Oral rifampicin interferes with urine dipstick tests for patients with pulmonary tuberculosis
Highlight box
Key findings
• The collection of urine samples within 12–24 h of administration of RMP can prevent the interference caused by RMP and dRMP.
What is known and what is new?
• RMP and dRMP can interfere with the detection of some urine chemistry analytes in UDTs and produce erroneous results.
• The anti-interference ability of Aution Sticks 10EA (Arkray, Japan), Combi-Screen 11SYS Plus (GIMA, France).
• Aution Sticks 10EA (Arkray, Japan) and Combi-Screen 11SYS Plus (GIMA, France) Plus urine dipsticks were evaluated by measuring the interference caused by the total RMP in the UDTs following an oral administration of RMP.
What is the implication, and what should change now?
• In the UDTs of patients with tuberculosis who receive oral RMP, protein, leukocyte esterase, ketone, and nitrite tests interfere with the false-positive results caused by RMP and dRMP in the urine samples. The affected analytes may vary across different types of urine test strips.
Introduction
The urine dipstick test (UDT) is a screening tool commonly used for the diagnosis of urinary tract infections (UTIs), kidney diseases, and other metabolic disorders (1). This cost-effective, easy-to-use, fast-reporting, non-invasive test is part of routine urinalysis in clinical settings. Given the long duration of therapy and potential adverse drug reactions (ADRs) associated with anti-tuberculosis (anti-TB) treatment (2,3), regular liver and kidney function tests and visual acuity assessments are required, particularly in instances of liver injury, UTIs, and diabetic nephropathy caused by anti-TB drugs, and urine chemistry analysis plays a significant role in disease screening (4,5).
Rifampicin (RMP) is a first-line anti-TB drug that has the highest bioavailability when taken on an empty stomach. Thus, individuals with TB should take RMP orally in the morning or before bedtime every day. The peak serum concentration of RMP can be measured 2–4 h after its oral administration. The metabolites of RMP are predominantly eliminated via the biliary route, but a fraction of such excreted metabolites remains unchanged or as desacetyl rifampicin (dRMP) in the urine. The urine concentration of RMP reaches its peak at 6 h after oral administration and returns to normal at 12 h post dose (6). Urine containing RMP and its major metabolite dRMP is orange red, with a darker color indicating a higher concentration. This finding can interfere with the detection of some urine chemistry analytes in UDTs and cause erroneous results that affect clinicians’ assessments of ADRs and treatment decisions.
In this study, the colorimetric method (7), which predicts the pharmacokinetics of RMP in plasma (8,9), was applied to measure the total concentration of RMP and its essential metabolite dRMP (hereinafter collectively referred to as the “total RMP concentration”) in urine. As the Combi-Screen 11SYS Plus test strips and Audio Sticks 10EA may have different test results, we evaluate the anti-interference ability of Arkray’s Aution Sticks 10EA and GIMA’s Combi-Screen 11SYS Plus urine dipsticks by measuring the interference caused by the total RMP in the UDTs following the oral administration of RMP. Anti interference ability is a capability of the urine test strip itself, which can counteract the influence of certain drug metabolites and other metabolites in the body, ensuring the accuracy of the results. The use of a preferable anti-interference urine test strip may improve the accuracy of the UDT results. This study sought to examine the effects of RMP and dRMP on UDTs using 2 different urine dipsticks (i.e., Arkray’s Aution Sticks 10EA and GIMA’s Combi-Screen 11SYS Plus sticks). We present this article in accordance with the MDAR reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-310/rc).
Methods
Determining the range of the total RMP concentration in the urine after oral administration of RMP
Urine sample collection
This study included 40 patients with TB who underwent a 2-week treatment regimen of RMP and at least 2 other anti-TB drugs, had normal kidney function and hemo-bilirubin and blood glucose levels, and denied a history of diabetes. Oral RMP was calculated as a dose of per 10 mg/kg body weight. Urine samples were collected randomly within 2–6 and 12–24 h of administration of RMP. Each urine sample was collected in a 10-mL urine collection tube with a cap, and 1 mL of each sample was centrifuged at 13,000 rpm for 10 min to harvest the supernatant, which was stored at −70 °C. The maximum storage time was 90 days. The patients’ clinical and demographic characteristics are presented in Table 1.
