Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ahsman, M. J. Articles by Mathot, R. A. Search for Related Content PubMed PubMed Citation Articles by Ahsman, M. J. Articles by Mathot, R. A.

 Previous Article  |  Next Article 

Antimicrobial Agents and Chemotherapy, January 2009, p. 75-80, Vol. 53, No. 1
0066-4804/09/$08.00+0     doi:10.1128/AAC.00636-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Microanalysis of β-Lactam Antibiotics and Vancomycin in Plasma for Pharmacokinetic Studies in Neonates

Maurice J. Ahsman,^1* Enno D. Wildschut,² Dick Tibboel,² and Ron A. Mathot¹

Department of Hospital Pharmacy (Clinical Pharmacology Unit), Erasmus University Medical Center, 's-Gravendijkwal 230, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands,¹ Department of Pediatric Surgery, Sophia Children's Hospital, Erasmus University Medical Center, ’s-Gravendijkwal 230, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands²

Received 15 May 2008/ Returned for modification 18 September 2008/ Accepted 18 October 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Rational dosing of antibiotics in neonates should be based on pharmacokinetic (PK) parameters assessed in specific populations. PK studies of neonates are hampered by the limited total plasma volume, which restricts the sample volume and sampling frequency. Available drug assay methods require large sample volumes and are labor-intensive or time-consuming. The objective of this study was to develop a rapid ultra-performance liquid chromatographic method with tandem mass spectrometry detection for simultaneous quantification of amoxicillin, meropenem, cefazolin, cefotaxime, deacetylcefotaxime, ceftriaxone, and vancomycin in 50 µl of plasma. Cleanup consisted of protein precipitation with cold acetonitrile (1:4) and solvent evaporation before reversed-phase chromatographic separation and detection using electrospray ionization tandem mass spectrometry. Standard curves were prepared over a large dynamic range with adequate limits of quantitation. Intra- and interrun accuracy and precision were within 100% ± 15% and 15%, respectively, with acceptable matrix effects. Coefficients of variation for matrix effects and recovery were <10% over six batches of plasma. Stability in plasma and aqueous stocks was generally sufficient, but stability of meropenem and ceftriaxone in extracts could limit autosampler capacity. The instrument run time was approximately 3.50 min per sample. Method applicability was demonstrated with plasma samples from an extracorporeal membrane oxygenation-treated neonate. Different β-lactam antibiotics can be added to this method with additional ion transitions. Using ultra-performance liquid chromatography mass spectrometry, this method allows simple and reliable quantification of multiple antibiotics in 50 µl of plasma for PK studies of neonates.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Successful drug therapy depends on the administration of the appropriate dose at the right dose interval, which implies an understanding of the drug’s pharmacokinetic (PK) profile. This is particularly relevant in the treatment of neonatal and pediatric patients, considering the potential variation in PK parameters due to an individual’s developmental stage or specific morbidity. Therefore, PK parameters should be assessed for the specific populations to which drugs are given to prevent either undertreatment or unnecessary toxicity. Unfortunately, PK studies of neonates are complicated by the limited amount of biological material (e.g., blood or plasma) available, which poses restrictions on sampling frequency and sample volume (20). Moreover, the number of patients participating in these studies is often low, especially for studies of patients of narrowly defined age groups, with specific (co)morbidity, or during treatment with extracorporeal techniques. Ideally, analytes of interest should be quantitated simultaneously in as little sample as possible. This limits the burden on individual patients caused by sampling while maintaining a sufficient number of data points for reliable data analysis.

For years, drugs have been quantified via high-performance liquid chromatography with UV detection (HPLC-UV). Many of the published bioanalytical methods require sample volumes of 250 µl or more (5, 8, 11, 13, 16). Some are labor-intensive (5, 7, 11) or require long run times (7, 11, 12), potentially leading to poor reproducibility for analytes that may decompose during analysis, such as certain β-lactam antibiotics. Analytes had to be separated chromatographically from the interfering endogenous and exogenous matrix components, and often an elaborate sample preparation was necessary to reach sufficient selectivity and specificity. Coeluting components would often interfere due to the low specificity of UV detection. With the advent of mass spectrometry (MS), selectivity greatly increased because specific analyte masses could be detected. This led to an even greater specificity when mass spectrometers were set up in sequence (tandem mass spectrometry [MS/MS]); now, not only a specific mass could be identified, but a specific fragmentation pattern could be monitored to differentiate between analytes with the same mass.

