Hindawi

International Journal of Analytical Chemistry

Volume 2019, Article ID 4848236, 7 pages
https://doi.org/10.1155/2019/4848236

Research Article
Fast and Simple LC-MS/MS Method for Rifampicin
Quantification in Human Plasma

Cane Temova Rakuša,1 Robert Roškar,1 Anita KlanIar Andrejc,1 Tina Trdan Lušin,1
Nataša Faganeli,1,2 Iztok Grabnar,1 Aleš Mrhar ,1 Albin Kristl,1 and Jurij Trontelj 1

1
University of Ljubljana, Faculty of Pharmacy, Ljubljana, Slovenia
2
Valdoltra Orthopaedic Hospital, Ankaran, Slovenia

Correspondence should be addressed to Jurij Trontelj; [email protected]

Received 30 November 2018; Revised 28 December 2018; Accepted 8 January 2019; Published 3 February 2019

Academic Editor: Stig Pedersen-Bjergaard

Copyright © 2019 Žane Temova Rakuša et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.

A simple, fast, and cost-effective LC-MS/MS method for quantification of rifampicin in human plasma was developed and fully
validated. The plasma samples containing rifampicin and isotopically labelled internal standard rifampicin D8, were cleaned up
using a Captiva ND Lipids filtration plate. Chromatographic separation was achieved on an 1290 Infinity liquid chromatograph
coupled to 6460 Triple Quadrupole operated in positive mode on a core-shell Kinetex C18 column (50 × 2.1 mm, 2.6 𝜇m) by
gradient elution using 0.1% formic acid in water and acetonitrile as a mobile phase. The proposed method is the fastest method
published by now, both in terms of sample preparation (approximately one minute per sample) and chromatographic analysis
(total run time 2.4 min). Another key benefit is the outstanding sensitivity and wide analytical range (5-40000 𝜇g/L) with good
linearity, accuracy, and precision. The method showed almost complete recovery (92%) and absence of any significant matrix effect
as demonstrated by uniform responses from QC samples prepared in blood plasma from 6 volunteers (RSD <5%). The proposed
method was successfully applied to rifampicin quantification in 340 patients’ plasma samples, thus demonstrating its suitability for
both therapeutic drug monitoring and pharmacokinetic analysis.

1. Introduction failure and development of drug resistance. Furthermore, it

has been reported that the pediatric population differs in
Tuberculosis is an aggressive infectious disease caused by RIF pharmacokinetic parameters, requiring almost twofold
Mycobacterium tuberculosis. According to WHO, it is cur- higher dosage in mg/kg body weight, compared to adults
rently the leading cause of death from a single infectious
to reach equivalent plasma concentrations [5]. Considering
agent, with a reported mortality of 1.6 million in 2017
its pharmacokinetic variability, therapeutic drug monitoring
[1]. Rifampicin (RIF) is highly effective bacteriostatic and
bactericidal macrocyclic antibiotic, primarily used as one (TDM) is a useful and valuable tool for improvement of RIF
of the first line drugs in the treatment of tuberculosis [2]. effectiveness by individual dose adjustment and consequently
It is also indicated in the treatment of leprosy, meningitis, prevention of therapeutic failure and drug resistance [9]. RIF
osteomyelitis, and staphylococcal endocarditis [3]. Further- TDM may be also beneficial in the sense of preventing or
more, the role of RIF is well-defined in patients with pros- reducing the incidence of potential toxicity and side effects,
thetic joint infections [4]. RIF pharmacokinetics has been including hepatotoxicity, which is the most common serious
reported as complex with significant interindividual vari- adverse effect, reported in 13-36% of the patients [10]. RIF
ability and plasma concentrations below the recommended determination in patients’ plasma is further recommended
therapeutic range (8000-24000 𝜇g/L) in majority of adult in multidrug therapies, since it is a potent inducer of drug
patients [5–8]. The latter is tightly associated with therapeutic transporters and metabolizing enzymes. As such, RIF may
2 International Journal of Analytical Chemistry

