- Email: [email protected]
Assay of dried blood spot from finger prick for sodium valproate via ink auxiliary headspace gas chromatography mass spectrometry
Journal of Chromatography A, 1601 (2019) 335–339
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Assay of dried blood spot from finger prick for sodium valproate via ink auxiliary headspace gas chromatography mass spectrometry Meng-zhe Guo a , Lili Shao a , Xi Chen a , Hai-juan Li a , Liang Wang a,b , Yuan-jiang Pan c,∗ , Dao-quan Tang a,∗ a
Jiangsu Key Laboratory of New Drug Research and Clinical Pharmacy, Xuzhou Medical University, Xuzhou, Jiangsu 221004, China Department of Bioinformatics, Xuzhou Medical University, Xuzhou, Jiangsu 221004, China c Department of Chemistry, Zhejiang University, Hangzhou, Zhejiang 310058, China b
a r t i c l e
i n f o
Article history: Received 30 April 2019 Received in revised form 21 May 2019 Accepted 22 May 2019 Available online 24 May 2019 Keywords: Ink auxiliary Dried blood spot Sodium valproate Epilepsy Headspace gas chromatography mass spectrometry
a b s t r a c t Sodium valproate is the most commonly used antiepileptic drug that patients need to keep taking over a long period of time or on a permanent basis. Its blood concentration should be accurately detected to avoid toxicity or side-effects, especially for children and the aged. Dried blood spot (DBS) sampling from finger prick is a minimally invasive and patient-friendly procedure for blood collection. However, there are few studies about rapid detection of sodium valproate in DBS samples in current literatures. In this work, we developed an ink auxiliary headspace gas chromatography mass spectrometry (GC–MS) strategy for direct detection of sodium valproate in DBS from epilepsy patients, which does not need extra solvent extraction or elution. It was discovered that carbon black ink could provide better capacity of heat absorption and dissociation, and higher quality of headspace sampling. The detection sensitivity has been improved with reported headspace GC–MS methods, and the limit of quantitation could reach to 200 ng/mL. Finally, this strategy was practically applied to quantify sodium valproate in DBS samples from 29 epilepsy patients. The result showed higher accuracy with lower relative errors by comparing with the clinical immunoassay results. In conclusion, we developed a direct detection method for DBS samples that is suitable for high-throughput clinical test with great potential for clinical application. © 2019 Elsevier B.V. All rights reserved.
1. Introduction Sodium valproate is the first anti-epilepsy medication approved by the Food and Drug Administration (FDA) and considered as the most commonly used antiepileptic drug for monotherapy treatment in the worldwide [1]. This drug needs to be taken for a long-term period and its concentration in blood should be closely monitored because the effective concentration is close to the toxic concentration [2,3]. In particular, personalized administration is necessary for sodium valproate to avoid toxicity on the basis of its efficacy [4]. However, blood concentration monitoring of sodium valproate is often difficult to perform because existing instrument requires at least 150 L of serum sample for a single test, which is equivalent to 0.5–2 mL of blood sample. In addition, external conditions such as collection, storage, and transportation of blood
∗ Corresponding authors. E-mail addresses: [email protected] (Y.-j. Pan), [email protected] (D.-q. Tang). https://doi.org/10.1016/j.chroma.2019.05.039 0021-9673/© 2019 Elsevier B.V. All rights reserved.
samples will also affect the accuracy of detection. Therefore, more advanced blood collection method needs to be developed. Dried blood spot (DBS) is an economical and easy-to-perform method, which enables self-sampling plausible for patients at home or outpatient service [5,6]. So far, the method has achieved numerous benefits and become acclaimed in bioanalytical, clinical and pharmaceutical fields [7–9]. For instance, stable whole blood sample can be collected in DBS with no need for anticoagulant or serum/plasma separation during shipment and storage [10]. In addition, DBS requires a small volume of blood from finger prick, which is an easy, minimally invasive, and patient-friendly approach for blood collection [11]. Although DBS sampling method has a series of advantages, there are great challenges for the detections of targets within the samples. Since the whole blood stored in DBS is low in content, the detection technique needs to be more sensitive and overcome the matrix effects of the blood [12]. Moreover, the compounds in DBS will go through a complex elution or extraction process, and the efficiency of desorption directly affects the analytical sensitivity. Thus, a DBS detection method with simple and efficient sample desorption is urgently needed.
336
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
The headspace gas chromatography mass spectrometry (GC–MS) technique provides a variety of advantages for DBS detection [13–16]. First of all, it promotes the separation of volatile contents from complex matrix samples with the introduction of headspace sampling [17,18]. Secondly, the headspace sampling can simply desorb the samples through heat [19]. Until now, some researchers have used this technique in the detection of sodium valproate in DBS [20]. However, the sensitivity of headspace GC–MS will be affected by the desorption degree of sodium valproate. To solve this issue, sample pre-treatment methods such as headspace solid-phase microextraction (HSSPME) or derivatization have been developed to improve the detection sensitivity [21–23]. However, these methods are time consuming with complicated operations, and need extra enrichment materials, which is not suitable for large-scale clinical testing [24]. In this study, we developed a straightforward headspace GC–MS method for fast detection of sodium valproate in DBS. After multiple rounds of trial-and-error screening, we identified black ink being the best assistance to enhance desorption of sodium valproate. In addition, various parameters of detection were also optimized, which includes the size of DBS, the type of black ink, the kind and content of extra infiltrating solvent, and the temperature of headspace sampling. The mechanism of ink-assisted desorption has also been attempted to explore. Finally, we applied this method for the detection of practical DBS samples from epilepsy patients. This method achieved better sensitivity and accuracy by comparing with the existing clinical laboratory methods. 2. Experiments 2.1. Materials and chemicals Acetone and ethanol were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Methanol, acetonitrile, acetic acid, and ammonium acetate were of chromatographic grade and purchased from MERDA (USA). Sodium valproate (95%) was purchased from Energy Chemical (Shanghai, China). Carbon black ink and prepared Chinese ink were purchased from HERO and “Yi De Ge”(China). The pencils were purchased from Mitsu-Bishi (Japan) (Fig. S1). 2.2. The selection of ink and optimization of headspace GC–MS method Firstly, different processing methods of DBS were compared, containing black ink of pen assisted, ink of Chinese brush assisted, printing black ink assisted, and pencils with different blackness assisted. 10 L of ink was added into the DBS. Secondly, the best method was chosen as the final processing method of DBS. Moreover, other parameters of headspace sampling were optimized, such as the size of filter paper, kinds of infiltrating solvent, as well as volume and additive, and the temperature of headspace sampling. 2.3. The investigation of the mechanism of black ink auxiliary improving the sensitivity of headspace GC–MS detection The mechanism of black ink auxiliary improving the sensitivity of headspace GC–MS detection has been investigated by several methods. Firstly, infrared camera was used to obtain the heat absorption of blank filter paper, ink assisted filter paper, and pencil assisted filter paper after laser radiation. Then scanning electron microscope (SEM) was used to achieve the fine surface structure of filter paper with ink and pencil assisted. Finally, three kinds of filter papers mentioned above were added sodium valproate and passed through the solvent elution and normal GC–MS detection.