Equipment capable of ultra-performance liquid chromatography MS (UPLC-MS/MS) has recently become available. With a smaller particle size and higher operating pressures than those for regular HPLC, UPLC provides a shorter run time and sharper peak shape, which improves sensitivity and reduces potential interference by matrix components (6, 10, 17). UPLC combined with MS/MS should therefore allow quantitative analysis of multiple analytes with minimal sample preparation and matrix effects.

Currently, clinical studies at the Sophia Children's Hospital include PK evaluations of multiple antibiotics in patients receiving extracorporeal membrane oxygenation (ECMO) treatment. In order to facilitate these studies, a simple and reliable method was developed to simultaneously quantify amoxicillin (AMX), meropenem (MEM), cefazolin (CFZ), cefotaxime (CTX), deacetylcefotaxime (DACT), ceftriaxone (CRO), and vancomycin (VAN) in 50 µl of plasma. In this report, the method is presented, validation results are reported, and method applicability is demonstrated with data from an ECMO-treated patient.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Reagents. LC-MS-grade water and LC-grade methanol and acetonitrile were obtained from Biosolve (Valkenswaard, The Netherlands). Formic acid (FA; Sigma, Schnelldorf, Germany) was analytical grade. The following reference standards were purchased from Sigma (Schnelldorf, Germany): CRO, VAN, CFZ, CTX, and oxacillin (OXA). DACT was kindly provided by Sanofi-Aventis (Gouda, The Netherlands). MEM was from Molekula (Wimborne, United Kingdom), and AMX was obtained from Certa (Braine-l'Alleud, Belgium).

QC samples and standard solutions. Standard stock solutions containing β-lactam antibiotics were prepared by dissolving the required amount of antibiotic (calculated as free base) in 25 ml of water. VAN solutions were prepared separately to prevent potential accelerated degradation of other antibiotics (3, 9). Various quantities of stock solution were diluted with water, resulting in eight working standards over the concentration range from the lower limit of quantitation (LLOQ) to the upper limit of quantitation (ULOQ). Calibration standards were prepared by diluting 1 part working standard with 9 parts human plasma. Quality control (QC) samples for intra- and interassay comparisons were similarly prepared, using a separate stock solution, and stored at –80°C; low, medium, and high controls were prepared at concentrations of three to four times the LLOQ, 40% of the ULOQ, and 75% of the ULOQ, respectively. A stock solution of the internal standard was prepared by dissolving 10 mg of OXA in 50 ml of water. Prior to analysis, 1 part stock solution was added to 99 parts chilled acetonitrile. This precipitant solution was freshly prepared before each analysis.

Sample preparation. To 50 µl of plasma, 200 µl of chilled acetonitrile containing the internal standard was added. The sample was mixed (5°C, 1,250 rpm) for at least 15 min to complete protein precipitation. After centrifugation at 16,000 x g for 10 min, 200 µl of the supernatant was transferred to a clean vial. The solvent was evaporated to dryness at 40°C under nitrogen gas flow, after which the residue was reconstituted in 100 µl of 0.1% (vol/vol) aqueous formic acid and left to mix for 30 min (5°C, 1,250 rpm). When cloudy, samples were centrifuged again at 16,000 x g for 10 min. The supernatant was transferred to a polypropylene autosampler vial and stored at 5°C until analysis by UPLC-MS/MS.