interact and thus reduce the plasma concentration of other was 1 𝜇L. Total run time including reequilibration was 2.4
drugs, including other tuberculosis drugs, which are often min. The mass spectrometer was operated in positive mode
used in combination with RIF [11]. using the following parameters of the ion source: gas flow
For routine monitoring of RIF plasma concentration a (and temperature): 5 L/min (275∘ C), nebulizer pressure: 310
simple, selective, robust, and cost-effective method is needed. kPa, and sheath gas flow (and temperature): 11 L/min (400∘ C),
There are several reported methods for RIF quantification in and the capillary voltage was 4000 V. The fragmentor voltage
human plasma, based primarily on HPLC-UV [12–24] and was 80 V. Both quadrupoles were set at the widest mass
LC-MS/MS [9, 10, 25–33]. However, all of them are associated resolution (2.5 amu) with the following m/z transitions: 823.4
with certain disadvantages to their widespread use in clinical 󳨀→ 107.1 at 61 eV and 823.4 󳨀→ 163.1 at 41 eV for qualifier
practice. The most commonly encountered issues include and quantifier ion of rifampicin, respectively. For IS, the m/z
lengthy (≥20 min) and/or complex extraction procedures [9, transition 831.5 󳨀→ 105.2 at collision energy of 85 eV was used.
12–19, 21–24, 28, 30, 31] and relatively long analysis run time Instrument control and data acquisition and processing were
(≥10 min) [12–19, 21–24, 27, 31], making them less suitable performed with Mass Hunter Workstation software B.06.00
for routine use. HPLC-UV methods are, in general, less (Agilent Technologies, Santa Clara, USA).
sensitive and require relatively large sample volumes (≥200
𝜇L) [12–19, 21, 23, 24], limiting their applicability in pediatric 2.3. Preparation of Calibration Standards and Quality Control
studies. Such methods are applicable for TDM of RIF, but Samples. Individual stock solutions of IS and RIF (primary
not for its pharmacokinetic studies due to insufficient limits stock solution, PS) with concentration 1000 mg/L were
of quantification [12, 13, 15–19, 21–24]. Sufficient sensitivity prepared by dissolving appropriate amounts in methanol.
and efficiency can be achieved with methods based on LC- RIF secondary, tertiary, and quaternary stock solutions were
MS/MS. However, the reported LC-MS/MS methods are prepared with appropriate PS dilutions with a mixture of
mostly associated with lack of isotopically labelled inter- methanol and water (1:1, v/v) to obtain solutions with 500,
nal standard [9, 28–31], persistent carry-over effect [29], 50, and 5 mg/L, respectively. These four stock solutions were
and inadequately implemented matrix effect determination, further diluted with the same solvent to obtain eight working
mostly in terms of lack of relative matrix effect determination solutions in the range 0.1-800 mg/L. Aliquots (50 𝜇L) of the
[9, 10, 25, 26, 28–31]. individual working solutions were diluted 20-fold with blank
Therefore, the aim of this work was to develop and plasma to obtain calibration standards with the following
validate a simple, fast, and sensitive LC-MS/MS method for RIF concentrations: 5, 25, 50, 100, 1000, 5000, 15000, and
rifampicin quantification in human plasma. The proposed 40000 𝜇g/L. Quality control (QC) samples were prepared on
method is undoubtedly appropriate for routine use in both six levels (15, 75, 750, 3000, 10000, and 30000 𝜇g/L) from
TDM and pharmacokinetic studies, which was confirmed on freshly prepared RIF PS, following the same protocol as for
340 samples from 56 different patients. calibration standards.
The extraction procedure combined protein precipitation
2. Experimental with solid-phase extraction (SPE) and was as follows: a 100
𝜇L aliquot of plasma sample was mixed with 300 𝜇L of ice
2.1. Chemicals and Reagents. Rifampicin (>97.0%) was pur- cold acetonitrile containing IS (2.5 mg/L). After precipitation,
chased from Fluka, Honeywell (Morris Plains, NJ, USA). the sample mixture was filtered through the Captiva ND
Stable-isotope labelled rifampicin D8, used as internal Lipids filtration plate (Agilent Technologies, Santa Clara,
standard (IS) was purchased from Alsachim (Illkirch- USA) using vacuum manifold. The collected filtrates were
Graffenstaden, France). HPLC grade methanol and acetoni- transferred to the autosampler and subjected to LC-MS/MS
trile, as well as LC-MS grade acetonitrile, were obtained analysis. Unless otherwise noted, RIF concentration was
from Honeywell (Morris Plains, NJ, USA). Formic acid was determined based on RIF/IS ratio.
obtained from Merck (Darmstadt, Germany). Milli-Q water
was generated by an Advantage A10 water purification system 2.4. Sample Collection and Preparation. A significant number
(Millipore Corp., Billerica, USA). of blood samples (340) were obtained from 56 patients,
receiving 450 mg RIF every 12 hours. The study was approved
2.2. Instrumentation and Chromatographic Conditions. The by the Republic of Slovenia National Medical Ethics Com-
pretreated samples were analyzed by an Agilent 1290 Infinity mittee (approval no. 48/06/11). Samples were collected at
liquid chromatograph coupled to 6460 Triple Quadrupole different predetermined time points schedule according to
detector (Agilent Technologies, Santa Clara, USA) equipped the site of the infection and were properly treated to obtain
with a Jetstream electrospray interface. The chromatographic plasma samples. Blank plasma samples were obtained from
separation was achieved on a core-shell Kinetex C18 column six healthy volunteers. Plasma samples were stored at -
(50 × 2.1 mm, 2.6 𝜇m) (Phenomenex, Torrance, USA) at 50∘ C, 80∘ C and thawed prior to analysis. The sample preparation
by gradient elution using 0.1% formic acid in Milli-Q water procedure was exactly the same as for the calibrator and QC
(mobile phase A) and 99.8% acetonitrile, 0.2% Milli-Q water samples. All patient samples were processed in at least two
(mobile phase B) with the following gradient (min, % B, flow parallels.
rate in mL/min): (0.0, 30, 0.800), (0.25, 35, 0.750), (0.5, 45,
0.650), (1.0, 55, 0.500), (1.25, 65, 0.500), (1.5-1.7, 95, 0.800), 2.5. Method Validation. The proposed method was validated
(1.9,30, 0.900), and (2.00, 30, 0.800). The injection volume according to the EMA guideline on bioanalytical method
International Journal of Analytical Chemistry 3