The result may indicate other enhancement mechanism by ink auxiliary. 2.4. Application of ink auxiliary headspace GC–MS method in the quantitative determination of sodium valproate from finger prick DBS of epilepsy patients The DBS samples including 29 epilepsy patients taking sodium valproate were provided by Xuzhou Central Hospital. The experiments were approved by the ethics committee. Method validation has been taken, containing quantitative linear, precision and accuracy, stability, and recovery. Then DBS samples were detected by our developed headspace GC–MS strategy. SPSS 16.0 was used for the statistical analysis, and the detection results were compared with the actual clinical detection results from immunoassay to confirm the accuracy of our method. The t test was used to identify the statistic differences. 2.5. GC–MS conditions Analysis was performed by GC–MS using an Agilent 7890A GC interfaced with the 5975C single-quadrupole mass spectrometry equipped with electron ionization (EI) (Santa Clara, CA, USA). The GC separation was performed on an HP-5MS column (60 m × 0.25 mm i.d. × 0.25 m film thickness, Agilent). The chromatographic separation mode was achieved using the following temperature program: initial temperature of 70 ◦ C for 1 min; raised to 200 ◦ C by 20 ◦ C min; raised to 260 ◦ C by 50 ◦ C min and kept for 5 min. The inlet temperature was 260 ◦ C. The carrier gas was helium. The flow rate was 1 mL/min and the split ratio was 10 to 1. The ion source was electron impact ion source with 70 eV of collision energy. The detection was SIM mode with 0–5 min of solvent delay. Three chosen reference ions were m/z 57, 73, and 102, respectively. The reference standard of sodium valproate was spiked into the blank whole blood from heather to prepare the 200 g/mL of standard solution. Then the standard solution was dilution using the blank matrix to 150, 120, 100, 80, 60, 50, 40, 15, 1 g/mL. Finally bring 10 L of each solution into the filter paper respectively to prepare the standards for calibration. The quality control (QC) samplers were similarly prepared. 3. Results and discussion 3.1. Optimization of ink auxiliary headspace GC–MS method Pilot study showed that filter paper with black surface could improve heat absorption, leading to enhance desorption of samples during headspace sampling. Therefore, the blackening process would contribute to DBS assay to strengthen the detection sensitivity of sodium valproate in headspace GC–MS. In this work, four types of black substances were investigated, which include fountain pen carbon black ink, prepared Chinese ink, black pencils with different blackness, and ink for printing. According to the result, the carbon black ink showed the best enhancement of sensitivity when compared with the other three materials in the headspace GC–MS detection. Thus, we used this substance as a supplement and developed an integrated strategy for the determination of DBS. In this strategy as shown in Fig. 1, other parameters have also been optimized. Firstly, the size of DBS and the volume of blood were studied. Optimal DBS size was 7 mm square. With this size, the best volume of blood was 10 L, which could be conveniently collected via finger prick. In addition, we found that the organic solvent could improve desorption in headspace sampling after the blackening of DBS. This auxiliary solvent could help to carry the sodium valproate into the gas by evaporation due
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
337
Fig. 1. The ink auxiliary headspace GC–MS strategy for the analysis of sodium valproate in DBS.
Table 1 The optimization of several parameters in ink auxiliary headspace GC–MS analysis. Parameters
Optimization
DBS treatment
White filter paper; Pencil auxiliary; Ink auxiliary Acetone; Methanol; Ethanol; Acetonitrile 0.1% Formic acid; 0.2% Formic acid; 0.3% Formic acid; 0.4% Formic acid; 0.5% Formic acid 4 × 4 mm; 7 × 7 mm; 10 × 10 mm; 12 × 12 mm; 14 × 14 mm 140; 160; 180; 190; 200
Auxiliary solvent Additive in solvent DBS Size Headspace temperature (◦ C)
to the low boiling point. Therefore, organic solvent was also optimized, which involves type, volume, and the additive. As shown in Table 1, acetone was identified as the preferred solvent by comparing with methanol, ethanol, and acetonitrile, and the required acetone volume was only 50 L. Additionally, 0.4% of formic acid was also chosen as an additive to improve desorption of sodium valproate through its ion inhibition effect. Finally, the temperature of headspace sampling was investigated and the best dissociation temperature was obtained at 180 ◦ C. After addition of the carbon black ink and adoption of optimized parameters, the detection sensitivity increased about 200 times than the direct DBS headspace GC–MS assay. 3.2. The mechanism of black ink auxiliary for improving the sensitivity of headspace GC–MS detection The mechanism of black ink auxiliary for improving desorption of sodium valproate in headspace GC–MS was investigated. Three types of filter papers, including blank control, blackened by ink and pencil were put into the headspace bottle and heated, respectively. After 30 s, these filter papers were detected by infrared camera to obtain the surface thermal values. As the result, the surface temperature of blank filter paper was 31 ◦ C while filter papers disposed by ink and pencil reached to 50 and 60 ◦ C, respectively. The increased heat absorption of filter paper can help desorption of sodium valproate. SEM was used to observe the microstructure of treated filter paper. As shown in Fig. S2, comparing with the blank filter paper, there were multilayer carbon powders covering the fiber surface of the filter papers treated by ink with small and evenly-distributed