UPLC-MS/MS conditions. The UPLC-MS/MS system consisted of a Waters Acquity UPLC instrument coupled to a Quattro Premier XE tandem-quadrupole mass spectrometer (Waters Corp., Milford, MA). The analytical column was an Acquity UPLC BEH C₁₈ 2.1-mm by 100-mm column with a 1.7-µm particle size (Waters Ltd., Dublin, Ireland), to which a 0.2-µm precolumn filter unit was added. The mobile phase was a gradient of solution A (0.1% FA in water) and solution B (0.1% FA in methanol), with an initial composition of 20% solution B. The mobile phase composition changed linearly from 20% B at 0.5 min to 40% B at 1.0 min and onward to 100% B at 2.0 min. The composition was switched back to 20% B at 2.5 min and maintained until 3.0 min. The flow rate was 0.4 ml/min, with a column temperature of 40°C. From each sample, 10 µl was injected onto the column. Analytes were detected via MS with an electrospray ionization interface in positive multiple reaction monitoring mode. Optimized multiple reaction monitoring settings for the individual drugs, including cone voltage and collision energy, are listed in Table 1. The acquisition settings were as follows: capillary voltage, 3.4 kV; source temperature, 120°C; desolvation temperature, 300°C; desolvation gas flow, 500 liters/h; cone gas flow, 50 liters/h; and dwell time, 80 ms.

View this table: [in this window] [in a new window]

TABLE 1. Acquisition parametersa

Data analysis. Data were acquired using Masslynx V4.1 software and processed using Quanlynx V4.1 software (Waters Inc.). For all analytes except CRO, calibration curves were obtained by plotting the peak area ratios of drug versus internal standard against the theoretical concentration. Peak height was used for CRO based on superior reproducibility. Standard curves were constructed for each analyte over the calibration range via weighted least-squares regression.

Validation. The method was validated based on FDA guidelines for bioanalytical method validation (2). VAN was added at a later stage with a reduced validation procedure. The full validation procedure included the following parameters.

(i) Specificity and selectivity. Chromatograms for three aqueous calibration standards were compared to those for six batches of blank plasma and 10 patient samples before and after spiking. Ion traces of each analyte’s mass transition were checked for interferences at the respective retention times.

(ii) Limit of quantification. The LLOQ was defined as the lowest concentration that could be quantified with an accuracy and precision of ±20%, as calculated from chromatograms for six independent samples.

(iii) Standard curves. Curves consisting of eight points were calculated by linear or polynomial regression, with each point consisting of independent triplicate measurements. Best fit was selected after exploration of different regression models and weighting factors.

(iv) Accuracy and precision. Intra- and interrun accuracy and precision were calculated for the three QC samples, with six duplicate measurements each or with measurements on six experiments done on different days. Accuracy was defined as the percent deviation from the theoretical concentration by quantifying QC samples with a freshly prepared calibration curve. Precision was defined as the coefficient of variation (CV) (standard deviation/mean of six measurements x 100%).

(v) Robustness. Variations in analytical conditions were mimicked based on observations of unexpected performance changes during method development, as follows: (a) signal intensities and 24-h autosampler stability of extracts (from low and high QC samples in duplicate) reconstituted in aqueous 0.1% FA were compared to those observed for samples reconstituted in water; (b) retention times and signal intensities in medium QC samples (in duplicate) at a column temperature of 40°C were compared to those observed at a column temperature of 30°C; and (c) signal intensities of medium QC samples (in duplicate) were compared to those observed in medium QC samples that were diluted preprecipitation (1:9 with water) to see whether dilution improved analyte recovery.

(vi) Matrix effects. Plasma and eluent components in the ionization chamber cause batch-specific ion suppression or enhancement, leading to interpatient and intrapatient signal variability (4, 19). These matrix effects (ME) were evaluated in two ways. First, extracts of six batches of blank plasma were injected while analytes were continuously infused into the mass spectrometer. Ion traces were recorded for each compound over the entire run time. Signal stability at the relevant retention time was visually assessed for each analyte over the six batches of blank plasma. Second, ME were quantified as proposed by Matuszewski et al. (14). In short, chromatograms were recorded for plasma spiked preextraction, plasma spiked postextraction, and spiked aqueous eluent. In total, six batches of blank plasma were spiked with low and high concentrations of each analyte in duplicate. Recovery (RE) was defined as the relative signal of samples spiked postextraction versus preextraction. ME was similarly defined as the relative signal of postextraction-spiked plasma samples versus spiked aqueous samples. Process efficiency (PE) was defined as the product of RE and ME, i.e., the overall signal of spiked plasma versus an aqueous standard solution. Average values and CVs for RE, ME, and PE were calculated for the six plasma batches.