validation [34]. Recovery, relative, and absolute matrix effect mean RIF concentration at each QC level of less than 15%
were evaluated in accordance with Matuszewski et al. [35, 36]. were considered acceptable.
Method selectivity was assessed by the analysis of plasma
samples from six different lots, which were evaluated for 3. Results and Discussion
interference with RIF or IS. Furthermore, RIF identity was
confirmed when the deviation of qualifier to quantifier ion 3.1. Method Optimization. Mass spectrometry ion source
ratio was less than 20%. conditions were optimized to achieve the highest possible
The lower limit of quantification (LLOQ) was determined responses for both RIF and its isotopically labelled IS (RIF-
as the lowest calibration standard, which reached acceptable D8). The optimal MRM transitions were obtained by auto-
accuracy (100 ± 20%) and precision (RSD below 20%). mated method development software (MassHunter Opti-
Additionally, the RIF signal in LLOQ was checked to be at mizer, Agilent Technologies). The most abundant fragment
least 5-fold greater compared to the response from blank ion was chosen as quantifier, and the second most abundant
sample. as qualifier; the ratio of qualifier to quantifier ions was used
Method linearity was evaluated on eight nonzero calibra- to confirm the RIF identity. For IS, only one mass transition
tion standards in the concentration range from 5 to 40000 was used in order to increase the sensitivity of the method
𝜇g/L for three consecutive days of the validation. Linearity towards RIF.
was determined based on the least-square linear regression. The chromatographic method utilizing a core-shell C18
The acceptance criterion was a correlation coefficient of more column was optimized to provide fast and robust RIF separa-
than 0.99. tion from background matrix with low system back-pressure,
Carry-over was assessed by injecting blank sample after making it suitable not only for UHPLC but for ordinary
the highest calibration standard (40000 𝜇g/L) and high HPLC systems as well.
concentration samples. The acceptance criterion was carry- Sample preparation procedure for RIF quantitation was
over of less than 20% of the LLOQ and 5% for the IS. based on a method described by Srivastava et al. [31], which
Accuracy and precision of the method were determined was further optimized in terms of substantial time reduction,
on QC samples at six concentration levels covering the which is especially important when large sample batches
calibration range and were prepared fresh on each of the need to be analyzed. Based on literature findings [37], as
three consecutive days of the validation. Within-run accuracy well as previous practical experience with plasma protein
and precision were evaluated on QC samples prepared in precipitation, acetonitrile was chosen as the optimal organic
five replicates and analyzed in a single run. Between-run solvent. Typical plasma samples preparation procedures
accuracy and precision were evaluated on QC samples (n=6) include vigorous vortex mixing to allow protein precipitation,
prepared and analyzed on each of the three consecutive days followed by centrifugation to remove the precipitates. These
of the validation. Accuracy was acceptable when the mean conventional and most time consuming steps in bioanalysis
concentration was within ± 15% of the nominal value for all were upgraded to in-well protein precipitation and cleanup
QC samples, except at the LLOQ, where ± 20% deviation was by Captiva ND Lipids filtration plate. With the introduction
accepted. Acceptance criterion for precision, expressed as the of this fast and simplified workflow, sample preparation
coefficient of variation (CV), was ± 15% for the QC samples, time was reduced from approximately 30 min to 1 minute
except at the LLOQ (± 20%). per sample, with high recovery and lipid matrix removal.
Recovery, absolute, and relative matrix effect were eval- Such sample preparation is cost-effective and can be fully
uated on four QC levels (15, 750, 10000, and 30000 𝜇g/L) automated for high-throughput sample analysis.
each prepared in triplicate. Relative matrix effect (RME) was
determined on QC samples prepared from blank plasma 3.2. Method Validation. Method selectivity was confirmed as
from six individual donors and was expressed as CV (%) no interference from endogenous compounds was observed
of RIF responses at each concentration level. CV (%) value at the retention time of RIF or IS in the individual blank
of standard line slopes obtained from six different plasma plasma from six different sources. Representative chro-
lots was also calculated. Absolute matrix effect (AME) was matograms of six blank plasma samples and LLOQ sample
determined at each QC concentration level as the ratio are presented in Figure 1. Rifampicin qualifier to quantifier
between blank plasma spiked after extraction with RIF and ion intensity ratio (average 0.69) was within the acceptance
IS (B) and standard RIF and IS solution in neat solvent (A): criterion (± 20%) in all tested samples.
AME = B/A × 100%. Recovery (RE) was determined as the Linearity of the method was confirmed over a very wide
ratio of RIF responses of extracted QC samples and blank concentration range: 5-40000 𝜇g/L, which completely covers
plasma spiked after extraction with the same nominal RIF the therapeutic range of rifampicin (8000-24000 𝜇g/L). The
and IS concentration as in the QC samples (B): RE = QC/B × obtained calibration curve from the first validation day,
100%. plotted as response ratio of RIF to IS on y-axis versus RIF
RIF stability was evaluated on four QC levels (15, 750, concentration on the x-axis of using weighting of 1/c, was
10000, and 30000 𝜇g/L), each prepared in triplicate. Stability y=0.003220x-0.004509. The corresponding determination
studies included freeze and thaw stability (after three cycles), coefficient (r2 ) was 0.9993. The method’s LLOQ was the
short-term stability (after 4h at room temperature) of RIF in lowest calibration standard (5 𝜇g/L) based on precision, accu-
human plasma, and autosampler stability of RIF in extracted racy, and LLOQ/blank sample signal ratio. The method was
samples (reanalyzed after 24 hours at 4∘ C). Deviations in therefore found sufficiently sensitive for RIF determination
4 International Journal of Analytical Chemistry

