Вы находитесь на странице: 1из 9

Research Article

For reprint orders, please contact reprints@future-science.com

Quantification of amikacin and kanamycin

in serum using a simple and validated
LC–MS/MS method

Background: Amikacin and kanamycin are frequently used in the treatment of Jacob A Dijkstra1, Marieke
multidrug-resistant TB. The current commercially available immunoassay is unable GG Sturkenboom1, Kai van
to analyze kanamycin and trough levels of amikacin. The objective was therefore to Hateren1, Remco A Koster1,
develop a LC–MS/MS method for the quantification of amikacin and kanamycin in Ben Greijdanus1 & Jan-Willem
C Alffenaar*,1
human serum. Materials & methods: Using apramycin as internal standard, selectivity, 1
University of Groningen, University
accuracy, precision, recovery, matrix effects and stability were evaluated. Results: The Medical Center Groningen, Department
presented LC–MS/MS method meets the recommendations of the US FDA with a low of Clinical Pharmacy & Pharmacology,
LLOQ of 250 ng/ml for amikacin and 100 ng/ml for kanamycin. No statistical significant PO Box 30.001, 9700 RB Groningen,
difference was found between the LC–MS/MS method and the immunoassay of The Netherlands
*Author for correspondence:
amikacin (Architect® assay, p = 0.501). Conclusion: The low LLOQ of amikacin and the
Tel.: +31 503 614 070
ability to analyze kanamycin makes the LC–MS/MS method the preferred method for Fax: +31 503 614 087
analyzing these aminoglycosides. j.w.c.alffenaar@umcg.nl

Background with human subjects showed a significant

TB is an infectious disease that kills almost correlation between the average trough
2 million patients every year [1] . Multidrug- level (Cmin) of aminoglycosides and audi-
resistant TB (MDR-TB) is caused by strains tory toxicity [9] . This effect could be more
of Mycobacterium tuberculosis resis- important in patients with TB, whom are
tant to at least two first-line drugs (rifampin often treated for months with aminoglyco-
and isoniazid) [1,2] . Annually, 500,000 sides. Although evidence is limited, moni-
patients are infected with an MDR-TB strain toring and minimizing trough levels seems
[1] . The aminoglycosides amikacin and justifiable.
kanamycin are both frequently used for the Therefore, both Cmax and Cmin are impor-
treatment of MDR-TB [1,2] , since minimal tant parameters in therapeutic drug
inhibitory concentrations (MIC) range from monitoring (TDM) of aminoglycosides.
1 to 4 mg/l [3] . However, a reliable, robust TDM may help to optimize the dose if mea-
method that does not require extensive sam- sured serum concentrations are not within
ple pre-processing to measure both amino- the desired therapeutic range [10] . To enable
glycosides in plasma or serum has not been TDM, a specific and sensitive method of
available until now. analysis is required. There is no commer-
The effective pharmacodynamic param- cially available immunoassay for the quan-
eter for aminoglycosides is the peak serum tification of kanamycin. Furthermore, the
concentration (Cmax) over MIC [4] . To reduce immuno­assay available for amikacin has a
the emergence of resistance and to optimize high LLOQ of 1500 ng/ml.
the efficacy, a Cmax /MIC ratio of ≥8 is pre- In addition, the available immunoassay is
ferred [3–5] . Common side effects associated less versatile than newer methods of analysis,
with aminoglycosides therapy are ototoxic- such as LC–MS/MS. The Stop TB Partner-
ity and nephrotoxicity, both of which are ship proposed in the ‘Global Plan to Stop
possibly related to treatment duration and TB 2011–2015’ that laboratory strength-
part of
cumulative dose [6–8] . Furthermore, a study ening is one of the key components in the

10.4155/BIO.14.191 © 2014 Future Science Ltd Bioanalysis (2014) 6(16), 2125–2133 ISSN 1757-6180 2125
Research Article  Dijkstra, Sturkenboom, van Hateren, Koster, Greijdanus & Alffenaar