(vii) Sample stability. Storage conditions and periods were chosen to mimic those at blood collection, during long-term storage of stock solutions and plasma, during freeze-thaw cycles, at the tabletop during processing, and in the autosampler awaiting analysis. QC samples were tested for stability over time (a) in aqueous stock solutions and plasma at –80°C (1 week and 1 to 2 months), (b) in aqueous stock solutions and plasma at 5 and 20°C and in EDTA-decoagulated whole blood at 5°C (6, 18, 24, 48, and 144 h), (c) in extracts at –80, 5, and 20°C (6, 18, 24, and 48 h), and (d) in extracts after three freeze-thaw cycles. Maximum storage periods were estimated based on an allowed concentration drop of 10%.

Measurement of plasma levels in a neonatal ECMO-treated patient. Patients receiving ECMO treatment were included after written parental consent. The study protocol was approved by the institutional ethics committee. We present data from a term neonate with persistent pulmonary hypertension after meconium aspiration. Antimicrobial treatment was given in accordance with the departmental protocol and included CTX at 50 mg/kg of body weight twice a day and AMX at 25 mg/kg four times a day for suspected sepsis and a single bolus injection of VAN at 20 mg/kg in preparation for decannulation. In total, 11 samples were taken extracorporeally from a preoxygenator access point during the 84-h ECMO run. After this period, the patient was successfully decannulated and transferred to the referring neonatal intensive care unit on conventional ventilation. Plasma levels of CTX, DACT, AMX, and VAN were simultaneously measured in 50 µl of plasma. Individual PK curves were constructed for CTX, AMX, and VAN by fitting measured plasma levels to previously reported PK parameters, using MW\Pharm software (MW\Pharm 3.58; Mediware, The Netherlands). CTX was modeled on a one-compartment model derived from data for non-ECMO-treated neonates (15), using iterative Bayesian fitting. AMX was modeled on a one-compartment model derived from non-ECMO-treated neonates (18), using iterative Bayesian fitting. VAN was modeled on a two-compartment model derived from ECMO-treated neonates (1), using non-Bayesian fitting.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Selectivity was achieved by the independent separation mechanisms of chromatography and MS/MS. None of the aqueous standards, plasma standards, or spiked patient samples contained interfering components. Representative ion traces can be seen in Fig. 1. A minor interference in the CTX ion trace was probably caused by (m/z + 1) isotopes of CFZ, but CTX and the interference were chromatographically separated, with retention times of 1.67 and 1.74 min, respectively. An alternate mass transition for CFZ and CTX was tried but rejected based on a loss of signal intensity. Standard curves were prepared with a weighting factor of 1/x. Most of the standard curves were best described via nonlinear regression. CFZ's standard curve could be divided into a linear and a nonlinear segment, resulting in a better fit. A linear calibration curve is generally more desirable and can be applied to samples with low to medium concentrations of CFZ. We tested whether high QC samples could be diluted 10-fold with blank plasma before extraction. This led to a 10-fold drop in concentration, while maintaining accuracy and precision (results not shown), but led to a proportional increase in the LLOQ as well. Coefficients of determination (R²) varied from 0.994 (VAN) up to 0.998 (CTX and DACT). See Table 1 for the dynamic range of each analyte. The low end of the dynamic range was considered to be the LLOQ; accuracy and precision for all analytes were within 100% ± 20%, and the CV was <20%. See Table 2 for intrarun accuracy and precision. Interrun accuracy and precision were similar, with an accuracy and precision of 100% ± 15% and a CV of <15% for all analytes.

View larger version (7K): [in this window] [in a new window]

FIG. 1. Representative chromatograms for a mixture of analytes in plasma, with an individual ion trace for each analyte. Most analytes have good peak shapes, with the exception of CTX (which has a peak at the retention time of CFZ) and CRO (which shows some tailing).