Table 1
| Characteristic | Values (n=40) |
|---|---|
| Age (years), median (range) | 44.5 (17 to 73) |
| Sex, n (%) | |
| Male | 28 (70.0) |
| Female | 12 (30.0) |
| Oral dose of rifampicin, n (%) | |
| 300 mg | 6 (15.0) |
| 450 mg | 9 (22.5) |
| 600 mg | 25 (62.5) |
| Serum total bilirubin, n (%) | |
| <21 μmol/L | 40 (100.0) |
| Renal function | Normal |
| History of diabetes | None |
TB, tuberculosis.
The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the Ethics Committee of Wuhan Pulmonary Hospital [No. (2021)8] and informed consent was taken from all the patients.
Measuring the total RMP concentration in the urine via colorimetry
A 500 mg/L mixture urine sample was prepared by adding the urine of 5 healthy individuals to RMP (SIGMA-ALDRICH, American), and 10 different concentrations were obtained by doubling the dilution. These mixtures (150 µL each) were subsequently inoculated into a 96-well plate, and the concentrations were measured by their absorbances at 475 nm. Next, an RMP absorbance-concentration standard curve was plotted, using the urine from the healthy participants as the blank control.
The frozen samples of the 40 patients were thawed in a water bath at 37 °C for 5 minutes, and the total RMP concentration was isolated from each urine sample using Sunahara’s method (6). Next, 150 µL of each sample was added to a 96-well plate, the absorbance was read at 475 nm, and the total RMP concentration in the urine was then calculated in combination with the standard curve.
Determining the range of the total RMP concentration in the urine
The total RMP concentration range in the urine reached the maximum concentration at 2–6 h post dose and the minimum concentration at 12–24 h post dose.
In vitro assays: simulating interference with UDT caused by RMP
Interference assays with false-positive results
The reference standard RMP (SIGMA-ALDRICH, America) was initially added to the mixture of the urine of the healthy individuals and then diluted to concentrations of 0, 100, 125, 200, 250, 300, 350, 400, 500, and 1,000 mg/L. Urine chemistry analyzers and urine test strips (i.e., Aution Sticks 10EA, Arkray, Japan; Combi-Screen 11SYS Plus, GIMA, France) were used for the urinalysis. Positive degrees and corresponding RMP concentrations were documented if any positive results were derived from the assays.
Interference assays with false-negative results
The positive control (BIO-RAD, California, American) was diluted until it reached a positive degree of “1+” provided in the different urine chemistry parameters, and the specific RMP concentrations of the diluted samples were 125, 250, 500, and 1,000 mg/L. Next, these samples were tested using Combi-Screen 11SYS Plus and Aution Sticks 10EA, and given urine test strips.
Definition of the in vitro interference threshold
The in vitro interference threshold was defined as the known RMP concentration that caused (false-positive or false-negative) interference with any urine chemistry analyte in a UDT.
Confirmatory test: revealing the interference with the UDT caused by RMP and dRMP in the urine of patients with TB
Patients and sampling
In total, 75 other patients who were diagnosed with pulmonary TB, had normal kidney function and serum total bilirubin (BIL) and glucose levels, denied a history of diabetes, and received oral RMP and at least 2 other anti-TB drugs for 2 weeks were enrolled in this study. Urine samples were obtained from these patients at any time within 2–12 h post dose and placed in a 10-mL urine collection tube with a cap. Next, 1 mL of the sample was centrifuged at 13,000 rpm for 10 min to harvest the supernatant, which was stored at −70 °C awaiting the subsequent determination of the total RMP concentration in the urine. The remaining samples were subjected to UDTs, urine microalbumin testing, and urine sediment microscopy using Combi-Screen 11SYS Plus and Aution Sticks 10EA. The patients’ clinical and demographic characteristics are presented in Table 2.