Key terms Aldrich). Water was purified with a Milli-Q system

(Millipore Corporation, MA, USA). Buffer solution
Mycobacterium tuberculosis: Causes TB, an infectious consisted of 1% HFBAA in purified water. The anhy-
disease that is responsible for the death of almost 2 million
patients every year.  dride formulation was already available in our labo-
ratory. Trichloroacetic acid was used in sample pro-
Aminoglycosides: Antibacterial drugs that inhibit protein
cessing and was purchased from Merck (NJ, USA).
synthesis and are used for a variety of infectious diseases. 
Eluent for liquid chromatography consisted of buffer
Amikacin: An aminoglycoside effective against solution, purified water and methanol (LiChrosolv ®,
Mycobacterium tuberculosis. 
Merck). IS stock solution consisted of 5000 ng/ml
Kanamycin: An aminoglycoside effective against apramycin (Sigma Aldrich) in HFBAA (1%).
Mycobacterium tuberculosis.  To 100 μl serum, 50 μl trichloroacetic acid and
Therapeutic drug monitoring: Optimizing the dose 50  μl IS stock solution was added. After each addi-
based on drug concentrations to ensure efficacy and tion, the sample was vortexed for 1 min. The subse-
prevent side effects.  quent sample was centrifuged (5 min at 11,000 rpm)
and the supernatant was transferred into a clear glass
fight against TB [11] . With one single LC–MS/MS, vial with a silicone/polytetrafluoroethylene septum. A
TDM could be performed for many drugs, improv- tandem mass spectrometer was used to perform the
ing efficacy and reducing toxicity, thereby hopefully analysis. The mass spectrometer was a Finnigan™
increasing compliance. TSQ Quantum Discovery MAX™ (TSQ Quantum,
Literature about analysis of amikacin and kanamy- Thermo Fisher, CA, USA), supplied with a Finni-
cin in serum using LC–MS/MS is scarcely available gan™ Surveyor™ MS Pump Plus and a Finnigan™
[12,13] . A published LC–MS/MS method used solid- Autosampler Plus. Positive electron spray ioniza-
phase extraction (SPE) with different types of col- tion was performed at 3500 V. The LC system was
umns, but this was a time-consuming method with equipped with a Thermo Scientific™ HyPURITY™
run times up to 12 min [12] . Furthermore, no inter- C18 5.0 × 2.1 mm column with a particle size of 3 μm.
nal standard (IS) was used. The authors used only Separation took place with gradient elution and a flow
external calibrations [12] . of 300 μl/min at approximately 100 bar. The gradient
Baietto et al. developed a sample preparation elution used is shown in Table 1.
method without SPE for LC–MS/MS [13] . Unfortu- The autosampler was configured to an injection
nately, this method was not validated for kanamy- volume of 5 μl and a tray temperature of 10°C. Nitro-
cin. Moreover, this method requires an additional gen was used as sheath and auxiliary gas, and argon
dilution step and used quinoxaline as the IS, which was used as collision gas. Sheath gas pressure was set
is not structurally related to amikacin or kanamycin to 35 arbitrary units and auxiliary gas to 5 arbitrary
[13] . Therefore, both described methods are less able units. Used mass transitions are shown in Table 2.
to compensate for sample and matrix variability [14] . The immunoassay analysis for amikacin was per-
A recent study described the simultaneous analysis formed using a validated immunoassay method (Archi-
of nine second-line anti-TB drugs [15] . Unfortunately, tect, Abbott Diagnostics, IL, USA). This method was
amikacin was not included in the method valida- validated based on the US FDA guidelines [16] . The
tion. Furthermore, the LLOQ of kanamycin was LLOQ of this method was 1500 ng/ml. Accuracy and
2500 ng/ml and therefore relatively high. precision was determined on four concentration levels
To enable the use of TDM in therapy with ami- (11,500, 6500, 15,000 and 27,500 ng/ml). The bias
kacin or kanamycin, a suitable, sensitive and robust varied from -1.5 to 6.7% over all concentration levels.
method is required to analyze amikacin and kanamy- Within-run coefficient of variation (CV) varied from
cin over a wide concentration range. The objective of 0.5 to 8.5% and between-run from 0.0 to 2.4%.
this study is to develop and validate a simple and fast
LC–MS/MS method for the quantification of ami- Analytical method validation
kacin and kanamycin in human serum for routine The method was validated based on the ‘Guidance
patient care and clinical studies. for Industry Bioanalytical Method Validation’ pub-
lished by the FDA [16] . Since the FDA guidelines lack
Material & methods requirements on the stability, a maximum bias of 15%
Analysis was used according to the EMA guidelines [17] .
Kanamycin and amikacin were purchased from Sigma The calibration line consisted of eight concentration
Aldrich (MO, USA). Heptafluorobutyric acid anhy- levels: 250, 500, 2500, 5000, 10,000, 15,000, 20,000
dride (HFBAA) was obtained from Fluka (Sigma and 25,000 ng/ml for amikacin and 100, 500, 2500,

