Secondary Logo

Journal Logo

Original Studies

Better Completion of Pediatric Latent Tuberculosis Treatment Using 4 Months of Rifampin in a US-based Tuberculosis Clinic

Gaensbauer, James MD, MScPH*†; Aiona, Kaylynn MPH*; Haas, Michelle MD*‡; Reves, Randall MD*‡; Young, Janine MD*§¶; Belknap, Robert MD*‡

Author Information
The Pediatric Infectious Disease Journal: March 2018 - Volume 37 - Issue 3 - p 224-228
doi: 10.1097/INF.0000000000001721


Antimicrobial treatment of children infected with Mycobacterium tuberculosis is an essential part of reducing the burden of tuberculosis (TB) disease. In the absence of early treatment, children—particularly the very young—who are infected with M. tuberculosis are at high risk of transition to active disease with significant morbidity and mortality. Additionally, children with latent TB infection (LTBI) may only become symptomatic and contagious many decades in the future, and thus represent a significant barrier to the global challenge of long-term TB elimination. While effective treatments for pediatric LTBI exist, there are significant barriers to effective and universal implementation. In the United States, the Centers for Disease Control and Prevention and the American Academy of Pediatrics preferentially recommend a treatment regimen of 9-month isoniazid (9H) therapy.1,2 However, this regimen can be difficult to complete, and the long duration adds significant resource requirements for TB control programs.3 In low- and middle-income countries, recent estimates from the World Health Organization (WHO) suggest that only 7% of children <5 years of age with LTBI receive appropriate treatment, and the resource requirements for the World Health Organization–recommended 6-month isoniazid (INH) treatment are likely a major contributor to this deficiency.4 Shorter course regimens have the potential to increase both treatment completion rates as well as allow for broader inclusion of pediatric LTBI treatment in TB control programs, including in low- and middle-income countries. One such potential regimen is a 4-month treatment with rifampin (4R), which is currently recommended in the United States as an alternative treatment for pediatric patients with INH resistance or inability to take INH.2 Evidence for efficacy of 4R is based primarily on data derived from small trials in high-risk adults, animal models and observational studies, but in this literature, 4R appears to be noninferior to 9-month INH regimens.5–7 Furthermore in adults, 4R is well tolerated and associated with improved adherence relative to longer courses.8–10 Little data on 4R in pediatric LTBI exist, though an observational study in a US TB control program that included 80 children with LTBI treated with 4R showed better adherence, high tolerance and equivalent effectiveness compared with a cohort of 9H-treated children.11 Better adherence with shorter LTBI treatments in pediatric patients has also been demonstrated in Greece, using a regimens of 3 or 4 months of daily isoniazid and rifampin.12

Given the direct relationship between adherence and effectiveness, the potentially higher rates of completion of 4R make this an attractive strategy for TB control programs. In 2012, the Denver Metro Tuberculosis Clinic (DMTBC), which provides pediatric TB care to 7 urban counties in Colorado and consultation to the Colorado Department of Public Health and Environment on all pediatric cases statewide, introduced 4R as the preferred (though not exclusive) regimen for pediatric LTBI. To assess the impact of this approach, we retrospectively compared treatment completion rates, side effects and efficacy between all pediatric patients treated for LTBI with 4R and those treated with 9H between January 1, 2006, and December 13, 2016, and assessed the influence of demographic characteristics on successful completion of the 2 regimens.


Study Population

Data were extracted from an electronic medical record database maintained by the DMTBC, which captures demographic and clinical information on all patients evaluated in the clinic. All patients <18 years of age with a diagnosis of LTBI were identified between January 1, 2006, and December 31, 2015. Demographic data included age, gender, race/ethnicity, primary language spoken in home, birth country and time in United States for foreign-born patients. The birth country of parents of US-born patients was not captured. Baseline clinical data captured included bacille Calmette-Guerin vaccination (by parental report or vaccine card review), results of interferon-gamma releasing assay (IGRA) or tuberculin skin tests (TST) and contact status to an individual with an active case of TB. Completion status and reasons for failed completion were assessed for initial LTBI treatments (defined as the first recommended treatment at the time of LTBI diagnosis). Routine laboratory monitoring is not performed on pediatric patients treated for LTBI and is not captured in the database. However, treatment noncompletion secondary to abnormal laboratory testing in selected individuals (eg, elevated liver enzymes in a symptomatic patient) is captured in the database.