Table 2
| Characteristic | Values (n=75) |
|---|---|
| Age (years), median (range) | 42 (17 to 82) |
| Sex, n (%) | |
| Male | 52 (69.3) |
| Female | 23 (30.7) |
| Urine microalbumin <150 mg/L, median (range) | 70 (0.6 to 88.2) |
| Urine sediment, n (%) | |
| RBC (<1/HPF) | 54 (72.0) |
| WBC (<1/HPF) | 33 (44.0) |
| Bacterial (negative) | 75 (100.0) |
| Serum total bilirubin, n (%) | |
| <21 μmol/L | 75 (100.0) |
| Renal function | Normal |
| History of diabetes | None |
TB, tuberculosis; RBC, red blood cell; WBC, white blood cell; HPF, high-power field.
UDT, urine sediment microscopy, and urine microalbumin tests
All the 75 urine collection tubes with urine samples were loaded into Arkray’s and Gima’s urine chemistry analyzers for the UDTs. The supernatants obtained from these samples via low-speed centrifugation were tested for microalbumin, and the sediments were examined by microscopy. All the procedures were completed within 2 h of the sampling.
Determination of the total RMP concentration in the urine
The total RMP concentration in the urine was determined as described in Section “Measuring the total RMP concentration in the urine via colorimetry”.
Interpretation criteria for the true-negative results for the urine chemistry analytes
In the urine sediment microscopy, “white blood cell count <1/high-power field” and “bacterial negative” were interpreted as true negatives for leukocyte esterase (LEU) and nitrite (NIT), respectively. “Urine microalbumin <150 mg/L” was classified as a true negative for protein (PRO), and “serum total BIL <21 mmol/L” was classified as a true negative for BIL. A normal blood glucose level was considered a true negative for ketone (KET).
Urine dipstick grading systems given by Arkray and Gima
Grading systems used for the Aution Sticks 10EA and the Combi-Screen 11SYS Plus dipsticks. And Grading systems included negative, weakly positive and positive (Table 3).
Table 3
| Projects | Aution Sticks 10EA | Combi-Screen 11SYS Plus | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Negative (−) | Positive | Negative (−) | Positive | ||||||||
| ± | 1+ | 2+ | 3+ | ± | 1+ | 2+ | 3+ | ||||
| Protein, g/L | – | 0.15 | 0.3 | 1 | 3 | − | 0.15 | 0.3 | 1 | 5 | |
| Leukocyte, μL | – | <25 | 25 | 75 | 250 | – | <25 | 25 | 75 | 500 | |
| Ketones, mmol/L | – | 0.5 | 1.5 | 4 | 8 | – | 1.0 | 2.5 | 10 | 30 | |
| Nitrite | – | + | – | + | |||||||
“–”: negative; “±”: weakly positive; “+”: positive.
Definition of interference threshold in the confirmatory test
The interference threshold was defined as the minimum level of increasing RMP concentration when the interference (the false-positive or false-negative results) reached 95% for a specific urine chemistry analyte.
Statistical analysis
SPSS 19.0 statistical software was used. P<0.05 was defined as a statistically significant difference. This study mainly applied descriptive statistics, with counting data represented by n (%), and measurement data represented by mean ± standard deviation (SD) or scatter diagram.
Results
Total RMP concentration range in the urine
As the RMP concentration-absorbance calibration curve in Figure 1 shows, a good linear correlation (r2=0.9982) was observed when the absorbance wavelength was 475 nm and the RMP concentration was between 1.95 and 500 µg/mL. Based on the test results of the 40 urine samples and the standard curve, the concentration of total RMP ranged from 88 to 376 µg/mL (median: 223 µg/mL) within 2–6 h of the oral administration of RMP, and between 22 and 112 µg/mL (median: 48 µg/mL) within 12–24 h of the oral administration of RMP (Figure 2).
In vitro interference assays
Aution Sticks 10EA
When the RMP had a concentration range of 0–500 µg/mL, false-positive interferences were observed at above 250 and 400 µg/mL for PRO (±) and LEU (1+), respectively. No false-negative interference was detected in the measurement of the given 10 parameters.