2126 Bioanalysis (2014) 6(16) future science group

Quantification of amikacin & kanamycin using LC–MS/MS Research Article

5000, 10,000, 15,000, 20,000 and 25,000 ng/ml for Table 1. Gradient elution.
kanamycin. The upper limit was chosen as the lin-
earity of the calibration line was not sufficient above Min  Heptafluorobutyric acid Water (%)  Methanol (%) 
anhydride, % (1% in water) 
25,000 ng/ml. Stock standards and working solu-
tions were diluted with serum to achieve the desired 0  5 77.5  17.5 
concentrations. Samples with concentrations above 0.50  5 50  45 
25,000 ng/ml were diluted to obtain a concentration 1.20  5 43  52 
within the limits of the calibration line. Four QC
5.51  5 77.5  17.5 
samples were used at LLOQ (100 and 250 ng/ml for
kanamycin and amikacin), LOW (500 ng/ml), MED 6.00  5 77.5  17.5 
(10,000 ng/ml) and HIGH (20,000 ng/ml). Cali-
bration lines, correlation coefficients and regression difference with the nominal concentration. Short-term
coefficients were calculated using one-way ANOVA. stability over 24 h was tested at room temperature.
Concentrations of the samples were calculated based The post-preparative stability was tested by analyzing
on the peak height ratio with the IS with the compiled spiked samples that had been placed in the autosam-
formula. pler for 24 h. Concentrations were compared with
Selectivity was determined by analyzing six serum calibration standards, which were freshly prepared on
samples, each obtained from a different pool of serum. the day of the analysis. The mean of each stability test
The extent of ion suppression or enhancement was should be within 15% of the nominal concentration,
tested, during a constant infusion of amikacin or according to the EMA guidelines [17] . To assess the
kanamycin with IS, by injecting six pooled serum dilution integrity, a solution of 25,000 ng/ml amika-
samples. Accuracy was evaluated by replicate analy- cin and kanamycin in serum was diluted in tenfold
sis of the QC samples at all four concentration levels with serum to a 2500 ng/ml solution and subsequently
(LLOQ, LOW, MED and HIGH) on three consecu- analyzed on three subsequent days.
tive days. The mean difference between the nominal
concentration and the experimental result should be Comparison between immunoassay
<15% (LLOQ <20%) [16] . Precision was determined & LC–MS/MS methods
using the measurements originating from the accu- To compare both methods, 17 clinical samples were
racy evaluation. The CV was calculated at each indi- measured with the new LC–MS/MS method and
vidual concentration level. The CV should be lower with the immunoassay. Only amikacin was com-
than 15% (LLOQ <20%) [16] . pared, since the immunoassay is unable to analyze
Recovery was determined by comparing the mean kanamycin.
peak area of spiked blank serum with the peak area of
the spiked extract serum. Recovery was determined at Clinical pilot
LOW, MED and HIGH levels of both amikacin and Routine TDM for TB drugs is performed for all
kanamycin. The matrix effect was assessed by divid- patients with MDR-TB in the Tuberculosis Centre
ing the peak area of spiked extract of blank serum by Beatrixoord, University Medical Center Groningen
the peak area of spiked extraction fluid. Matrix effects (Haren, The Netherlands). Routine TDM consists of
were also determined at LOW, MED and HIGH sampling at 0, 1, 2, 3, 4 and 7 h after administration.
concentration levels of amikacin and kanamycin. Medical charts were reviewed to detect patients that
Both recovery and matrix effects were determined in received amikacin or kanamycin as part of their TB
fivefold. treatment at our TB center and that were subjected to
Blank serum was spiked with amikacin or kanamy- routine TDM. This study was evaluated by the local
cin at two concentration levels: LOW and HIGH. Sta- ethics committee (IRB 2013–2492) and was accord-
bility was determined after three freeze–thaw cycles ing to the Dutch law allowed due to its retrospective
by measuring the spiked samples and assessing the nature.