Treatment Regimens

Diagnosis of LTBI at the DMTBC requires a positive TST (with size of induration based on current US guidelines) or IGRA and normal symptom review and chest radiograph.2 All decisions related to treatment regimen are made on site by DMTBC nurses, with physician consultation when required and following the general clinic procedures; before 2012, 9H was preferred, but 4R was acceptable for drug interactions or nurse concern for adherence with the longer regimen. After 2012, 4R was preferred, but 9H was still a secondary option, particularly when concerns for drug interactions existed. Cost was not a decision factor as all medications are provided at DMTBC through a Public Health grant. In general, the suggested regimen by nurses is selected by patients; in-depth discussion of the relative strength of the literature underlying each regimen is not undertaken. Factors that inform the decision (concerns for adherence, medication interactions, patient preference, etc.) are not captured in the database. It is the DMTBC policy to administer daily LTBI regimens by directly observed preventive therapy to children under 5 years of age, and in these cases, it is given by clinic personnel 5 days/week (Monday to Friday; typically at the patients home), and not on weekends. In older children, it is not administered by directly observed preventive therapy, but parents are instructed and monitored to administer medication 7 days/week for consistency. Thus the primary comparison groups include both 5 and 7 days/week regimens, depending on age. Regimens utilizing fewer than 5 days/week (eg, twice weekly INH) were not included in the primary comparison.

Statistical Analysis

The primary comparison in the study was between patients treated with 4R and 9H. Chi-square tests were used to compare rates of completion (plus odds ratios), reasons for treatment noncompletion, drug toxicity and successful clinical outcomes. Results were stratified by age, language, region of origin, whether the diagnosis of LTBI was made using an IGRA and whether the patient was a contact to an active case. To account for a lack of randomization, multivariable logistic regression was performed. Comparison of characteristics between the 2 treatment regimens was done using rank-sum tests and χ2 tests. The impact of individual demographic and clinical variables on treatment completion rates (listed above and converted to dichotomous variables where appropriate) was initially assessed in univariate analysis, using χ2 tests or Fisher exact tests [for variables with few (<5) data points]. Variables associated with a significant effect on LTBI completion status and variables with significant differences (P < 0.05) between treatment regimens were included in the multivariate analysis. Significance for all statistical tests was set at a 2-sided P value of <0.05. Analysis was performed using Stata, Version 13 (StataCorp, College Station, TX).

Ethical Approvals

The study was approved by the Colorado Multiple Institutional Review Board. Informed consent was not obtained or required by the Institutional Review Board as all data were de-identified and analyzed in aggregate.



The study population was comprised of 1174 patients treated with either 4 months of rifampin (n = 395) or 9 months of isoniazid (n = 779) (Figure 1). Demographic and clinical characterization of the 2 groups is presented in Table 1. There were some significant differences between the 2 comparison groups: on average, 4R-treated children were slightly older, more likely to have been tested for LTBI with an IGRA and be a contact to an active case of TB. There were also differences in the birth region and language distribution, with more 4R patients having family country of origin in North America and Asia, and more commonly speaking English as a preferred language. Origin from Latin America and speaking Spanish as preferred language was more common in the 9H cohort.

Demographic Characteristics of LTBI Patients Treated With 4R or 9H
Latent tuberculosis infection treatment completion rates by treatment regimen, Denver Metro Tuberculosis Clinic, 2006–2015. *Numbers represent patients treated by year; no patients received 4R in 2008 and 2009 and 9H in 2015.

LTBI Treatment Completion

Patients were significantly more likely to complete the LTBI regimen in the 4R group than in the 9H group (330/395; 83.5% vs. 536/779; 68.8%, P < 0.001) (Table, Supplemental Digital Content 1, When assessed by age, gender, country of origin, preferred individual language or common language (which we defined as speaking any language spoken by >5% of all patients, as a variable intended to capture the impact of more established language/interpreter service at the clinic; languages included English, Spanish, Nepali and Burmese), country of origin, IGRA status and contact status, 4R completion rates were statistically better compared with 9H for most variables. In no category was 9H significantly better than 4R. Because the majority of 4R treatment courses occurred after 2011 and the majority of 9H cases occurred before 2011, temporal trends in completion rates between 2006 and 2015 for both regimens were examined (Figure 1). In both cases, no significant trend was noted over a 10-year period [4R: Pearson correlation coefficient (r) = 0.19, P = 0.66; 9H: r = 0.46, P = 0.21].