Combi-Screen 11SYS Plus
For the 0–500 µg/mL range, RMP was associated with false-positive interferences that appeared in the measurements of LEU (1+) and KET (±) when the RMP concentration was ≥125 µg/mL, in the measurements of PRO (3+) when the RMP concentration was ≥200 µg/mL, and in the measurements of NIT (+) when the RMP concentration was ≥500 µg/mL. No false-negative interferences were observed in the measurements of the given substances.
Confirmatory tests
Aution Sticks 10EA
When the total RMP concentration in the urine reached 219 µg/mL, interferences emerged in the measurement of PRO, and the false-positive rate increased from 88.9% to 100% at a concentration of 250 µg/mL. When the total RMP concentration in the urine reached 267 µg/mL, interference emerged in the measurement of LEU, and the false-positive rate increased from 77.8% to 100% at a concentration of 300 µg/mL and above (Table 4).
Table 4
| Analyte | Urinary total rifampicin concentration range (μg/mL) | Negative No. | Interference results | |||
|---|---|---|---|---|---|---|
| ± | 1+ | 2+ | Subtotal, n (%) | |||
| Protein | 219–400 | 18 | 9 | 7 | – | 16 (88.9) |
| 250–400 | 13 | 8 | 5 | – | 13 (100.0) | |
| Leukocyte esterase | 267–400 | 9 | – | 6 | 1 | 7 (77.8) |
| 300–400 | 4 | – | 3 | 1 | 4 (100.0) | |
Combi-Screen 11SYS Plus
The false-positive results were found to interfere with KET when the total RMP concentration in the urine reached ≥123 µg/mL (with a false-positive rate of 85%), with LEU at 131 µg/mL (with a false-positive rate of 90.9%), with KET and LEU at 150 µg/mL (with false-positive rates of 100% and 94.4%, respectively), with PRO at 289 µg/mL (with a false-positive rate of 87.5% and a false-positive rate of 100% if the concentration ≥300 µg/mL), and with NIT at 346 µg/mL (with a false-positive rate of 100%) (Table 5).
Table 5
| Analyte | Urinary total rifampicin concentration range (μg/mL) | Negative No. | Interference results | |||
|---|---|---|---|---|---|---|
| 1+ | 2+ | 3+ | Subtotal, n (%) | |||
| Ketones | 123–400 | 27 | 23 | – | – | 23 (85.2) |
| 150–400 | 19 | 19 | – | – | 19 (100.0) | |
| Leukocyte esterase | 131–400 | 22 | 7 | 11 | 2 | 20 (90.9) |
| 150–400 | 18 | 5 | 10 | 2 | 17 (94.4) | |
| Protein | 289–400 | 8 | – | – | 7 | 7 (87.5) |
| 300–400 | 6 | – | – | 6 | 6 (100.0) | |
| Nitrite | 346–400 | 3 | 3 | – | – | 3 (100.0) |
| 350–400 | 2 | 2 | – | – | 2 (100.0) | |
Comparison of the degrees of interference between the 2 urine dipsticks
In relation to the Aution Sticks 10EA, false-positive interferences were only observed for PRO and LEU. Moreover, the false-positive results were not detectable until after the total RMP concentration reached 250 µg/mL in the confirmatory test. In relation to the Combi-Screen 11SYS Plus urine strips, false-positive interferences were observed in some analytes, including NIT, PRO, LEU, and KET. However, the phenomenon was not observed until the total RMP concentration reached ≥150 µg/mL in the confirmatory test (Table 6). Thus, the Aution Sticks 10EA were more resistant to the interference caused by RMP than the Combi-Screen 11SYS Plus urine dipsticks.
Table 6
| Tests | In vitro interference assays (μg/mL) | Confirmatory tests (μg/mL) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| NIT | PRO | LEU | KET | NIT | PRO | LEU | KET | ||
| Aution Sticks 10EA | – | 250 | 400 | – | – | 250 | 300 | – | |
| Combi-Screen 11SYS Plus | 500 | 200 | 125 | 125 | 350 | 300 | 150 | 150 | |
NIT, nitrite; PRO, protein; LEU, leukocyte esterase; KET, ketones.