Table 2. MS/MS conditions and mass transitions.

Compound  Tube lens (V)  Collision energy (eV)  Mass transition (m/z) 
Amikacin  109  33  586.2 → 163.0 
Kanamycin  100  27  485.3 → 163.1 
Apramycin (internal standard)  109  14  540.3 → 378.2 

future science group www.future-science.com 2127

Research Article  Dijkstra, Sturkenboom, van Hateren, Koster, Greijdanus & Alffenaar

Statistics Shapiro–Wilk test). Test results from the immuno-

Validation parameters were calculated using a vali- assay, which were below the detection limit, were
dated Excel sheet (Microsoft, WA, USA). Both the test assumed to be 0. No significant difference was found
for normality (Shapiro–Wilk test) and the Wilcoxon between the outcomes of both methods (p = 0.501;
signed rank test were performed using SPSS 20 (SPSS, n = 17). However, the LLOQ of amikacin is lower
IL, USA). when the LC–MS/MS method is used compared with
the immunoassay.
Method validation Clinical pilot
Three chromatograms of the LOW concentration Full pharmacokinetic concentrations curves were
(500 ng/ml) of amikacin, kanamycin and apramy- obtained from two patients using amikacin and four
cin are displayed in Figure 1. The retention times of patients using kanamycin, as shown in Figure 2. This
amikacin, kanamycin and apramycin were deter- population consisted of five male patients and one female
mined to be 2.53, 2.62 and 3.33 min. At these reten- patient, with a mean age of 33.7 years (range:  19–67).
tion times, no other peaks are observed in thepooled The mean BMI was 20.2 (range: 13.0–26.5).
serum samples. The validated LLOQs were 250 ng/ml
for amikacin and 100 ng/ml for kanamycin. Ion sup- Discussion
pression or enhancement was tested using the constant This is the first LC–MS/MS method in the literature to
infusion test [18] . No ion suppression or enhancement analyze amikacin and kanamycin in serum with min-
was observed at the retention times of amikacin and imal sample preprocessing. This LC–MS/MS method
kanamycin as displayed in Figure 1. has been validated for accuracy, precision, recovery,
Calibration of both amikacin and kanamycin was dilution integrity, matrix effect, autosampler stability,
performed with eight concentration levels. Resulting bench top stability and freeze–thaw stability, based
calibration lines are displayed in Table 3. Accuracy was on the FDA guidelines [16] . Concerning the stabil-
determined by analyzing five different spiked samples ity, the FDA guidelines do not contain exact limits.
at the four different QC concentration levels. Calcu- Therefore, the limits of the EMA guidelines were
lated biases were -8.8 to +7.0% and -6.7 to +5.6% for applied [17] .
amikacin and kanamycin, respectively. All results are The requirements that were demanded for this
displayed in Table 4. Precision was determined using method complicated the development. In order
data originating from the accuracy evaluation. The CV to minimize the total processing time, SPE was
of the within-run and the between-run were calculated avoided and a simple sample preprocessing method
and are displayed in Table 4. The CV varied between was developed. Furthermore, gradient elution was
1.7 and 6.7% for amikacin, and between 0.0 and 4.9% optimized to shorten the time of analysis. However,
for kanamycin. this method still has an adequate LLOQ for TDM,
The recovery of amikacin and kanamycin varied and provides high sensitivity and selectivity.
between 93.6 and 98.8%, as displayed in Table 4. The LLOQs are 250 ng/ml for amikacin and
Matrix effects were determined in fivefold and varied 100 ng/ml for kanamycin. These LLOQs are compa-
between 98.5 and 103.0%. The stability of both ana- rable to a previous study (100 ng/ml) [12] , but signifi-
lytes was assessed using the above proposed method. cantly lower compared with the method developed by
Measured concentrations differed from the nominal Baietto et al. (2340 ng/ml) [13] and another recently
concentration with a maximal difference of -11.7 to published method (2500 ng/ml) [15] . Furthermore, the
+1.2% at room temperature, 1.3–7.4% in the auto­ LLOQ of amikacin was lower than the LLOQ of the
sampler and -2.6–10.8% after three freeze–thaw immunoassay. The LLOQs found in this study are
cycles. Dilution integrity was assessed on three dif- sufficient to detect any clinically relevant trough lev-
ferent days in fivefold. The bias found was -8.0% els. The ability to measure below 1500 ng/ml makes
and 5.0% for amikacin and kanamycin, respectively. this method suitable for early detection of trends in
Within-run CV was 2.6% and between-run CV varied rising trough levels and to adjust the dose accordingly.
from 1.7% to 2.9%. Another advantage of this method is that other
second-line TB drugs, such as linezolid [19] , clarithro-
Comparison between immunoassay mycin [20] and moxifloxacin [21] can be monitored using
(Architect®) & LC–MS/MS LC–MS/MS. This is in contrast to the immunoassay,
Results of both methods for quantification of amikacin which provides only the quantification of amikacin.
were compared using the Wilcoxon signed rank test, As described in the ‘Global Plan to Stop TB’, labora-
since the data was not normally distributed (p < 0.001, tory strengthening is one of the core components [11] .