Multivariable Analysis

Outcomes of the multivariable analysis are presented in Table 2. Variables included in the model based on significance in univariate analysis included the following: treatment regimen, age, speaking a commonly encountered primary language, speaking Nepali language, global region (Africa, South East Asia, Other), LTBI diagnosed using an IGRA and LTBI diagnosis in context of contact with an active case. In the multivariable model, treatment with 4R [odds ratio (OR): 1.64, 95% confidence interval (CI): 1.07–2.52, P = 0.023], contact with an active case (OR: 1.82, 95% CI: 1.13–2.93, P = 0.013) and speaking commonly encountered language (OR: 1.58, 95% CI: 1.02–2.45, P = 0.04) were associated with higher odds of completion of LTBI treatment.

Impact on LTBI Treatment Completion: Multivariable Logistic Regression

Reasons for Completion Failure

The majority of patients in both treatment groups who did not complete the initial LTBI regimen did so because of personal or social barriers, including parent or patient refusal to continue treatment, loss to follow-up or family relocation out of the DMTBC catchment area (majority out of the state of Colorado) (Table 3). No patients were lost to follow-up in the 4R cohort, whereas 27 9H patients (3.3% of total) were lost to follow-up. Drug toxicity was an uncommon reason for treatment noncompletion in either group: 1.5% of all subjects in the 4R group and 0.7% in the 9H group, a difference that was not statistically significant. The 6 instances of drug toxicity leading to failure to complete therapy included 3 patients with nonspecific rash, 2 with urticarial rash and 1 patient with presumed allergic facial swelling. There were no instances of symptomatic hepatotoxicity.

Reasons for Failure to Complete Initial LTBI Treatment Regimen


The study population represents the largest reported cohort of pediatric patients treated for LTBI with 4 months of daily rifampin, and demonstrates that the regimen is well tolerated and associated with higher rates of completion than a comparison group treated with 9 months of isoniazid. The observation of better completion was consistent across ages, languages, ethnicity/country of origin in our pediatric population, as well as in high-risk groups including contacts of active cases, young children and infants and patients who were IGRA positive.

The critical question in evaluating the potential benefit of improved completion rates of 4R over 9H is: to what extent is the efficacy of these regimens equivalent? Unfortunately, data comparing efficacy, particularly in an otherwise healthy North American population, are lacking. However, despite this limitation, the 14.5% absolute improvement in completion rates will result in equal or better overall effectiveness, as long as the efficacy of 4R was not worse by more than an equivalent percentage—with fewer required resources. Rifampin has bacteriocidal activity against M. tuberculosis and is a well-established component of treatment regimens for TB disease.13 In an early animal model, better clearance of M. tuberculosis from the spleen in a mouse model of latent TB was demonstrated in mice treated with 3 months rifampin (80% clearance) versus 6 months of isoniazid (0% clearance).5 In humans, 2 early trials provide efficacy data for rifampin in treatment of LTBI in adults. Among adult LTBI patients with silicosis in Hong Kong, 3 months of rifampin was twice as effective as 6 months of isoniazid in prevention of development of TB disease in a 5-year period.6 Rifampin was 100% effective at preventing progression to TB disease among 49 homeless adults infected with an INH-resistant strain (followed for a mean duration of 26.8 months) compared with 7.9%–8.6% in untreated or INH-treated patients.7 Similar results were observed in 157 adolescents exposed to a suspected INH-resistant organism, none of whom developed TB disease after a 24-week course of rifampin, with a 2-year follow-up.14

Though no cases of active TB developed in the 395 children treated with 4R, our findings make only a limited additional contribution to effectiveness data of 4R regimens because we do not have long-term follow-up data to assess for the emergence symptomatic disease. However, all pediatric patients with active TB disease in the state of Colorado are managed in consultation with the Denver Metro TB clinic, and we have encountered no cases of active disease as of February 2017 in any of the 395 patients in this study started on 4R, with a median follow-up time since treatment initiation of 4 years and 6 months. Perhaps more convincing evidence for effectiveness in our population can be found in subpopulations at higher risk of progression to TB disease in whom the lack of active cases is more notable: a third of 4R-treated pediatric patients were contacts to an active case, and 16% were under 5 years of age.

