmonary tuberculosis. Overall, administration of high-dose rifampicin resulted in significantly lower residual culture positivity after 2 months of ATT (RR: 0.74, 95% CI: 0.62–0.89) but was associated with a significantly higher incidence of severe hepatotoxicity (RR: 2.06, 95% CI: 1.15–3.68) and treatment interruption/discontinuation (RR: 1.89, 95% CI: 1.09–3.28). On subgroup analysis, these outcomes were particularly evident with rifampicin doses of ≥30 mg/kg/day. A major strength of our systematic review is the inclusion of recent good-quality RCTs that reflect the current global standards of tuberculosis care, including daily administration of ATT, use of both rifampicin and isoniazid throughout treatment duration, and additional administration of ethambutol and/or pyrazinamide during the initial intensive phase. We also included clinical trials with clearly higher daily rifampicin doses (≥15 mg/kg/day) and excluded studies with intermittent dosing regimens or those with borderline rifampicin doses (e.g. 12 mg/kg/day). This improves the relevance and applicability of our observations in the currently prevalent clinical context. In general, our subgroup analyses found that outcomes after administration of 15 mg/kg/day of rifampicin were not significantly different across any efficacy parameter (even though the number of studies is rather small). We therefore believe that further research should be concentrated on higher rifampicin dosing. We used DerSimonian–Laird random effects modelling as our primary meta-analytic procedure. We also checked the robustness of summary risk-ratio estimates for our primary outcomes through alternative frequentist and Bayesian meta-analytic techniques. The Hartung-Knapp-Sidik-Jonkman method, based on a t-distribution rather than normal distribution, may perform better when there is heterogeneity and the number of studies in the meta-analysis is small. The significant reduction in sputum culture positivity rates at the end of intensive phase of ATT was observed across all these models and offers promising clinical implications for treating pulmonary tuberculosis. Quicker bacillary clearance during ATT has an enormous epidemiological significance in reducing disease transmission. Much of this improvement was significant only for the higher rifampicin doses (≥20 mg/kg/day). Previous data also suggest that a far greater proportion of patients receiving higher rifampicin doses achieve serum drug concentrations in the therapeutic range.[1146] Although increasing rifampicin dose can drive quicker and more effective mycobacterial clearance, potentially shortening ATT duration, it may also increase the risk and severity of adverse effects. While there was some evidence (of moderate certainty) for greater incidence of severe hepatoxicity with higher rifampicin doses, this was probably important only for daily doses of 30 mg/kg or higher. Overall, the benefit from early sputum conversion tended to outweigh the potential harm from severe hepatotoxicity (NNTB 18 versus NNTH 45, Table 4). Such data can aid policymakers and clinicians once they further explore the feasibility, utility, and risk of higher rifampicin doses during implementation in programmatic settings. Perhaps it might be prudent to gather additional experience with 20–25 mg/kg/day dosing before moving on to still higher doses. Whether the higher sputum conversion rates during intensive phase of high-dose rifampicin-based ATT will permit shortening of treatment duration from the standard 6-month regimen is still not clear.[6] Two recent studies included in our analysis administered ATT for a shorter duration to their experimental groups; both demonstrated comparable safety but lack of clear non-inferiority for their primary composite efficacy outcomes.[4145] However, in one study, culture-confirmed recurrence rates were 4.5 times higher with 4 months of high-dose rifampicin as compared to the standard regimen [Figure 4].[41] This is an area that requires further exploration. Our findings are at a slight variance with previous systematic reviews evaluating the safety and efficacy of high-dose rifampicin for treating tuberculosis.[474849] One old systematic review reported better sputum culture conversion rates with higher rifampicin doses in pulmonary tuberculosis.[47] However, all included studies used ATT regimens with intermittent rifampicin administration without pyrazinamide. Another meta-analysis also showed a higher sputum culture conversion rate at end of intensive phase of therapy but no difference in treatment failure, severe hepatotoxicity, or other adverse effects.[48] However, this analysis also included studies offering higher rifampicin doses for only 2 weeks, and those with a marginally higher dose of 12 mg/kg/day, which currently is the upper limit of standard daily rifampicin dosing in adults.[50] A more recent meta-analysis looked at high-dose rifamycins (rifampicin, rifabutin, or rifapentine) for treating pulmonary and extrapulmonary tuberculosis.[49] It found no significant difference in bacteriological conversion rates or adverse events between pulmonary or meningeal tuberculosis patients receiving high-dose versus standard-dose rifamycin-based ATT.[49] However, a separate detailed analysis of rifampicin-based ATT for pulmonary tuberculosis was not available, and studies with ATT regimens not providing rifampicin throughout treatment were also evaluated. None of these meta-analyses systematically evaluated the impact of progressively increasing rifampicin dose on treatment efficacy or incidence of adverse effects. Our analysis is not without limitations. The 12 included studies used very different ATT regimens, in both the standard-dose and high-dose rifampicin treatment arms. Although there was no substantial statistical heterogeneity for several outcomes studied [Table 4], the number and nature of drugs used, daily rifampicin doses, period of high-dose rifampicin administration, and total treatment duration were all highly variable [Table 3]. The various RCTs also looked at different sets of outcomes, which were sometimes measured or expressed using somewhat different tools. The number of studies in each dose subgroup was also limited. These variations preclude generalisability of any outcome to any higher rifampicin dose or regimen. The sample size of the included studies ranged from below 100[3945] to more than 600.[4041] We could not study publication bias for most outcomes due to a small number of studies. However, since all our included studies were reported in English language, we may have missed some small clinical trials published in foreign languages. In conclusion, we found a higher rate of microbiological clearance at 2 months with the use of high-dose rifampicin regimens, with some increase in severe hepatoxicity events, particularly at rifampicin doses of 30 mg/kg/day or higher. There is need to further assess high-dose rifampicin regimens in routine programmatic settings, particularly with daily rifampicin doses of 20–25 mg/kg/day, and to explore specific patient populations that might benefit more from such treatment schedules. Author contributions ANA: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Data Curation, Writing – Original Draft, Writing – Review & Editing, Visualization, Supervision. RA: Methodology, Validation, Formal analysis, Investigation, Data Curation, Writing – Original Draft, Writing – Review & Editing, Visualization, Supervision. SD: Methodology, Validation, Formal analysis, Investigation, Data Curation, Writing – Original Draft, Writing – Review & Editing, Visualization. PKB: Formal analysis, Writing – Original Draft, Writing – Review & Editing. KTP: Formal analysis, Writing – Original Draft, Writing – Review & Editing. IPS: Formal analysis, Writing – Original Draft, Writing – Review & Editing. VM: Formal analysis, Writing – Original Draft, Writing – Review & Editing. Data sharing The data supporting the findings of this study are already available in public domain within the publications included for the systematic review. Any additional information will be made available by the corresponding author upon reasonable request. Conflicts of interest There are no conflicts of interest. Risk of bias assessment using Cochrane's Risk of Bias Tool version 2.0 Residual sputum mycobacterial culture positivity during treatment Miscellaneous efficacy outcomes Adverse events Meta-regression analysis to assess impact of progressively increasing rifampicin doses on the primary outcomes