View this table: [in this window] [in a new window]

TABLE 2. Intrarun accuracy and precision (n = 6 for each concentration)a

Robustness. Reconstitution with water instead of 0.1% FA improved the signal intensities of CFZ (+8%), DACT (+10%), and CRO (+300%); water had no effect on the signal intensities of the other analytes. Twenty-four-hour degradation (autosampler, 5°C), however, was considerably worse in water for MEM (–67% versus –40%), AMX (–50% versus –7%), CTX (–45% versus –19%), and DACT (–38% versus –20%). A decrease in column temperature to 30°C led to longer retention times without a deterioration of signal intensity, peak shape, and resolution. Tenfold sample dilution led to a correspondingly decreased signal intensity for all analytes except VAN; dilution of samples containing VAN likely improved sample cleanup, resulting in relatively high signal intensities.

ME. Visual inspection of chromatograms of plasma injected during T infusion revealed a signal loss of roughly 30 to 50% due to matrix components. Interbatch variability for plasma appeared small, and there were no sudden signal losses or peaks around the respective retention times. ME, RE, and PE were similar for low and high QC samples (see Fig. 2 for the high QC samples). PE varied between 20% (VAN) and 75% (CRO), with notable ME and RE for each analyte. CRO was the only analyte with ion enhancement due to matrix components, as opposed to the ion suppression seen with the other analytes. ME and RE CVs over the six different plasma batches were under 10% for each analyte.

View larger version (10K): [in this window] [in a new window]

FIG. 2. ME (A), RE (B), and overall PE (C) of analytes in high QC samples. Values are averages with corresponding 95% confidence intervals. For VAN, only PE was tested.

Sample stability. Analytes were stable for at least 2 months in water and plasma at –80°C. Maximum storage periods for aqueous solutions, plasma, whole blood, and extracts (Table 3) were based on a maximum amount of degradation of 10%. After three freeze-thaw cycles (n = 2 for both low and high QC concentrations), average remaining concentrations in extracts were at least 90% of the initial concentration, except for MEM (79%).

View this table: [in this window] [in a new window]

TABLE 3. Maximum in-process and autosampler storage periods

Measurement of plasma levels in a neonatal ECMO-treated patient. Plasma levels of CTX, DACT, AMX, and VAN were measured simultaneously in the 50-µl plasma samples and successfully fitted to the previously reported models. Figure 3 contains individual concentration-time curves and measured concentrations. Individual parameters were as follows: elimination rate constant (k_el) = 0.197 h^–1 and volume of distribution (V_d) = 0.912 liter/kg for CTX; k_el = 0.330 h^–1 and V_d = 0.713 liter/kg for AMX; clearance = 7.63 liters/h/1.85 m², V₁ = 1.03 liters/kg, and the intercompartmental clearance rates from central to peripheral (k₁₂) and vice versa (k₂₁) for VAN were 1.5 h^–1 and 2.634 h^–1, respectively.

View larger version (11K): [in this window] [in a new window]

FIG. 3. Measured concentrations (open circles) and individually fitted curves (lines) for CTX (A), AMX (B), and VAN (C) in an ECMO-treated neonate. Deacetylcefotaxime concentrations (closed circles) were not fitted.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We report the development of a fast UPLC-MS/MS method with an analytical performance meeting FDA specifications. The required plasma volume is 50 µl, which is sufficient to allow reinjection. If the latter is not deemed necessary, the sample volume could probably even be reduced to 20 µl.

This method can be expanded for the quantification of other β-lactam antibiotics by scanning additional mass transitions. We have, for instance, tested ceftazidime and cefuroxime, with sufficient retention, signal intensity, and peak shape. Among the presented group of antibiotics, only CRO had a potentially problematic peak shape, but the accuracy and precision were adequate. We tested an adjusted gradient but were unsuccessful in removing CRO tailing while maintaining resolution between CTX and the CFZ isotope peak.

The limited availability of a reference standard complicated a full validation with respect to DACT quantitation. Considering the structural, chromatographic, and mass spectrometric similarities between DACT and CTX, we assumed that their analytical performances would be similar and limited the validation procedure to the critical aspects of ME and sample stability.