2128 Bioanalysis (2014) 6(16) future science group

Quantification of amikacin & kanamycin using LC–MS/MS Research Article

Intensity (%)




Intensity (%)




Ion supression test
Intensity (%)


20 Amikacin

Intensity (%)




0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Time (min)

Figure 1. Three chromatograms of the LOW (500 ng/ml) concentration of amikacin, kanamycin and apramycin,
and the ion suppression test.
With this single LC–MS/MS method, TDM of these The sample processing method used in this study
anti-TB drugs can be performed to monitor efficacy is less time consuming than the SPE methods previ-
and possibly reduce toxicity. Further research could ously described [12] . The recoveries found in this study
elucidate how to use dried blood spots to monitor were higher than in the study of Baietto (85.2%), who
aminoglycoside concentrations, thus simplifying stor- used similar sample preprocessing [13] . Baietto’s method
age and transport of blood samples, as has already required an extra dilution step, which may introduce
been achieved for linezolid and moxifloxacin [22–25] . additional errors [13] .

future science group www.future-science.com 2129

Research Article  Dijkstra, Sturkenboom, van Hateren, Koster, Greijdanus & Alffenaar

Table 3. Calibration curves.

Compound  Slope (± SD)  Intercept (± SD)  Correlation coefficient 
Amikacin  0.761 ± 0.0168  -0.0545 ± 0.0105  0.995 
Kanamycin  0.997 ± 0.0143  0.00390 ± 0.00391  0.997 
SD: Standard deviation.