In addition to efficacy, other factors to be considered in comparison of regimens for pediatric LTBI include tolerance/toxicity, cost, drug interactions and potential for developing antimicrobial resistance. Our patient population demonstrated high tolerability of 4R, with less than 2% of recipients experiencing adverse effects necessitating a change in regimen. This is consistent with adult literature in which rates of significant hepatotoxicity are lower than for 9-month INH regimens and with pediatric literature and clinical practice in which 4R is very well tolerated.8,9,11,14

The impact of 4R on cost relative to 9H has been variably estimated. In the United States, a 4-month course rifampin is likely to be more expensive than 9H in terms of direct cost to patients or TB programs.11 However, any comprehensive estimate of the impact on cost of 4R would need to include potential savings associated with successful prevention of active disease through higher rates of completed therapy, reduced side effects and blood monitoring and personnel time requirements, which may serve to make 4R the more cost-effective regimen.10 This more complete calculus may also be relevant in low- and middle-income countries, where resources required to manage longer INH courses for pediatric LTBI often mean that such programs are not well supported, particularly for children.15 4R could make tackling pediatric LTBI in such places a more attractive and feasible public health activity, with the potential for a long-term contribution to TB elimination.

Drug interactions represent another potential disadvantage of 4R, in particular, adverse impacts on metabolism of oral contraceptive medications and antiepileptics.16 However, adverse outcomes due to drug–drug interaction are generally avoidable with education, adequate information gathering and use of alternate options.

A final area of uncertainty with 4R regimens for LTBI is the potential for development of antimicrobial resistance. Monotherapy with rifampin has been associated with rapid development of resistance in typical bacterial infections, most notably Staphylococcus aureus.17 Selection for rifampin resistance during monotherapy has been clearly demonstrated in active TB.18 The lower bacillary burden in pediatric LTBI should serve to significantly lower the risk of development of rifampin resistance, but this has not been explicitly demonstrated in the case of 4R regimens. Page et al8 report 1 4R LTBI treatment failure out of 1379 patients, in whom rifampin-susceptible M. tuberculosis was cultured from a lymph node despite 4 months of reported compliant exposure. The possibility of selecting rifampin resistance through use of 4R regimens for LTBI in a patient that actually has active TB dictates the need for careful discrimination of LTBI from active disease at initiation of therapy, which can be challenging in infants and young children. Approaches in uncertain cases may be better addressed with INH- or combination-based regimens. Additionally, because higher doses of rifampin have been associated with improved bacteriologic outcomes, our practice is to aim for the higher end of recommended dose for children (typically targeting 20 mg/kg/dose, within the constraints of available formulations) that may serve to limit risk of development of resistance through more rapid bacterial clearance.19

Some important limitations of our study should be acknowledged. Perhaps most notably we do not have long-term follow-up to formally assess efficacy, as discussed above. Additionally, just under 22% of children treated with 4R were diagnosed with LTBI on the basis of TST without IGRA, and of these 45% (9.6% of the total) were bacille Calmette-Guerin vaccinated, and thus a small percentage may not have been infected with TB at all. However, an alternate perspective on this observation is that such patients will be encountered and treated for LTBI by TB controllers, and 4 months of “unnecessary treatment” may well be better than 9, particularly from the standpoint of resource utilization. Given the restraints of the retrospective dataset, it is possible that there are unmeasured factors beyond the treatment regimen that are influencing the difference in completion rates. We were unable to measure rates of treatment uptake, because the dataset did not capture patients who refused all treatment from the outset nor the provider and patient factors leading to the selection of one regimen over another, raising the potential for unidentified bias in completion rates. Of these, perhaps most likely would be selection of 4R by clinic providers for patients who were deemed less likely to complete 9H in the time frame before 4R becoming the preferred regimen; if present, this would be more likely to lower the completion rate of 4R. Another potential confounding factor could be that there have been general improvements in case management that increased completion rates in 4R, which is our more recent regimen. However, we did not identify any temporal trend in improvement for either regimen over the study period. Similarly, the selection of 4R regimens has coincided with increasing use of IGRA testing to establish the diagnosis of TB infection in our center. Though not statistically significant in the multivariate analysis, the increased specificity of this test may serve as an additional motivator for completion that we have not quantified in our dataset. In fact, it is likely that the more specific test and a shorter LTBI regimen are complementary in motivating patients to complete therapy.