Table 1: Accuracy and precision data for RIF.

Accuracy (%) Precision (CV%)

QC (𝜇g/L)
Within-run Between-run Within-run Between-run
15 105.04 100.47 1.95 6.70
75 99.23 97.93 1.25 3.95
750 87.21 88.67 0.63 7.92
3000 98.43 98.09 1.27 2.35
10000 98.23 101.80 3.52 6.43
30000 103.87 106.45 2.97 5.32

+ESI MRM Frag=80.0V [email protected] (823.4 -> 163.1) 2UM 5.d Table 2: Relative matrix effect data for RIF.
×102 Noise (Peak to Peak) = 17.56; SNR (1.09min) = 68.4
5 1.09 Relative matrix effect (CV%)
4.8 1201
4.6 QC (𝜇g/L) IS correction Without IS correction
4.4
4.2 15 4.75 12.00
4
3.8 750 2.97 13.34
3.6
3.4 10000 3.26 11.03
3.2
3 30000 3.65 2.66
2.8
2.6 slope 2.71 2.94
2.4
2.2
2
1.8 Table 3: Stability data for RIF.
1.6
1.4
1.2 Stability (%)
1
0.8 QC (𝜇g/L) Freeze and thaw Short-term Autosampler
0.6
0.4 15 3.03 5.06 -9.88
0.2
0 750 -10.23 -1.89 N.D.
0.9 0.95 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 10000 3.00 -3.80 N.D.
Counts vs. Acquisition Time (min) 30000 1.61 2.24 -2.18
Figure 1: Representative LC-MS/MS chromatograms of six blank N.D. = not determined
plasma samples and LLOQ sample. Retention time of RIF is 1.1 min.