During this study, apramycin was used as the IS. may be because of the different retention time and ion-
Since apramycin is structurally related to amikacin and ization characteristics. However, the ion suppression
kanamycin (Figure 3), it is our opinion that apramycin tests showed that there was less than ±3% ion sup-
is suitable as an IS [14] . Apramycin is neither regis- pression, indicating that the developed method is free
tered within the EU nor in the USA for human use. from ion suppression and that the IS used is suitable.
A structural analog as the IS may not compensate for Identical samples were measured with the immuno-
ion suppression effects, as well as a deuterated IS. This assay (Architect) and the new LC–MS/MS method.

Table 4. Validation results of the LC–MS/MS method.

Criteria Concentration level
Nominal concentration (ng/ml)
Amikacin 250 500 10,000 20,000
Kanamycin 100 500 10,000 20,000
Accuracy (bias, %; n = 15)
Amikacin 7.0 -8.8 2.6 -1.3
Kanamycin -3.9 -6.7 5.6 -0.9
Within-day precision (CV, %; n = 15)
Amikacin 2.9 2.5 4.5 1.9
Kanamycin 4.9 2.9 4.0 2.1
Between-day precision (CV, %; n = 15)
Amikacin 4.0 1.7 4.0 6.7
Kanamycin 0.0 3.5 1.8 2.8
Recovery (%)
Amikacin NA 98.8 96.4 93.6
Kanamycin NA 98.4 97.1 93.9
Matrix effect (%)
Amikacin NA 99.6 103.0 98.9
Kanamycin NA 99.2 98.7 98.5
Autosampler stability (24 h; bias, %)
Amikacin NA 7.4 NA 5.5
Kanamycin NA 5.6 NA 1.3
Bench top stability (24 h; bias, %)
Amikacin NA -11.7 NA 1.2
Kanamycin NA 1.1 NA -0.1
Freeze–thaw stability (after three freeze–thaw cycles; bias, %)
Amikacin NA -0.1 NA -2.6
Kanamycin NA 10.8 NA -2.1
n = 15 means five replicate analyses on each of the three validation days. Within-run and between-run CV were calculated with one-way
analysis of variance.
CV: Coefficient of variation; NA: Not available.

2130 Bioanalysis (2014) 6(16) future science group

Quantification of amikacin & kanamycin using LC–MS/MS Research Article

400 mg
350 mg
Kanamycin concentration (mg/l)




400 mg
Amikacin concentration (mg/l)



0 2 4 6 8 10 12 14 16
Time (h) after start infusion

Figure 2. Concentration–time curve of multidrug-resistant TB patients receiving amikacin and kanamycin

measured with LC–MS/MS.

A Wilcoxon signed rank test was used to compare the use of TDM with amikacin or kanamycin. Moreover,
two methods. With this statistical test, no significant the new method is more sensitive compared with the
difference was found in the measured concentrations. immunoassay, providing information about trough
The newly validated LC–MS/MS method enables the levels.

future science group www.future-science.com 2131

Research Article  Dijkstra, Sturkenboom, van Hateren, Koster, Greijdanus & Alffenaar

(a) (b) (c) NH2
H 2N H2 N

Figure 3. Chemical structures. (A) Apramycin, (B) amikacin and (C) kanamycin.

Conclusion effective analysis of all anti-TB drugs. This will

This new approach enables the analysis of both ami- enable TDM to be available in developing countries,
kacin and kanamycin using a single method in a where TB is a major health problem. Ultimately, this
robust and reliable way. Moreover, the LLOQ of this will lead to improved cure rates and reduce resistance.
method is sufficient to measure trough levels of both
aminoglycosides. Financial & competing interests disclosure
The authors have no relevant affiliations or financial in-
Future perspective volvement with any organization or entity with a financial
Resistance to anti-TB drugs is emerging and adequate interest in or financial conflict with the subject matter or
therapy is essential in eradicating all mycobacteria. materials discussed in the manuscript. This includes employ-
By analyzing the pharmacokinetic parameters of ment, consultancies, honoraria, stock ownership or options,
anti-TB drugs, the effect of the provided therapy can expert testimony, grants or patents received or pending, or
be quantified. This will allow a physician to provide royalties.
individual therapy by optimizing the dosages of all No writing assistance was utilized in the production of this
anti-TB drugs taken. Together with other improve- manuscript.
ments in the field of TB, this could optimize TB care
by minimizing side effects and improving efficacy. Ethical conduct of research
Long-term stability should be evaluated for full The authors state that they have obtained appropriate institu-
validation of the developed method. Yet, with the tional review board approval or have followed the principles
extending use of LC–MS/MS, all anti-TB drugs outlined in the Declaration of Helsinki for all human or animal
could be analyzed within a single sample using a experimental investigations. In addition, for investigations in-
single LC–MS/MS method. LC–MS/MS should be volving human subjects, informed consent has been obtained
installed in a central laboratory to provide a cost- from the participants involved.