particles. Therefore, we hypothesized that the carbon powder could provide extra assistance for desorption of the sodium valproate from filter paper. To prove this assumption, 100 L of acetone was used as the solvent for the elution of sodium valproate from DBS and the eluant was directly detected by GC–MS without headspace sampling. The results showed that the solvent from ink auxiliary DBS had the highest content of sodium valproate, demonstrating that nanoscale carbon powder in the surface of filter paper could indeed help the elution and desorption. 3.3. Application of ink auxiliary headspace GC–MS in quantitative determination of sodium valproate in DBS from epilepsy patients Following the optimization of ink auxiliary headspace GC–MS strategy, method validation was then performed. As shown in Tables S1 and S2, the result demonstrated a good linearity in the range of 1–200 g/mL. The correlation coefficient (r) was calculated to be greater than 0.99 and the limit of quantitation was 200 ng/mL. This linear range can fully meet the detection requirement of sodium valproate in DBS, which is better than reported headspace GC–MS method [21–23,25,26]. In addition, both interbatch and intra-batch accuracy and precision showed the favorable accuracy from 96.55% to 105.14%, with relative standard deviation (RSD) from 4.76% to 10.18%. In addition, we also applied the headspace GC–MS strategy into the quantification of sodium valproate in DBS from epilepsy patients. To investigate the accuracy of our method in practical assay, drug concentrations of these patients were also analyzed by existing clinical tests of immunoassay. The 29 patients have the ages from 4 to 87 years and are gender-balanced, which indicates a sufficient coverage of all kinds of people taking the drug. Moreover, drug concentrations were measured from 19 to 131 g/mL, representing the concentration range of the chemical in clinical tests of immunoassay. According to the results (Table 2 and Fig. 2), our method had higher accuracy when compared with immunoassay clinical tests. The median of errors was 3.57%. Only two blood concentrations were obviously different from the immunoassay clinical tests. These two concentrations were the lowest among the 29 samples (19 g/mL and 20 g/mL from the immunoassay clinical tests, respectively). However, it was not necessary to mean that our method had the wrong detection results because the exist-
338
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
Table 2 The determination of 29 DBS samples from epilepsy patients and its comparison with the clinical test. Number
Age
Sex
Clinical concentration (g/ml)
Measured concentration (Mean ± SD)/ (g/ml)
Deviation(%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
46 64 45 30 17 36 63 13 50 17 44 35 43 33 76 57 44 24 10 87 4 45 55 4 29 42 42 56 72
Female Female Female Male Male Male Female Female Male Male Female Male Male Male Male Male Male Male Male Male Male Male Female Female Male Male Male Male Male
131.00 130.00 124.00 120.00 117.00 112.00 104.00 97.00 89.00 85.00 77.00 75.00 73.00 65.00 64.00 64.00 59.00 58.00 56.00 46.00 43.00 41.00 39.00 38.00 36.00 33.00 27.00 20.00 19.00
130.46 ± 14.15 132.97 ± 8.45 124.07 ± 6.30 131.07 ± 7.32 117.50 ± 11.03 112.92 ± 9.22 99.55 ± 2.08 81.73 ± 3.08 80.06 ± 3.87 79.58 ± 4.76 75.83 ± 4.71 71.23 ± 6.67 72.44 ± 4.40 63.18 ± 4.17 62.20 ± 6.41 64.64 ± 1.33 61.11 ± 1.81 58.54 ± 4.91 55.88 ± 1.67 41.17 ± 2.90 48.70 ± 2.41 45.65 ± 0.66 35.37 ± 1.53 40.37 ± 6.45 42.04 ± 2.79 40.59 ± 2.93 27.03 ± 3.02 35.04 ± 0.79 28.35 ± 0.62
−0.41% 2.28% 0.05% 9.22% 0.43% 0.82% −4.28% −15.74% −10.05% −6.38% −1.52% −5.02% −0.77% −2.79% −2.81% 1.00% 3.57% 0.93% −0.21% −10.51% 13.26% 11.35% −9.32% 6.23% 16.77% 22.99% 0.12% 75.18% 49.21%
drug concentration due to the effect of structural analogs such as metabolites. 4. Conclusion In this study, we developed an ink auxiliary headspace GC–MS strategy for the analysis of sodium valproate in DBS from epilepsy patients. The detection sensitivity was improved about 200 times with the reported direct DBS headspace GC–MS assay, and the limit of quantitation could reach to 200 ng/mL. In addition, this strategy has been successfully applied into the determination of sodium valproate from 29 epilepsy patients. By comparing with the clinical test results, our detection showed higher accuracy with lower relative errors. In sum, we developed a novel method for quantitative analysis of sodium valproate in DBS from finger prick with convenient storage and transportation, which is suitable for high-throughput clinical tests and shows great potential for clinical applications in future. Conflict of interest The authors declare that they have no conflict of interest. Acknowledgements We appreciate the support of Natural Science Foundation of China (No. 21505116), Jiangsu Natural Science Foundation of China (Nos. BK20171183 and BK20181147), Jiangsu University Natural Science Foundation of China (No. 16KJA350001), and Natural Science Foundation of Xuzhou City (Nos. KC17187 and KC18050).
Fig. 2. Comparison of the headspace GC–MS method and clinical test (a) and its relative errors (b) based on 29 epilepsy patients, the error bar stands for the standard deviation.
Appendix A. Supplementary data
ing clinical immunoassay method also lacks of enough accuracy in the detection of very low concentrations in DBS samples. Moreover, the result of immunoassay is commonly higher than the real
Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.chroma.2019. 05.039.