essential in order to obtain reliable and useful results for

further pharmacokinetic analysis. Therefore, relative matrix
in bioequivalence and pharmacokinetic studies. The wide effect was quantitatively assessed as proposed by Matuszewski
concentration range enables RIF quantification in various et al. [35]. The obtained results for relative matrix effect
samples without the need of any additional sample processing (CV <5%) in combination with the low CV of standard
(e.g., dilution). line slopes obtained from six different plasma lots (Table 2)
Carry-over effect has previously been reported as a trou- are an excellent demonstration of the absence of relative
blesome part of RIF LC-MS/MS analysis [29, 38]. However, matrix effect. The obtained average absolute matrix effect ±
the analysis of blank solutions immediately after the highest SE for RIF/IS ratio was 101.89 ± 1.95%. Based on the obtained
calibration standard and high concentration samples revealed results for both relative and absolute matrix effect, it can be
no quantifiable carry-over. concluded that the proposed method is free from any major
The obtained results for within-run and between-run ion suppression or enhancement. The mean recoveries (RE ±
accuracy and precision are within the acceptance criterion, SE) for both RIF (92.48 ± 3.97%) and IS (88.39 ± 3.07%) were
as presented in Table 1. consistent and reproducible, thus confirming the suitability
The evaluation of recovery and matrix effect is a very of the extraction procedure.
important part of bioanalytical LC-MS/MS method valida- The results for freeze and thaw stability, short-term
tion. It is, however, somewhat neglected or undervalued stability, and autosampler stability, expressed as % of change
aspect, due to incomplete experimental protocol by the from the initial sample, are summarized in Table 3. The
FDA and EMA guidelines in terms of relative matrix effect stability of RIF in all QC samples was acceptable (± 15%
evaluation [34, 39]. The latter is a crucial parameter in bioan- change) under all tested conditions. Long-term stability was
alytical assays validation. The greatest drawback of previously not assessed since RIF has previously been shown stable
reported methods is RIF quantification in human plasma in plasma stored at -20∘ C for at least 1 month [9] and
from different patients, without appropriate demonstration for 4 months after storage at -80∘ C [2]. Patients’ plasma
that the relative MS/MS response is not affected by the matrix. samples were stored at -80∘ C and were analyzed as soon as
Elimination of relative matrix effect uncertainty is, thus, possible.
International Journal of Analytical Chemistry 5

×104 +ESI MRM Frag=170.0V [email protected] (831.5 -> 105.2) 25R. 13-4 II.d
12 3
4

0
×102 +ESI MRM Frag=80.0V [email protected] (823.4 -> 107.1) 25R. 13-4 II.d
12 3
8

0
×103 +ESI MRM Frag=80.0V [email protected] (823.4 -> 163.1) 25R. 13-4 II.d
1.4 3
12
1.2
1
0.8
0.6
0.4
0.2
0
0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
Counts vs. Acquisition Time (min)

Figure 2: LC-MS/MS chromatogram of a real plasma sample from a patient, receiving 450 mg RIF every 12 hours. The determined RIF
concentration is 28.3 𝜇g/L. Top trace represents IS; middle trace represents RIF qualifier ion, and bottom trace represents RIF quantifier ion.

3.3. Application and Prospect of the Method. The applicability limited. This assay is therefore undoubtedly applicable for
of the proposed method was confirmed on 340 plasma pharmacokinetic, bioequivalence, or bioavailability studies of
samples from 56 patients with orthopedic endoprosthesis RIF, as well as for TDM.
infections, receiving 450 mg RIF every 12 hours. The deter-
mined RIF concentrations ranged between 5.92 and 21447
𝜇g/L. Calculated mean RIF concentration (with standard 4. Conclusions
deviation) in the assayed samples with RIF concentrations
The developed LC-MS/MS method was fully validated in
above LLOQ was found 2433 ± 3324 𝜇g/L. An example of
terms of selectivity, linearity, accuracy, precision, dilution
a real plasma sample chromatogram, with determined RIF
integrity, carry-over, recovery, matrix effect, and stability. Its
concentration 28.3 𝜇g/L, is shown in Figure 2. The analysis of
suitability for routine analyses was confirmed by rifampicin
such a number of plasma samples was facilitated by the fast
quantification in 340 patients’ plasma samples. The fast,
and simple sample preparation, which took approximately
simple, and efficient sample preparation, along with the short
one minute per sample. Sample preparation time can be
LC-MS/MS run time, enables the analysis of a large number
further reduced by the use of a fully automated approach.
of plasma samples in a short time, thus providing a fast,
This, in combination with the fast LC-MS/MS method with
reliable, and cost-effective analytic tool for RIF therapeutic
total run time of 2.4 min, is the most important advantage
monitoring and pharmacokinetic studies.
compared to other reported methods for RIF quantification.
Additional advantages include low LLOQ and wide analytical
range, making it suitable for routine analysis of plasma sam- Data Availability
ples with diverse RIF concentrations. Due to the small plasma
volume (100 𝜇L), the proposed method is also suited for The data used to support the findings of this study are
pediatric studies, where sample volumes are commonly quite included within the article.
6 International Journal of Analytical Chemistry