Discussion
In this study, the total RMP concentration in the urine fell within the range of 88–376 µg/mL at 2–6 h post dose, which is similar to the results reported by Acocella (6). At 475 nm, the RMP concentration was less affected by other anti-TB drugs and their metabolites, while the maximum urine BIL absorption peak at 460 nm can affect the determination of RMP concentrations by colorimetry. The true negativity of urine BIL was defined on the basis of the serum total BIL within the normal range to eliminate the effect of BIL, and all the TB patients included in the study had a normal serum BIL level.
The study results suggested that RMP and dRMP caused false-positive interference in the urine dipstick tests, including in relation to PRO, LEU, KET, and NIT, but they were independent of pH, specific gravity, glucose, occult blood, urobilinogen, and BIL. In this study, patients with normal blood glucose levels and without diabetes were considered a true negative for urine ketone. This result might have affected the accuracy of the study results, as an individual can test positive for urine KET after fasting or drinking (10,11). Thus, positivity should not be ruled out simply based on urine and blood glucose results that fall within the normal ranges before a blood KET test. There were 19 urine samples with a total RMP concentration >150 µg/mL. All these samples tested positive for KET with the Combi-Screen 11SYS Plus test strips and negative with the Aution Sticks 10EA, which indicates that there is a high probability of false-positive results in the KET test when the Combi-Screen 11SYS Plus urine strips are used.
Despite its advantages, the easy-to-operate in vitro interference assay does not accurately describe the degree of interference with UDT. Differences existed in the corresponding interference thresholds, but the affected analytes were consistent between the in vitro interference assays and confirmatory tests. This might be due to the presence of dRMP in the urine samples obtained from the patients with TB in the confirmatory tests. Apart from RMP and dRMP, the urine samples of the participants treated with combination therapies contained metabolites, such as isoniazid, pyrazinamide, ethambutol, and/or other anti-TB agents. These substances might have contributed to the differences in the interference thresholds between the confirmatory tests and in vitro interference assays. Thus, the in vitro interference assay cannot be used as a substitute for the confirmatory test.
The test principles of the urine test strips are another significant factor contributing to the difference in the interference threshold for LEU. In relation to the Aution Sticks 10EA, LEU catalyzes the hydrolysis of 3-(N toluene sulfonyl-L-alanyloxy) indole to form indoxyl, which further reacts with 2-methoxy-4-(N-morpholino) benzene diazonium to generate a purple product. When tested by the Combi-Screen 11SYS Plus urine strips, LEU splits heterocyclic carboxylates, and the product ethanol reacts with diazonium salt, producing a violet color. The interference caused by RMP and dRMP likely occurs in tests using the Combi-Screen 11SYS Plus urine strips.
The anti-interference performance of the urine test strips vary in the presence of RMP. When the total RMP concentration in the urine was 150 µg/mL, interference occurred simultaneously in the LEU and KET tests using the Combi-Screen 11SYS Plus urine strips. When the Aution 10EA Sticks were used, interference with PRO was not detectable until the total RMP concentration in the urine increased to 250 µg/mL. Notably, fewer analytes interfered with RMP when the AUTION Sticks 10EA were used for the urinalysis. The highest degree of interference with PRO was limited to “1+” for the Aution Sticks 10EA, which was significantly lower than that (“3+”) for the Combi-Screen 11SYS Plus urine strips. Moreover, a lower degree of interference was observed in the LEU test with the Aution Sticks 10EA than the Combi-Screen 11SYS Plus urine strips (“2+” vs. “3+”). Thus, the Aution Sticks 10EA outperformed the Combi-Screen 11SYS Plus urine strips in terms of resisting RMP-induced interference. In clinical settings, highly resistant urine test strips are needed to obtain the most accurate test results.
The resistance of urine test strips to the interference caused by RMP should be evaluated, particularly in medical institutions for TB prevention and control. As reported by numerous clinical studies, high doses of RMP are widely used to shorten the course of anti-TB treatments and may be included in clinical treatments for TB (12,13). Our findings highlight the importance of evaluating the anti-interference performance of test strips, as patients with TB are likely to have higher concentrations of RMP in their urine. In this study, the oral RMP recipients had a total RMP concentration in their urine ranging from 22 to 112 µg/mL within 12–24 h post dose, but no interference was observed with any of the UDTs. Thus, the suspension of therapy is not necessary for every patient with TB. However, UDTs can be conducted within 12–24 h following the administration of oral RMP.