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
References [1] L. Gu, B.Y. Liang, Q. Chen, J.X. Long, J.J. Xie, G.L. Liang, Y. Yan, J.J. Tan, W.H. Dou, W. Chen, P. Wu, J.P. Wang, L. Su, Prevalence of epilepsy in the People’s Republic of China: a systematic review, Epilepsy Res. 105 (2013) 195–205. [2] S. Balestrini, S. Sisodiya, Pharmacogenomics in epilepsy, Neurosci. Lett. 667 (2018) 27–39. [3] L. Walker, N. Mirza, V. Yip, A. Marson, M. Pirmohamed, Personalized medicine approaches in epilepsy, J. Intern. Med. 227 (2015) 218–234. [4] S. Cremers, N. Guha, B. Shine, Therapeutic drug monitoring in the era of precision medicine: opportunities, Br. J. Clin. Pharmacol. 82 (2016) 900–902. [5] C. Linder, M. Neideman, K. Wide, M. von Euler, L.L. Gustafsson, A. Pohanka, Dried blood spot self-sampling by guardians of children with epilepsy is feasible: comparison with plasma for multiple antiepileptic drugs, Ther. Drug Monit. (2019), http://dx.doi.org/10.1097/FTD.0000000000000605. [6] C. Linder, A. Hansson, S. Sadek, L.L. Gustafsson, A. Pohanka, Carbamazepine, lamotrigine, levetiracetam and valproic acid in dried blood spots with liquid chromatography tandem mass spectrometry; method development and validation, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 1072 (2018) 116–122. [7] C. Linder, K. Wide, M. Walander, O. Beck, L.L. Gustafsson, A. Pohanka, Comparison between dried blood spot and plasma sampling for therapeutic drug monitoring of antiepileptic drugs in children with epilepsy: a step towards home sampling, Clin. Biochem. 50 (2017) 418–424. [8] C. Lander, M. Andersson, K. Wide, O. Beck, A. Pohanka, A LC–MS/MS method for therapeutic drug monitoring of carbamazepine, lamotrigine and valproic acid in DBS, Bioanalysis 7 (2015) 2031–2039. [9] A. Pohanka, M. Mahindi, M. Masquelier, L.L. Gustafsson, O. Beck, Quantification of valproic acid in dried blood spots, Scand. J. Clin. Lab. Invest. 74 (2014) 648–652. [10] S.T. Kong, S.H. Lim, E. Chan, P.C. Ho, Estimation and comparison of carbamazepine population pharmacokinetics using dried blood spot and plasma concentrations from people with epilepsy: the clinical implication, J. Clin. Pharmacol. 54 (2014) 225–233. [11] I. Ryona, J. Henion, A book-type dried plasma spot card for automated flow-through elution coupled with on-line SPE-LC-MS/MS bioanalysis of opioids and stimulants in blood, Anal. Chem. 88 (2016) 11229–11237. [12] J. Deglon, A. Thomas, Y. Daali, E. Lauer, C. Samer, J. Desmeules, P. Dayer, P. Mangin, C. Staub, Automated system for on-line desorption of dried blood spots applied to LC/MS/MS pharmacokinetic study of flurbiprofen and its metabolite, J. Pharm. Biomed. Anal. 54 (2011) 359–367. [13] A. Maruti, E. Duran-Guerrero, C. Barroso, R. Castro, Optimization of a multiple headspace sorptive extraction method coupled to gas chromatography-mass spectrometry for the determination of volatile compounds in macroalgae, J. Chromatogr. A 1551 (2018) 41–51. [14] J.R. Belmonte-Sanchez, S. Gherghel, J. Arrebola-Liebanas, R.R. Gonzalez, J. Vidal, I. Parkin, A.G. Frenich, Rum classification using fingerprinting analysis
[15]
[16]
[17]
[18]
[19]
[20]
[21] [22]
[23]
[24]
[25]
[26]
339
of volatile fraction by headspace solid phase microextraction coupled to gas chromatography-mass spectrometry, Talanta 187 (2018) 348–356. N. Kamarulzaman, N. Le-Minh, R. Stuetz, Identification of VOCs from natural rubber by different headspace techniques coupled using GC–MS, Talanta 191 (2019) 535–544. D. Li, H. Duan, Y. Ma, W. Deng, Headspace-sampling paper-based analytical device for colorimetric/SERS dual sensing of sulfur dioxide in wine, Anal. Chem. 90 (2018) 5719–5727. D. Orazbayeva, B. Kenessov, E. Psillakis, D. Nassyrova, M. Bektassov, Determination of transformation products of unsymmetrical dimethylhydrazine in water using vacuum-assisted headspace solid-phase microextraction, J. Chromatogr. A 1555 (2018) 30–36. R. Jain, M.K. Gupta, A. Chauhan, V. Pandey, M.K. Reddy-Mudiam, Ultrasound-assisted dispersive liquid-liquid microextraction followed by GC–MS analysis for the determination of valproic acid in urine samples, Bioanalysis 7 (2015) 2451–2459. D. Nechaeva, A. Shishov, S. Ermakov, A. Bulatov, A paper-based analytical device for the determination of hydrogen sulfide in fuel oils based on headspace liquid-phase microextraction and cyclic voltammetry, Talanta 183 (2018) 290–296. M.A. Farajzadeh, K. Farhadi, A.A. Matin, P. Hashemi, A. Jouyban, Headspace solid-phase microextraction-gas chromatography method for the determination of valproic acid in human serum, and formulations using hollow-fiber coated wire, Anal. Sci. 25 (2009) 875–879. J.B. Stephenson, M.L. Flater, L.T. Bain, Analysis of valproic acid, salicylic acid and lbuprofen in whole blood by GC–MS, J. Anal. Toxicol. 40 (2016) 649–652. C.H. Deng, N. Li, J. Ji, B. Yang, G.L. Duan, X.M. Zhang, Development of water-phase derivatization followed by solid-phase microextraction and gas chromatography/mass spectrometry for fast determination of valproic acid in human plasma, Rapid Commun. Mass Spectrom. 20 (2006) 1281–1287. A.A. Matin, P. Biparva, H. Amanzadeh, K. Farhadi, Zinc/aluminum layered double hydroxide-titanium dioxide composite nano sheet film as novel solid phase microextraction fiber for the gas chromatographic determination of valproic acid, Talanta 103 (2013) 207–213. I. Dominguez, F. Arrebola, R. Gavara, J. Vidal, A. Frenich, Automated and simultaneous determination of priority substances and polychlorinated biphenyls in wastewater using headspace solid phase microextraction and high resolution mass spectrometry, Anal. Chim. Acta 1002 (2018) 39–49. L. Rhoden, M.V. Antunes, P. Hidalgo, C.A. da Silva, R. Linden, Simple procedure for determination of valproic acid in dried blood spots by gas chromatography-mass spectrometry, J. Pharm. Biomed. Anal. 96 (2014) 207–212. R. Markus, W. Jessica, A.W. Armin, H.M. Hans, Development, validation, and application of a fast and simple GC–MS method for determination of some therapeutic drugs relevant in emergency toxicology, Ther. Drug Monit. 33 (2011) 649–653.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Assay of dried blood spot from finger prick for sodium valproate via ink auxiliary headspace gas chromatography mass spectrometry Meng-zhe Guo a , Lili Shao a , Xi Chen a , Hai-juan Li a , Liang Wang a,b , Yuan-jiang Pan c,∗ , Dao-quan Tang a,∗ a
Jiangsu Key Laboratory of New Drug Research and Clinical Pharmacy, Xuzhou Medical University, Xuzhou, Jiangsu 221004, China Department of Bioinformatics, Xuzhou Medical University, Xuzhou, Jiangsu 221004, China c Department of Chemistry, Zhejiang University, Hangzhou, Zhejiang 310058, China b
a r t i c l e
i n f o
Article history: Received 30 April 2019 Received in revised form 21 May 2019 Accepted 22 May 2019 Available online 24 May 2019 Keywords: Ink auxiliary Dried blood spot Sodium valproate Epilepsy Headspace gas chromatography mass spectrometry