Conflicts of Interest [12] S. Goutal, S. Auvity, T. Legrand et al., “Validation of a simple

HPLC-UV method for rifampicin determination in plasma:
The authors declare that there are no conflicts of interest Application to the study of rifampicin arteriovenous concen-
regarding the publication of this article. tration gradient,” Journal of Pharmaceutical and Biomedical
Analysis, vol. 123, pp. 173–178, 2016.

Acknowledgments [13] Z. Zhou, L. Chen, P. Liu, M. Shen, and F. Zou, “Simultaneous

Determination of Isoniazid, Pyrazinamide, Rifampicin and
This research was financially supported by Slovenian Acetylisoniazid in Human Plasma by High-Performance Liquid
Research Agency (ARRS) [P1-0189]. Chromatography,” Analytical Sciences, vol. 26, no. 11, pp. 1133–
1138, 2010.
[14] R. Panchagnula, A. Sood, N. Sharda, K. Kaur, and C. Kaul,
References “Determination of rifampicin and its main metabolite in plasma
and urine in presence of pyrazinamide and isoniazid by HPLC
[1] WHO (World Health Organization), Global Tuberculosis Report
method,” Journal of Pharmaceutical and Biomedical Analysis,
2018, Geneva, Switzerland, 2018, http://www.who.int/tb/publi-
vol. 18, no. 6, pp. 1013–1020, 1999.
cations/global report/tb18 ExecSum web 4Oct18.pdf?ua=1.
[15] L. Melo, R. Queiroz, and M. Queiroz, “Automated determina-
[2] K. Govender, J. H. Adamson, and P. Owira, “The development
tion of rifampicin in plasma samples by in-tube solid-phase
and validation of a LC-MS/MS method for the quantitation of
microextraction coupled with liquid chromatography,” Journal
metformin, rifampicin and isoniazid in rat plasma using HILIC
of Chromatography B, vol. 879, no. 24, pp. 2454–2458, 2011.
chromatography,” Journal of Chromatography B, vol. 1095, pp.
127–137, 2018. [16] D. Fox, R. O’Connor, P. Mallon, and G. McMahon, “Simulta-
[3] R. Jogi and S. K. Tyring, “Therapy of nontuberculous mycobac- neous determination of efavirenz, rifampicin and its metabolite
terial infections,” Dermatologic Therapy, vol. 17, no. 6, pp. 491– desacetyl rifampicin levels in human plasma,” Journal of Phar-
498, 2004. maceutical and Biomedical Analysis, vol. 56, no. 4, pp. 785–791,
2011.
[4] T. F. Moriarty, R. Kuehl, T. Coenye et al., “Orthopaedic device-
related infection: current and future interventions for improved [17] B. Louveau, C. Fernandez, N. Zahr et al., “Determination of
prevention and treatment,” EFORT Open Reviews, vol. 1, no. 4, rifampicin in human plasma by high-performance liquid chro-
pp. 89–99, 2016. matography coupled with ultraviolet detection after automa-
tized solid-liquid extraction,” Biomedical Chromatography, vol.
[5] P. Donald, J. Maritz, and A. Diacon, “The pharmacokinetics
30, no. 12, pp. 2009–2015, 2016.
and pharmacodynamics of rifampicin in adults and children in
relation to the dosage recommended for children,” Tuberculosis, [18] A. Allanson, M. Cotton, J. Tettey, and A. Boyter, “Determination
vol. 91, no. 3, pp. 196–207, 2011. of rifampicin in human plasma and blood spots by high
performance liquid chromatography with UV detection: A
[6] L. Aı̈t Moussa, O. El Bouazzi, S. Serragui, D. Soussi Tanani,
potential method for therapeutic drug monitoring,” Journal of