Executive summary
Background: quantification of aminoglycosides for treatment of multidrug-resistant TB
• Amikacin and kanamycin are essential for the treatment of multidrug-resistant TB.
• The current available immunoassay is unable to analyze kanamycin and has a high LLOQ for amikacin.
• A new LC–MS/MS method was developed based on the US FDA guidelines.
• The new LC–MS/MS method is able to analyze both amikacin and kanamycin.
• The results of the LC–MS/MS method are comparable to those of the immunoassay.
• The sample preparation of this new method is simple and time efficient.
• This new method is able to analyze aminoglycosides with a sufficient LLOQ.
• The analytical range is extended compared with the immunoassay.

2132 Bioanalysis (2014) 6(16) future science group

Quantification of amikacin & kanamycin using LC–MS/MS Research Article

References 13 Baietto L, D’Avolio A, De Rosa FG et al. Development

Papers of special note have been highlighted as: and validation of a simultaneous extraction procedure
• of interest; •• of considerable interest for HPLC–MS quantification of daptomycin, amikacin,
gentamicin, and rifampicin in human plasma. Anal. Bioanal.
1 Keshavjee S, Farmer PE. Tuberculosis, drug resistance, and
Chem. 396(2), 791–798 (2010).
the history of modern medicine. N. Engl. J. Med. 367(10),
931–936 (2012). 14 Lowes S, Jersey J, Shoup R et al. Recommendations
on: internal standard criteria, stability, incurred sample
2 World Health Organization. Group of drugs to treat
reanalysis and recent 483s by the Global CRO Council for
MDR-TB. In: Treatment of Tuberculosis Guidelines. WHO
Bioanalysis. Bioanalysis 3(12), 1323–1332 (2011).
Press, Switzerland, 84 (2010).
15 Han M, Jun SH, Lee JH, Park KU, Song J, Song SH.
3 Jureen P, Angeby K, Sturegard E et al. Wild-type MIC
Method for simultaneous analysis of nine second-line anti-
distributions for aminoglycoside and cyclic polypeptide
tuberculosis drugs using UPLC–MS/MS. J. Antimicrob.
antibiotics used for treatment of Mycobacterium tuberculosis
Chemother. 68(9), 2066–2073 (2013).
infections. J. Clin. Microbiol. 48(5), 1853–1858 (2010).
16 US Department of Health and Human Services, Food and
4 Craig WA. Does the dose matter?. Clin. Infect. Dis.
Drug Administration. Guidance for Industry, Bioanalytical
33(Suppl. 3), S233–S237 (2001). 
Method Validation. Center for Drug Evaluation and
• Considers the connection between PK parameters and Research, Rockville, MD, USA (2001).
17 European Medicines Agency, Committee for Medicinal
5 Nuermberger E, Grosset J. Pharmacokinetic and Products for Human Use. Guideline on Bioanalytical
pharmacodynamic issues in the treatment of mycobacterial Validation (EMEA/CHMP/EWP/192217/2009). European
infections. Eur. J. Clin. Microbiol. Infect. Dis. 23(4), Medicines Agency, London, UK (2011).
243–255 (2004).
18 Mallet CR, Lu Z, Mazzeo JR. A study of ion suppression
6 Peloquin CA, Berning SE, Nitta AT et al. Aminoglycoside effects in electrospray ionization from mobile phase additives
toxicity: daily versus thrice-weekly dosing for treatment of and solid-phase extracts. Rapid Commun. Mass Spectrom.
mycobacterial diseases. Clin. Infect. Dis. 38(11), 1538–1544 18(1), 49–58 (2004).