a b s t r a c t Sodium valproate is the most commonly used antiepileptic drug that patients need to keep taking over a long period of time or on a permanent basis. Its blood concentration should be accurately detected to avoid toxicity or side-effects, especially for children and the aged. Dried blood spot (DBS) sampling from finger prick is a minimally invasive and patient-friendly procedure for blood collection. However, there are few studies about rapid detection of sodium valproate in DBS samples in current literatures. In this work, we developed an ink auxiliary headspace gas chromatography mass spectrometry (GC–MS) strategy for direct detection of sodium valproate in DBS from epilepsy patients, which does not need extra solvent extraction or elution. It was discovered that carbon black ink could provide better capacity of heat absorption and dissociation, and higher quality of headspace sampling. The detection sensitivity has been improved with reported headspace GC–MS methods, and the limit of quantitation could reach to 200 ng/mL. Finally, this strategy was practically applied to quantify sodium valproate in DBS samples from 29 epilepsy patients. The result showed higher accuracy with lower relative errors by comparing with the clinical immunoassay results. In conclusion, we developed a direct detection method for DBS samples that is suitable for high-throughput clinical test with great potential for clinical application. © 2019 Elsevier B.V. All rights reserved.
1. Introduction Sodium valproate is the first anti-epilepsy medication approved by the Food and Drug Administration (FDA) and considered as the most commonly used antiepileptic drug for monotherapy treatment in the worldwide [1]. This drug needs to be taken for a long-term period and its concentration in blood should be closely monitored because the effective concentration is close to the toxic concentration [2,3]. In particular, personalized administration is necessary for sodium valproate to avoid toxicity on the basis of its efficacy [4]. However, blood concentration monitoring of sodium valproate is often difficult to perform because existing instrument requires at least 150 L of serum sample for a single test, which is equivalent to 0.5–2 mL of blood sample. In addition, external conditions such as collection, storage, and transportation of blood
∗ Corresponding authors. E-mail addresses: [email protected] (Y.-j. Pan), [email protected] (D.-q. Tang). https://doi.org/10.1016/j.chroma.2019.05.039 0021-9673/© 2019 Elsevier B.V. All rights reserved.
samples will also affect the accuracy of detection. Therefore, more advanced blood collection method needs to be developed. Dried blood spot (DBS) is an economical and easy-to-perform method, which enables self-sampling plausible for patients at home or outpatient service [5,6]. So far, the method has achieved numerous benefits and become acclaimed in bioanalytical, clinical and pharmaceutical fields [7–9]. For instance, stable whole blood sample can be collected in DBS with no need for anticoagulant or serum/plasma separation during shipment and storage [10]. In addition, DBS requires a small volume of blood from finger prick, which is an easy, minimally invasive, and patient-friendly approach for blood collection [11]. Although DBS sampling method has a series of advantages, there are great challenges for the detections of targets within the samples. Since the whole blood stored in DBS is low in content, the detection technique needs to be more sensitive and overcome the matrix effects of the blood [12]. Moreover, the compounds in DBS will go through a complex elution or extraction process, and the efficiency of desorption directly affects the analytical sensitivity. Thus, a DBS detection method with simple and efficient sample desorption is urgently needed.
336
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
The headspace gas chromatography mass spectrometry (GC–MS) technique provides a variety of advantages for DBS detection [13–16]. First of all, it promotes the separation of volatile contents from complex matrix samples with the introduction of headspace sampling [17,18]. Secondly, the headspace sampling can simply desorb the samples through heat [19]. Until now, some researchers have used this technique in the detection of sodium valproate in DBS [20]. However, the sensitivity of headspace GC–MS will be affected by the desorption degree of sodium valproate. To solve this issue, sample pre-treatment methods such as headspace solid-phase microextraction (HSSPME) or derivatization have been developed to improve the detection sensitivity [21–23]. However, these methods are time consuming with complicated operations, and need extra enrichment materials, which is not suitable for large-scale clinical testing [24]. In this study, we developed a straightforward headspace GC–MS method for fast detection of sodium valproate in DBS. After multiple rounds of trial-and-error screening, we identified black ink being the best assistance to enhance desorption of sodium valproate. In addition, various parameters of detection were also optimized, which includes the size of DBS, the type of black ink, the kind and content of extra infiltrating solvent, and the temperature of headspace sampling. The mechanism of ink-assisted desorption has also been attempted to explore. Finally, we applied this method for the detection of practical DBS samples from epilepsy patients. This method achieved better sensitivity and accuracy by comparing with the existing clinical laboratory methods. 2. Experiments 2.1. Materials and chemicals Acetone and ethanol were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Methanol, acetonitrile, acetic acid, and ammonium acetate were of chromatographic grade and purchased from MERDA (USA). Sodium valproate (95%) was purchased from Energy Chemical (Shanghai, China). Carbon black ink and prepared Chinese ink were purchased from HERO and “Yi De Ge”(China). The pencils were purchased from Mitsu-Bishi (Japan) (Fig. S1). 2.2. The selection of ink and optimization of headspace GC–MS method Firstly, different processing methods of DBS were compared, containing black ink of pen assisted, ink of Chinese brush assisted, printing black ink assisted, and pencils with different blackness assisted. 10 L of ink was added into the DBS. Secondly, the best method was chosen as the final processing method of DBS. Moreover, other parameters of headspace sampling were optimized, such as the size of filter paper, kinds of infiltrating solvent, as well as volume and additive, and the temperature of headspace sampling. 2.3. The investigation of the mechanism of black ink auxiliary improving the sensitivity of headspace GC–MS detection The mechanism of black ink auxiliary improving the sensitivity of headspace GC–MS detection has been investigated by several methods. Firstly, infrared camera was used to obtain the heat absorption of blank filter paper, ink assisted filter paper, and pencil assisted filter paper after laser radiation. Then scanning electron microscope (SEM) was used to achieve the fine surface structure of filter paper with ink and pencil assisted. Finally, three kinds of filter papers mentioned above were added sodium valproate and passed through the solvent elution and normal GC–MS detection.