A. Soulaymani, and R. Soulaymani, “Rifampicin and isoniazid
Pharmaceutical and Biomedical Analysis, vol. 44, no. 4, pp. 963–
plasma concentrations in relation to adverse reactions in tuber-
969, 2007.
culosis patients: a retrospective analysis,” Therapeutic Advances
in Drug Safety, vol. 7, no. 6, pp. 239–247, 2016. [19] A. Walubo, P. Smith, and P. I. Folb, “Comprehensive assay
[7] F. Fahimi, P. Tabarsi, F. Kobarfard et al., “Isoniazid, rifampicin for pyrazinamide, rifampicin and isoniazid with its hydrazine
and pyrazinamide plasma concentrations 2 and 6 h post dose metabolites in human plasma by column liquid chromatog-
in patients with pulmonary tuberculosis,” The International raphy,” Journal of Chromatography B: Biomedical Sciences and
Journal of Tuberculosis and Lung Disease, vol. 17, no. 12, pp. 1602– Applications, vol. 658, no. 2, pp. 391–396, 1994.
1606, 2013. [20] M. Ishii and H. Ogata, “Determination of rifampicin and its
[8] L. Van Tongeren, S. Nolan, V. J. Cook, J. M. FitzGerald, and J. main metabolites in human plasma by high-performance liquid
C. Johnston, “Therapeutic drug monitoring in the treatment of chromatography,” Journal of Chromatography B: Biomedical
tuberculosis: a retrospective analysis,” The International Journal Sciences and Applications, vol. 426, pp. 412–416, 1988.
of Tuberculosis and Lung Disease, vol. 17, no. 2, pp. 221–224, 2013. [21] H. Seifart, P. Kruger, D. Parkin, P. van Jaarsveld, and P. Donald,
[9] S. Gao, Z. Wang, X. Xie et al., “Rapid and sensitive method “Therapeutic monitoring of antituberculosis drugs by direct in-
for simultaneous determination of first-line anti-tuberculosis line extraction on a high-performance liquid chromatography
drugs in human plasma by HPLC-MS/MS: Application to system,” Journal of Chromatography B: Biomedical Sciences and
therapeutic drug monitoring,” Tuberculosis, vol. 109, pp. 28–34, Applications, vol. 619, no. 2, pp. 285–290, 1993.
2018. [22] B. Ratti, R. R. Parenti, A. Toselli, and L. F. Zerilli, “Quan-
[10] K. H. Hee, J. J. Seo, and L. S. Lee, “Development and validation titative assay of rifampicin and its main metabolite 25-
of liquid chromatography tandem mass spectrometry method desacetylrifampicin in human plasma by reversed-phase high-
for simultaneous quantification of first line tuberculosis drugs performance liquid chromatography,” Journal of Chromatogra-
and metabolites in human plasma and its application in clinical phy B: Biomedical Sciences and Applications, vol. 225, no. 2, pp.
study,” Journal of Pharmaceutical and Biomedical Analysis, vol. 526–531, 1981.
102, pp. 253–260, 2015. [23] P. J. Smith, J. van Dyk, and A. Fredericks, “Determination of
[11] G. Maartens, M. Boffito, and C. W. Flexner, “Compatibility of rifampicin, isoniazid and pyrazinamide by high performance
next-generation first-line antiretrovirals with rifampicin-based liquid chromatography after their simultaneous extraction from
antituberculosis therapy in resource limited settings,” Current plasma,” International Journal of Tuberculosis and Lung Disease,
Opinion in HIV and AIDS, vol. 12, no. 4, pp. 355–358, 2017. vol. 3, no. 11, Suppl 3, Article ID 10593712, pp. S325–S328, 1999.
International Journal of Analytical Chemistry 7