Conclusions
In the UDTs of TB patients who are receiving oral RMP, tests of PRO, LEU, KET, and NIT interfere with the false-positive results caused by RMP and dRMP in the urine samples, and the affected analytes may vary across different types of urine test strips. Given that in vitro interference assays cannot comprehensively reveal the interference in UDTs based on the urine samples of oral RMP recipients, laboratory testing is required to evaluate the resistance of urine dipsticks to the interference caused by RMP and dRMP. False-positive interference can be avoided by conducting UDTs within 12–24 h of the oral administration of RMP.
Acknowledgments
Funding: None.
Footnote
Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-310/rc
Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-310/dss
Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-310/prf
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-310/coif). The authors have no conflicts of interest to declare.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the Ethics Committee of Wuhan Pulmonary Hospital [No. (2021)8] and informed consent was taken from all the patients.
Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
References
- Simerville JA, Maxted WC, Pahira JJ. Urinalysis: a comprehensive review. Am Fam Physician 2005;71:1153-62. [PubMed]
- Schaberg T, Rebhan K, Lode H. Risk factors for side-effects of isoniazid, rifampin and pyrazinamide in patients hospitalized for pulmonary tuberculosis. Eur Respir J 1996;9:2026-30. [Crossref] [PubMed]
- Chung SJ, Byeon SJ, Choi JH. Analysis of Adverse Drug Reactions to First-Line Anti-Tuberculosis Drugs Using the Korea Adverse Event Reporting System. J Korean Med Sci 2022;37:e128. [Crossref] [PubMed]
- Saukkonen JJ, Cohn DL, Jasmer RM, et al. An official ATS statement: hepatotoxicity of antituberculosis therapy. Am J Respir Crit Care Med 2006;174:935-52. [Crossref] [PubMed]
- Chu CM, Lowder JL. Diagnosis and treatment of urinary tract infections across age groups. Am J Obstet Gynecol 2018;219:40-51. [Crossref] [PubMed]
- Acocella G, Mattiussi R, Segre G. Multicompartmental analysis of serum, urine and bile concentrations of rifampicin and desacetyl-rifampicin in subjects treated for one week. Pharmacol Res Commun 1978;10:271-88. [Crossref] [PubMed]
- Sunahara S, Nakagawa H. Metabolic study and controlled clinical trials of rifampin. Chest 1972;61:526-32. [Crossref] [PubMed]
- Zentner I, Schlecht HP, Khensouvann L, et al. Urine colorimetry to detect Low rifampin exposure during tuberculosis therapy: a proof-of-concept study. BMC Infect Dis 2016;16:242. [Crossref] [PubMed]
- Szipszky C, Van Aartsen D, Criddle S, et al. Determination of Rifampin Concentrations by Urine Colorimetry and Mobile Phone Readout for Personalized Dosing in Tuberculosis Treatment. J Pediatric Infect Dis Soc 2021;10:104-11. [Crossref] [PubMed]
- Bashir B, Fahmy AA, Raza F, et al. Non-diabetic ketoacidosis: a case series and literature review. Postgrad Med J 2021;97:667-71. [Crossref] [PubMed]
- Shah P, Isley WL. Ketoacidosis during a low-carbohydrate diet. N Engl J Med 2006;354:97-8. [Crossref] [PubMed]
- Boeree MJ, Heinrich N, Aarnoutse R, et al. High-dose rifampicin, moxifloxacin, and SQ109 for treating tuberculosis: a multi-arm, multi-stage randomised controlled trial. Lancet Infect Dis 2017;17:39-49. [Crossref] [PubMed]
- Velásquez GE, Brooks MB, Coit JM, et al. Efficacy and Safety of High-Dose Rifampin in Pulmonary Tuberculosis. A Randomized Controlled Trial. Am J Respir Crit Care Med 2018;198:657-66. [Crossref] [PubMed]
(English Language Editor: L. Huleatt)