The result may indicate other enhancement mechanism by ink auxiliary. 2.4. Application of ink auxiliary headspace GC–MS method in the quantitative determination of sodium valproate from finger prick DBS of epilepsy patients The DBS samples including 29 epilepsy patients taking sodium valproate were provided by Xuzhou Central Hospital. The experiments were approved by the ethics committee. Method validation has been taken, containing quantitative linear, precision and accuracy, stability, and recovery. Then DBS samples were detected by our developed headspace GC–MS strategy. SPSS 16.0 was used for the statistical analysis, and the detection results were compared with the actual clinical detection results from immunoassay to confirm the accuracy of our method. The t test was used to identify the statistic differences. 2.5. GC–MS conditions Analysis was performed by GC–MS using an Agilent 7890A GC interfaced with the 5975C single-quadrupole mass spectrometry equipped with electron ionization (EI) (Santa Clara, CA, USA). The GC separation was performed on an HP-5MS column (60 m × 0.25 mm i.d. × 0.25 m film thickness, Agilent). The chromatographic separation mode was achieved using the following temperature program: initial temperature of 70 ◦ C for 1 min; raised to 200 ◦ C by 20 ◦ C min; raised to 260 ◦ C by 50 ◦ C min and kept for 5 min. The inlet temperature was 260 ◦ C. The carrier gas was helium. The flow rate was 1 mL/min and the split ratio was 10 to 1. The ion source was electron impact ion source with 70 eV of collision energy. The detection was SIM mode with 0–5 min of solvent delay. Three chosen reference ions were m/z 57, 73, and 102, respectively. The reference standard of sodium valproate was spiked into the blank whole blood from heather to prepare the 200 g/mL of standard solution. Then the standard solution was dilution using the blank matrix to 150, 120, 100, 80, 60, 50, 40, 15, 1 g/mL. Finally bring 10 L of each solution into the filter paper respectively to prepare the standards for calibration. The quality control (QC) samplers were similarly prepared. 3. Results and discussion 3.1. Optimization of ink auxiliary headspace GC–MS method Pilot study showed that filter paper with black surface could improve heat absorption, leading to enhance desorption of samples during headspace sampling. Therefore, the blackening process would contribute to DBS assay to strengthen the detection sensitivity of sodium valproate in headspace GC–MS. In this work, four types of black substances were investigated, which include fountain pen carbon black ink, prepared Chinese ink, black pencils with different blackness, and ink for printing. According to the result, the carbon black ink showed the best enhancement of sensitivity when compared with the other three materials in the headspace GC–MS detection. Thus, we used this substance as a supplement and developed an integrated strategy for the determination of DBS. In this strategy as shown in Fig. 1, other parameters have also been optimized. Firstly, the size of DBS and the volume of blood were studied. Optimal DBS size was 7 mm square. With this size, the best volume of blood was 10 L, which could be conveniently collected via finger prick. In addition, we found that the organic solvent could improve desorption in headspace sampling after the blackening of DBS. This auxiliary solvent could help to carry the sodium valproate into the gas by evaporation due
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
337
Fig. 1. The ink auxiliary headspace GC–MS strategy for the analysis of sodium valproate in DBS.
Table 1 The optimization of several parameters in ink auxiliary headspace GC–MS analysis. Parameters
Optimization
DBS treatment
White filter paper; Pencil auxiliary; Ink auxiliary Acetone; Methanol; Ethanol; Acetonitrile 0.1% Formic acid; 0.2% Formic acid; 0.3% Formic acid; 0.4% Formic acid; 0.5% Formic acid 4 × 4 mm; 7 × 7 mm; 10 × 10 mm; 12 × 12 mm; 14 × 14 mm 140; 160; 180; 190; 200
Auxiliary solvent Additive in solvent DBS Size Headspace temperature (◦ C)
to the low boiling point. Therefore, organic solvent was also optimized, which involves type, volume, and the additive. As shown in Table 1, acetone was identified as the preferred solvent by comparing with methanol, ethanol, and acetonitrile, and the required acetone volume was only 50 L. Additionally, 0.4% of formic acid was also chosen as an additive to improve desorption of sodium valproate through its ion inhibition effect. Finally, the temperature of headspace sampling was investigated and the best dissociation temperature was obtained at 180 ◦ C. After addition of the carbon black ink and adoption of optimized parameters, the detection sensitivity increased about 200 times than the direct DBS headspace GC–MS assay. 3.2. The mechanism of black ink auxiliary for improving the sensitivity of headspace GC–MS detection The mechanism of black ink auxiliary for improving desorption of sodium valproate in headspace GC–MS was investigated. Three types of filter papers, including blank control, blackened by ink and pencil were put into the headspace bottle and heated, respectively. After 30 s, these filter papers were detected by infrared camera to obtain the surface thermal values. As the result, the surface temperature of blank filter paper was 31 ◦ C while filter papers disposed by ink and pencil reached to 50 and 60 ◦ C, respectively. The increased heat absorption of filter paper can help desorption of sodium valproate. SEM was used to observe the microstructure of treated filter paper. As shown in Fig. S2, comparing with the blank filter paper, there were multilayer carbon powders covering the fiber surface of the filter papers treated by ink with small and evenly-distributed