19 Harmelink IM, Alffenaar JWC, Wessels AMA, Greijdanus
•• Reports a relation between the cumulative dose of B, Uges DRA. A rapid and simple liquid chromatography-
aminoglycosides and the occurrence of toxicity. tandem mass spectrometry method for the determination
7 Perletti G, Vral A, Patrosso MC et al. Prevention and of linezolid in human serum. Eur. J. Hosp. Pharm. 14, 5–8
modulation of aminoglycoside ototoxicity (Review). Mol. (2008).
Med. Rep. 1(1), 3–13 (2008). 20 de Velde F, Alffenaar JW, Wessels AM, Greijdanus B,
8 de Jager P, van Altena R. Hearing loss and nephrotoxicity Uges DR. Simultaneous determination of clarithromycin,
in long-term aminoglycoside treatment in patients with rifampicin and their main metabolites in human plasma
tuberculosis. Int. J. Tuberc. Lung Dis. 6(7), 622–627 (2002). by liquid chromatography-tandem mass spectrometry.
J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877(18–
• Demonstrates a relationship between nephrotoxicity 19), 1771–1777 (2009).
prevalence and treatment duration and cumulative
21 Pranger AD, Alffenaar JW, Wessels AM, Greijdanus B,
dose of aminoglycosides in the treatment of
Uges DR. Determination of moxifloxacin in human plasma,
multidrug-resistant TB.
plasma ultrafiltrate, and cerebrospinal fluid by a rapid and
9 Gatell JM, Ferran F, Araujo V et al. Univariate and simple liquid chromatography-tandem mass spectrometry
multivariate analyses of risk factors predisposing to auditory method. J. Anal. Toxicol. 34(3), 135–141 (2010).
toxicity in patients receiving aminoglycosides. Antimicrob.
22 Alffenaar JW. Dried blood spot analysis combined with
Agents Chemother. 31(9), 1383–1387 (1987). 
limited sampling models can advance therapeutic drug
• Indicates that patients with ototoxicity have a significantly monitoring of tuberculosis drugs. J. Infect. Dis. 205(11),
higher percentage of trough levels. 1765–1766; author reply 1766 (2012).
10 Peloquin CA. Therapeutic drug monitoring in the treatment 23 Vu DH, Alffenaar JW, Edelbroek PM, Brouwers JR,
of tuberculosis. Drugs 62(15), 2169–2183 (2002).  Uges DR. Dried blood spots: a new tool for tuberculosis
• Assesses the added value of therapeutic drug monitoring in treatment optimization. Curr. Pharm. Des. 17(27),
the treatment of TB. 2931–2939 (2011).

11 Stop TB Partnership. WHO. The Global Plan to Stop TB 24 Vu DH, Bolhuis MS, Koster RA et al. Dried blood spot
2011–2013 (2015). analysis for therapeutic drug monitoring of linezolid in
patients with multidrug-resistant tuberculosis. Antimicrob.
12 Oertel R, Neumeister V, Kirch W. Hydrophilic interaction
Agents Chemother. 56(11), 5758–5763 (2012).
chromatography combined with tandem-mass spectrometry
to determine six aminoglycosides in serum. J. Chromatogr. A 25 Vu DH, Koster RA, Alffenaar JW, Brouwers JR, Uges DR.
1058(1–2), 197–201 (2004). Determination of moxifloxacin in dried blood spots using
LC–MS/MS and the impact of the hematocrit and blood
• Describes the analysis of various aminoglycosides in serum volume. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.
using MS/MS with sufficient LLOQ. 879(15–16), 1063–1070 (2011).

future science group www.future-science.com 2133