1. Kimberlin DW, Brady MT, Jackson MA, et al.; American Academy of Pediatrics, Committee on Infectious Diseases. Tuberculosis. In: Red Book: 2015 Report of the Committee on Infectious Diseases. 2015:30th ed. Elk Grove Village, IL; American Academy of Pediatrics, 805831.
2. Centers for Disease Control and Prevention. Treatment Regimens for Latent TB Infection (LTBI) [CDC website]. April 5, 2016. Available at: Accessed March 23, 2017.
3. Sumartojo E. When tuberculosis treatment fails. A social behavioral account of patient adherence. Am Rev Respir Dis. 1993;147:13111320.
4. World Health Organization. Global Tuberculosis Report 2016. 2016. Geneva: World Health Organization; Available at: Accessed May 18, 2016.
5. Lecoeur HF, Truffot-Pernot C, Grosset JH. Experimental short-course preventive therapy of tuberculosis with rifampin and pyrazinamide. Am Rev Respir Dis. 1989;140:11891193.
6. A double-blind placebo-controlled clinical trial of three antituberculosis chemoprophylaxis regimens in patients with silicosis in Hong Kong. Hong Kong Chest Service/Tuberculosis Research Center, Madras/British Medical Research Council. Am Rev Respir Dis. 1992;145:3641.
7. Polesky A, Farber HW, Gottlieb DJ, et al. Rifampin preventive therapy for tuberculosis in Boston’s homeless. Am J Respir Crit Care Med. 1996;154:14731477.
8. Page KR, Sifakis F, Montes de Oca R, et al. Improved adherence and less toxicity with rifampin vs isoniazid for treatment of latent tuberculosis: a retrospective study. Arch Intern Med. 2006;166:18631870.
9. Menzies D, Long R, Trajman A, et al. Adverse events with 4 months of rifampin therapy or 9 months of isoniazid therapy for latent tuberculosis infection: a randomized trial. Ann Intern Med. 2008;149:689697.
10. Ziakas PD, Mylonakis E. 4 months of rifampin compared with 9 months of isoniazid for the management of latent tuberculosis infection: a meta-analysis and cost-effectiveness study that focuses on compliance and liver toxicity. Clin Infect Dis. 2009;49:18831889.
11. Cruz AT, Starke JR. Safety and completion of a 4-month course of rifampicin for latent tuberculous infection in children. Int J Tuberc Lung Dis. 2014;18:10571061.
12. Spyridis NP, Spyridis PG, Gelesme A, et al. The effectiveness of a 9-month regimen of isoniazid alone versus 3- and 4-month regimens of isoniazid plus rifampin for treatment of latent tuberculosis infection in children: results of an 11-year randomized study. Clin Infect Dis. 2007;45:715722.
13. Norton BL, Holland DP. Current management options for latent tuberculosis: a review. Infect Drug Resist. 2012;5:163173.
14. Villarino ME, Ridzon R, Weismuller PC, et al. Rifampin preventive therapy for tuberculosis infection: experience with 157 adolescents. Am J Respir Crit Care Med. 1997;155:17351738.
15. Kasprowicz VO, Churchyard G, Lawn SD, et al. Diagnosing latent tuberculosis in high-risk individuals: rising to the challenge in high-burden areas. J Infect Dis. 2011;204(suppl 4):S1168S1178.
16. Finch CK, Chrisman CR, Baciewicz AM, et al. Rifampin and rifabutin drug interactions: an update. Arch Intern Med. 2002;162:985992.
17. Tang HJ, Lai CC, Hsueh PR, et al. RNA polymerase B subunit gene mutations in biofilm-embedded methicillin-resistant Staphylococcus aureus following rifampin treatment. J Microbiol Immunol Infect. 2016;49:394401.
18. Kayigire XA, Friedrich SO, van der Merwe L, et al. Acquisition of rifampin resistance in pulmonary tuberculosis. Antimicrob Agents Chemother. 2017;61: e0222016.
19. Boeree MJ, Diacon AH, Dawson R, et al.; PanACEA Consortium. A dose-ranging trial to optimize the dose of rifampin in the treatment of tuberculosis. Am J Respir Crit Care Med. 2015;191:10581065.

latent tuberculosis infection; rifampin; pediatric

Supplemental Digital Content

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.