Initially, this method was designed for β-lactam antibiotics only. VAN was later added with a reduced validation procedure to simultaneously quantify commonly used combinations of antibiotics. Since VAN stability has been demonstrated (13), we did not include stability testing.

Many reported LC-MS methods contain an elaborate cleanup procedure, such as liquid-liquid extraction or solid-phase extraction, to provide a sufficient response without interfering ME. Our method shows that a simple protein precipitation with acetonitrile in combination with the narrow peak shape provided by UPLC-MS/MS can be used to quantify antibiotics with acceptable accuracy and precision, despite the presence of ME. At least 80% of most analytes was recovered after protein precipitation. This could imply that either a small fraction of analyte was not displaced from protein binding sites or the analytes did not readily dissolve into the highly organic solvent. Despite incomplete RE and ME, the method accuracy and precision comply with predefined specifications.

The limited stability of certain β-lactam antibiotics potentially limits analysis of large sample runs for the least stable antibiotics, MEM and CRO. The total analysis time was less than 4 min per sample, however, which allows large sample runs for MEM and CRO, provided that samples are processed and analyzed without delay.

We measured CTX, DACT, AMX, and VAN simultaneously in 50-µl samples taken from an ECMO-treated neonate. Compared to the data for non-ECMO-treated neonates (15), CTX showed similar clearance, with a twofold increase in the V_d, which can be explained by the added volume of the ECMO circuit and edema. AMX clearance did not differ from the clearance found in non-ECMO-treated neonates (18); V_d was also slightly increased, but this may have been underestimated because of the few AMX concentrations available from this particular patient directly following injection. VAN clearance and V_d were similar to those reported previously for ECMO-treated neonates (1), although clearance was higher in our patient.

With PK software, we were able to construct concentration-time curves and to calculate individual PK parameters for this neonate, using existing models. The high sampling frequency during classic PK studies can be problematic with neonates. We expect to be able to compute population PK parameters for this specific population by using sophisticated nonlinear mixed effects modeling (NONMEM) software, combining sparse and randomly sampled concentration data from multiple patients. This complements the microanalysis method, making maximum use of as little and as few samples as possible.

This UPLC-MS/MS method for quantification of AMX, MEM, CFZ, CTX, DACT, CRO, and VAN in 50 µl of plasma provides reliable concentration data. In combination with sophisticated PK modeling software, this enables efficient PK studies of neonates.

 
* Corresponding author. Mailing address: Department of Hospital Pharmacy (Clinical Pharmacology Unit), Erasmus University Medical Center, 's-Gravendijkwal 230, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. Phone: 31 10 7033202. Fax: 31 10 7032400. E-mail: m.ahsman{at}erasmusmc.nl