particles. Therefore, we hypothesized that the carbon powder could provide extra assistance for desorption of the sodium valproate from filter paper. To prove this assumption, 100 L of acetone was used as the solvent for the elution of sodium valproate from DBS and the eluant was directly detected by GC–MS without headspace sampling. The results showed that the solvent from ink auxiliary DBS had the highest content of sodium valproate, demonstrating that nanoscale carbon powder in the surface of filter paper could indeed help the elution and desorption. 3.3. Application of ink auxiliary headspace GC–MS in quantitative determination of sodium valproate in DBS from epilepsy patients Following the optimization of ink auxiliary headspace GC–MS strategy, method validation was then performed. As shown in Tables S1 and S2, the result demonstrated a good linearity in the range of 1–200 g/mL. The correlation coefficient (r) was calculated to be greater than 0.99 and the limit of quantitation was 200 ng/mL. This linear range can fully meet the detection requirement of sodium valproate in DBS, which is better than reported headspace GC–MS method [21–23,25,26]. In addition, both interbatch and intra-batch accuracy and precision showed the favorable accuracy from 96.55% to 105.14%, with relative standard deviation (RSD) from 4.76% to 10.18%. In addition, we also applied the headspace GC–MS strategy into the quantification of sodium valproate in DBS from epilepsy patients. To investigate the accuracy of our method in practical assay, drug concentrations of these patients were also analyzed by existing clinical tests of immunoassay. The 29 patients have the ages from 4 to 87 years and are gender-balanced, which indicates a sufficient coverage of all kinds of people taking the drug. Moreover, drug concentrations were measured from 19 to 131 g/mL, representing the concentration range of the chemical in clinical tests of immunoassay. According to the results (Table 2 and Fig. 2), our method had higher accuracy when compared with immunoassay clinical tests. The median of errors was 3.57%. Only two blood concentrations were obviously different from the immunoassay clinical tests. These two concentrations were the lowest among the 29 samples (19 g/mL and 20 g/mL from the immunoassay clinical tests, respectively). However, it was not necessary to mean that our method had the wrong detection results because the exist-
338
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
Table 2 The determination of 29 DBS samples from epilepsy patients and its comparison with the clinical test. Number
Age
Sex
Clinical concentration (g/ml)
Measured concentration (Mean ± SD)/ (g/ml)
Deviation(%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
46 64 45 30 17 36 63 13 50 17 44 35 43 33 76 57 44 24 10 87 4 45 55 4 29 42 42 56 72
Female Female Female Male Male Male Female Female Male Male Female Male Male Male Male Male Male Male Male Male Male Male Female Female Male Male Male Male Male
131.00 130.00 124.00 120.00 117.00 112.00 104.00 97.00 89.00 85.00 77.00 75.00 73.00 65.00 64.00 64.00 59.00 58.00 56.00 46.00 43.00 41.00 39.00 38.00 36.00 33.00 27.00 20.00 19.00
130.46 ± 14.15 132.97 ± 8.45 124.07 ± 6.30 131.07 ± 7.32 117.50 ± 11.03 112.92 ± 9.22 99.55 ± 2.08 81.73 ± 3.08 80.06 ± 3.87 79.58 ± 4.76 75.83 ± 4.71 71.23 ± 6.67 72.44 ± 4.40 63.18 ± 4.17 62.20 ± 6.41 64.64 ± 1.33 61.11 ± 1.81 58.54 ± 4.91 55.88 ± 1.67 41.17 ± 2.90 48.70 ± 2.41 45.65 ± 0.66 35.37 ± 1.53 40.37 ± 6.45 42.04 ± 2.79 40.59 ± 2.93 27.03 ± 3.02 35.04 ± 0.79 28.35 ± 0.62
−0.41% 2.28% 0.05% 9.22% 0.43% 0.82% −4.28% −15.74% −10.05% −6.38% −1.52% −5.02% −0.77% −2.79% −2.81% 1.00% 3.57% 0.93% −0.21% −10.51% 13.26% 11.35% −9.32% 6.23% 16.77% 22.99% 0.12% 75.18% 49.21%
drug concentration due to the effect of structural analogs such as metabolites. 4. Conclusion In this study, we developed an ink auxiliary headspace GC–MS strategy for the analysis of sodium valproate in DBS from epilepsy patients. The detection sensitivity was improved about 200 times with the reported direct DBS headspace GC–MS assay, and the limit of quantitation could reach to 200 ng/mL. In addition, this strategy has been successfully applied into the determination of sodium valproate from 29 epilepsy patients. By comparing with the clinical test results, our detection showed higher accuracy with lower relative errors. In sum, we developed a novel method for quantitative analysis of sodium valproate in DBS from finger prick with convenient storage and transportation, which is suitable for high-throughput clinical tests and shows great potential for clinical applications in future. Conflict of interest The authors declare that they have no conflict of interest. Acknowledgements We appreciate the support of Natural Science Foundation of China (No. 21505116), Jiangsu Natural Science Foundation of China (Nos. BK20171183 and BK20181147), Jiangsu University Natural Science Foundation of China (No. 16KJA350001), and Natural Science Foundation of Xuzhou City (Nos. KC17187 and KC18050).
Fig. 2. Comparison of the headspace GC–MS method and clinical test (a) and its relative errors (b) based on 29 epilepsy patients, the error bar stands for the standard deviation.
Appendix A. Supplementary data
ing clinical immunoassay method also lacks of enough accuracy in the detection of very low concentrations in DBS samples. Moreover, the result of immunoassay is commonly higher than the real
Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.chroma.2019. 05.039.