[24] M. S. Balbão, C. Bertucci, M. M. Bergamaschi et al., “Rifampicin chromatography–tandem mass spectrometry,” Journal of Chro-
determination in plasma by stir bar-sorptive extraction and liq- matography B, vol. 785, no. 2, pp. 263–275, 2003.
uid chromatography,” Journal of Pharmaceutical and Biomedical [38] D. Vu, R. Koster, A. Wessels, B. Greijdanus, J. Alffenaar, and D.
Analysis, vol. 51, no. 5, pp. 1078–1083, 2010. Uges, “Troubleshooting carry-over of LC–MS/MS method for
[25] H. Kim, K. Seo, H. Kim et al., “Simple and accurate quanti- rifampicin, clarithromycin and metabolites in human plasma,”
tative analysis of 20 anti-tuberculosis drugs in human plasma Journal of Chromatography B, vol. 917-918, pp. 1–4, 2013.
using liquid chromatography–electrospray ionization–tandem [39] U.S. FDA, Guidance for Industry, Bioanalytical Method
mass spectrometry,” Journal of Pharmaceutical and Biomedical Validation, 2018, http://www.fda.gov/Drugs/GuidanceCompli-
Analysis, vol. 102, pp. 9–16, 2015. anceRegulatoryInformation/Guidances/default.htm.
[26] R. Hartkoorn, S. Khoo, D. Back et al., “A rapid and sensitive
HPLC–MS method for the detection of plasma and cellular
rifampicin,” Journal of Chromatography B, vol. 857, no. 1, pp. 76–
82, 2007.
[27] W. A. Korfmacher, J. Bloom, M. I. Churchwell, E. B. Hansen,
T. A. Getek, and K. T. McManus, “Characterization of three
rifamycins via electrospray mass spectrometry and hplc-
thermospray mass spectrometry,” Journal of Chromatographic
Science (JCS), vol. 31, no. 12, pp. 498–501, 1993.
[28] L. Baietto, A. D’Avolio, F. G. De Rosa et al., “Development
and validation of a simultaneous extraction procedure for
HPLC-MS quantification of daptomycin, amikacin, gentamicin,
and rifampicin in human plasma,” Analytical and Bioanalytical
Chemistry, vol. 396, no. 2, pp. 791–798, 2010.
[29] F. de Velde, J. C. Alffenaar, A. M. Wessels, B. Greijdanus, and
D. R. Uges, “Simultaneous determination of clarithromycin,
rifampicin and their main metabolites in human plasma by
liquid chromatography–tandem mass spectrometry,” Journal of
Chromatography B, vol. 877, no. 18-19, pp. 1771–1777, 2009.
[30] L. Baietto, A. Calcagno, I. Motta et al., “A UPLC-MS-MS
method for the simultaneous quantification of first-line anti-
tuberculars in plasma and in PBMCs,” Journal of Antimicrobial
Chemotherapy, vol. 70, no. 9, pp. 2572–2575, 2015.
[31] A. Srivastava, D. Waterhouse, A. Ardrey, and S. A. Ward, “Quan-
tification of rifampicin in human plasma and cerebrospinal fluid
by a highly sensitive and rapid liquid chromatographic–tandem
mass spectrometric method,” Journal of Pharmaceutical and
Biomedical Analysis, vol. 70, pp. 523–528, 2012.
[32] J. B. Prahl, M. Lundqvist, J. M. Bahl et al., “Simultaneous
quantification of isoniazid, rifampicin, ethambutol and pyrazi-
namide by liquid chromatography/tandem mass spectrometry,”
APMIS-Acta Pathologica, Microbiologica et Immunologica Scan-
dinavica, vol. 124, no. 11, pp. 1004–1015, 2016.
[33] S. H. Song, S. H. Jun, K. U. Park et al., “Simultaneous
determination of first-line anti-tuberculosis drugs and their
major metabolic ratios by liquid chromatography/tandem mass
spectrometry,” Rapid Communications in Mass Spectrometry,
vol. 21, no. 7, pp. 1331–1338, 2007.
[34] Committee for Medicinal Products for Human Use
(CHMP), Guideline on Bioanalytical Method Validation,
European Medicines Agency (EMA), London, UK, 2011,
https://www.ema.europa.eu/contact.
[35] B. Matuszewski, “Standard line slopes as a measure of a relative
matrix effect in quantitative HPLC–MS bioanalysis,” Journal of
Chromatography B, vol. 830, no. 2, pp. 293–300, 2006.
[36] B. K. Matuszewski, M. L. Constanzer, and C. M. Chavez-Eng,
“Strategies for the assessment of matrix effect in quantitative
bioanalytical methods based on HPLC-MS/MS,” Analytical
Chemistry, vol. 75, no. 13, pp. 3019–3030, 2003.
[37] C. Polson, P. Sarkar, B. Incledon, V. Raguvaran, and R.
Grant, “Optimization of protein precipitation based upon
effectiveness of protein removal and ionization effect in liquid