Published ahead of print on 27 October 2008.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Amaker, R. D., J. T. DiPiro, and J. Bhatia. 1996. Pharmacokinetics of vancomycin in critically ill infants undergoing extracorporeal membrane oxygenation. Antimicrob. Agents Chemother. 40:1139-1142.[Abstract]
2
2. Anonymous. 2001. Guidance for industry—bioanalytical method validation. Biopharmaceutics Coordinating Committee, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Rockville, MD.
3
3. Basma, V. 2004. Stability and compatibility of vancomycin for administration by continuous infusion in intensive care patients. European Congress of Clinical Microbiology and Infectious Diseases, Prague, Czech Republic.
4
4. Careri, M., and A. Mangia. 2006. Validation and qualification: the fitness for purpose of mass spectrometry-based analytical methods and analytical systems. Anal. Bioanal. Chem. 386:38-45.[Medline]
5
5. Chou, Y. W., Y. H. Yang, J. H. Chen, C. C. Kuo, and S. H. Chen. 2007. Quantification of meropenem in plasma and cerebrospinal fluid by micellar electrokinetic capillary chromatography and application in bacterial meningitis patients. J. Chromatogr. B 856:294-301.
6
6. Churchwell, M. I., N. C. Twaddle, L. R. Meeker, and D. R. Doerge. 2005. Improving LC-MS sensitivity through increases in chromatographic performance: comparisons of UPLC-ES/MS/MS to HPLC-ES/MS/MS. J. Chromatogr. B 825:134-143.
7
7. Denooz, R., and C. Charlier. 2008. Simultaneous determination of five beta-lactam antibiotics (cefepim, ceftazidim, cefuroxim, meropenem and piperacillin) in human plasma by high-performance liquid chromatography with ultraviolet detection. J. Chromatogr. B 864:161-167.
8
8. Farin, D., R. Kitzes-Cohen, G. Piva, and I. Gozlan. 1999. High performance liquid chromatography method for the determination of meropenem in human plasma. Chromatographia 49:253-255.
9
9. Glew, R. H., and R. A. Pavuk. 1983. Stability of gentamicin, tobramycin, and amikacin in combination with four beta-lactam antibiotics. Antimicrob. Agents Chemother. 24:474-477.[Abstract/Free Full Text]
10
10. Guillarme, D., D. T. Nguyen, S. Rudaz, and J. L. Veuthey. 2007. Recent developments in liquid chromatography—impact on qualitative and quantitative performance. J. Chromatogr. A 1149:20-29.
11
11. Kraemer, H. J., R. Gehrke, A. Breithaupt, and H. Breithaupt. 1997. Simultaneous quantification of cefotaxime, desacetylcefotaxime, ofloxacine and ciprofloxacine in ocular aqueous humor and in plasma by high-performance liquid chromatography. J. Chromatogr. B 700:147-153.
12
12. Lindegardh, N., T. Singtoroj, A. Annerberg, N. J. White, and N. P. Day. 2005. Development and validation of a solid-phase extraction-liquid chromatographic method for determination of amoxicillin in plasma. Ther. Drug Monit. 27:503-508.[CrossRef][Medline]
13
13. Luksa, J., and A. Marusic. 1995. Rapid high-performance liquid chromatographic determination of vancomycin in human plasma. J. Chromatogr. B 667:277-281.[Medline]
14
14. Matuszewski, B. K., M. L. Constanzer, and C. M. Chavez-Eng. 2003. Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS. Anal. Chem. 75:3019-3030.[Medline]
15
15. McCracken, G. H., Jr., N. E. Threlkeld, and M. L. Thomas. 1982. Pharmacokinetics of cefotaxime in newborn infants. Antimicrob. Agents Chemother. 21:683-684.[Abstract/Free Full Text]
16
16. Piva, G., D. Farin, I. Gozlan, and R. Kitzes-Cohen. 2000. HPLC method for determination of cefuroxime in plasma. Chromatographia 51:154-156.
17
17. Plumb, R., J. Castro-Perez, J. Granger, I. Beattie, K. Joncour, and A. Wright. 2004. Ultra-performance liquid chromatography coupled to quadrupole-orthogonal time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 18:2331-2337.[CrossRef][Medline]
18
18. Pullen, J., L. M. Stolk, F. H. Nieman, P. L. Degraeuwe, F. H. van Tiel, and L. J. Zimmermann. 2006. Population pharmacokinetics and dosing of amoxicillin in (pre)term neonates. Ther. Drug Monit. 28:226-231.[Medline]
19
19. Taylor, P. J. 2005. Matrix effects: the Achilles heel of quantitative high-performance liquid chromatography-electrospray-tandem mass spectrometry. Clin. Biochem. 38:328-334.[CrossRef][Medline]
20
20. Ward, R. M., W. E. Benitz, D. K. Benjamin, Jr., L. Blackmon, G. P. Giacoia, M. Hudak, T. Lasky, W. Rodriguez, and A. Selen. 2006. Criteria supporting the study of drugs in the newborn. Clin. Ther. 28:1385-1398.[Medline]

---

Antimicrobial Agents and Chemotherapy, January 2009, p. 75-80, Vol. 53, No. 1
0066-4804/09/$08.00+0     doi:10.1128/AAC.00636-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ahsman, M. J. Articles by Mathot, R. A. Search for Related Content PubMed PubMed Citation Articles by Ahsman, M. J. Articles by Mathot, R. A.