M.-z. Guo et al. / J. Chromatogr. A 1601 (2019) 335–339
References [1] L. Gu, B.Y. Liang, Q. Chen, J.X. Long, J.J. Xie, G.L. Liang, Y. Yan, J.J. Tan, W.H. Dou, W. Chen, P. Wu, J.P. Wang, L. Su, Prevalence of epilepsy in the People’s Republic of China: a systematic review, Epilepsy Res. 105 (2013) 195–205. [2] S. Balestrini, S. Sisodiya, Pharmacogenomics in epilepsy, Neurosci. Lett. 667 (2018) 27–39. [3] L. Walker, N. Mirza, V. Yip, A. Marson, M. Pirmohamed, Personalized medicine approaches in epilepsy, J. Intern. Med. 227 (2015) 218–234. [4] S. Cremers, N. Guha, B. Shine, Therapeutic drug monitoring in the era of precision medicine: opportunities, Br. J. Clin. Pharmacol. 82 (2016) 900–902. [5] C. Linder, M. Neideman, K. Wide, M. von Euler, L.L. Gustafsson, A. Pohanka, Dried blood spot self-sampling by guardians of children with epilepsy is feasible: comparison with plasma for multiple antiepileptic drugs, Ther. Drug Monit. (2019), http://dx.doi.org/10.1097/FTD.0000000000000605. [6] C. Linder, A. Hansson, S. Sadek, L.L. Gustafsson, A. Pohanka, Carbamazepine, lamotrigine, levetiracetam and valproic acid in dried blood spots with liquid chromatography tandem mass spectrometry; method development and validation, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 1072 (2018) 116–122. [7] C. Linder, K. Wide, M. Walander, O. Beck, L.L. Gustafsson, A. Pohanka, Comparison between dried blood spot and plasma sampling for therapeutic drug monitoring of antiepileptic drugs in children with epilepsy: a step towards home sampling, Clin. Biochem. 50 (2017) 418–424. [8] C. Lander, M. Andersson, K. Wide, O. Beck, A. Pohanka, A LC–MS/MS method for therapeutic drug monitoring of carbamazepine, lamotrigine and valproic acid in DBS, Bioanalysis 7 (2015) 2031–2039. [9] A. Pohanka, M. Mahindi, M. Masquelier, L.L. Gustafsson, O. Beck, Quantification of valproic acid in dried blood spots, Scand. J. Clin. Lab. Invest. 74 (2014) 648–652. [10] S.T. Kong, S.H. Lim, E. Chan, P.C. Ho, Estimation and comparison of carbamazepine population pharmacokinetics using dried blood spot and plasma concentrations from people with epilepsy: the clinical implication, J. Clin. Pharmacol. 54 (2014) 225–233. [11] I. Ryona, J. Henion, A book-type dried plasma spot card for automated flow-through elution coupled with on-line SPE-LC-MS/MS bioanalysis of opioids and stimulants in blood, Anal. Chem. 88 (2016) 11229–11237. [12] J. Deglon, A. Thomas, Y. Daali, E. Lauer, C. Samer, J. Desmeules, P. Dayer, P. Mangin, C. Staub, Automated system for on-line desorption of dried blood spots applied to LC/MS/MS pharmacokinetic study of flurbiprofen and its metabolite, J. Pharm. Biomed. Anal. 54 (2011) 359–367. [13] A. Maruti, E. Duran-Guerrero, C. Barroso, R. Castro, Optimization of a multiple headspace sorptive extraction method coupled to gas chromatography-mass spectrometry for the determination of volatile compounds in macroalgae, J. Chromatogr. A 1551 (2018) 41–51. [14] J.R. Belmonte-Sanchez, S. Gherghel, J. Arrebola-Liebanas, R.R. Gonzalez, J. Vidal, I. Parkin, A.G. Frenich, Rum classification using fingerprinting analysis
[15]
[16]
[17]
[18]
[19]
[20]
[21] [22]
[23]
[24]
[25]
[26]
339
of volatile fraction by headspace solid phase microextraction coupled to gas chromatography-mass spectrometry, Talanta 187 (2018) 348–356. N. Kamarulzaman, N. Le-Minh, R. Stuetz, Identification of VOCs from natural rubber by different headspace techniques coupled using GC–MS, Talanta 191 (2019) 535–544. D. Li, H. Duan, Y. Ma, W. Deng, Headspace-sampling paper-based analytical device for colorimetric/SERS dual sensing of sulfur dioxide in wine, Anal. Chem. 90 (2018) 5719–5727. D. Orazbayeva, B. Kenessov, E. Psillakis, D. Nassyrova, M. Bektassov, Determination of transformation products of unsymmetrical dimethylhydrazine in water using vacuum-assisted headspace solid-phase microextraction, J. Chromatogr. A 1555 (2018) 30–36. R. Jain, M.K. Gupta, A. Chauhan, V. Pandey, M.K. Reddy-Mudiam, Ultrasound-assisted dispersive liquid-liquid microextraction followed by GC–MS analysis for the determination of valproic acid in urine samples, Bioanalysis 7 (2015) 2451–2459. D. Nechaeva, A. Shishov, S. Ermakov, A. Bulatov, A paper-based analytical device for the determination of hydrogen sulfide in fuel oils based on headspace liquid-phase microextraction and cyclic voltammetry, Talanta 183 (2018) 290–296. M.A. Farajzadeh, K. Farhadi, A.A. Matin, P. Hashemi, A. Jouyban, Headspace solid-phase microextraction-gas chromatography method for the determination of valproic acid in human serum, and formulations using hollow-fiber coated wire, Anal. Sci. 25 (2009) 875–879. J.B. Stephenson, M.L. Flater, L.T. Bain, Analysis of valproic acid, salicylic acid and lbuprofen in whole blood by GC–MS, J. Anal. Toxicol. 40 (2016) 649–652. C.H. Deng, N. Li, J. Ji, B. Yang, G.L. Duan, X.M. Zhang, Development of water-phase derivatization followed by solid-phase microextraction and gas chromatography/mass spectrometry for fast determination of valproic acid in human plasma, Rapid Commun. Mass Spectrom. 20 (2006) 1281–1287. A.A. Matin, P. Biparva, H. Amanzadeh, K. Farhadi, Zinc/aluminum layered double hydroxide-titanium dioxide composite nano sheet film as novel solid phase microextraction fiber for the gas chromatographic determination of valproic acid, Talanta 103 (2013) 207–213. I. Dominguez, F. Arrebola, R. Gavara, J. Vidal, A. Frenich, Automated and simultaneous determination of priority substances and polychlorinated biphenyls in wastewater using headspace solid phase microextraction and high resolution mass spectrometry, Anal. Chim. Acta 1002 (2018) 39–49. L. Rhoden, M.V. Antunes, P. Hidalgo, C.A. da Silva, R. Linden, Simple procedure for determination of valproic acid in dried blood spots by gas chromatography-mass spectrometry, J. Pharm. Biomed. Anal. 96 (2014) 207–212. R. Markus, W. Jessica, A.W. Armin, H.M. Hans, Development, validation, and application of a fast and simple GC–MS method for determination of some therapeutic drugs relevant in emergency toxicology, Ther. Drug Monit. 33 (2011) 649–653.