Quantitative Measurement of Plasma Free Metanephrines by a Simple and Cost-Effective Microextraction Packed Sorbent with Porous Graphitic Carbon and Liquid Chromatography-Tandem Mass Spectrometry

Plasma free metanephrines are widely used for the diagnosis of pheochromocytoma and paraganglioma (PPGL), yet quantifying metanephrines using a simple and cost-effective approach may be challenging due to preanalytical and analytical constraints. In this study, we established and validated a new method for quantitative measurement of plasma free metanephrines based on microextraction by packed sorbent (MEPS) with porous graphitic carbon (PGC) and liquid chromatography-tandem mass spectrometry (HILIC-MS/MS). The elution step was fully compatible with HILIC mode without evaporation and reconstitution. The analytes were well resolved, and potential interferences (54 substances) were investigated. This method was linear from 24.7–2717 pg/mL for metanephrine (MN) and 24.5–4010 pg/mL for normetanephrine (NMN) with a coefficient of determination (R2) higher than 0.994. The limit of MN and NMN detection were 12.4 pg/mL and 12.3 pg/mL, respectively. The intra- and interassay impressions were ≤12.8% for spiked quality controls and ≤13.6% for commercial quality controls; the method recoveries ranged within 88.0–109.0%, respectively. The area under the receiver operating characteristic (ROC) curve was 0.848 ± 0.047 for MN and 0.979 ± 0.021 for NMN. Validation that was performed by comparing clinical specimens with various biochemical results showed that plasma free metanephrines in a seated position had comparable sensitivity and lower specificity to urinary free metanephrines, which could be compensated by combining other biochemical tests. The newly developed MEPS method resulted as a time-saving, reliable, and cost-effective microextraction technique that can be applied for a successful screening of PPGL.


Introduction
Pheochromocytoma and paraganglioma (PPGL) are rare neuroendocrine tumors characterized by the overproduction of catecholamines [1], which can lead to serious cardiovascular complications, such as hypertensive crisis, myocardial infarction, and stroke [2]. e biochemical testing is crucial for diagnosing PPGL. Compared with plasma catecholamines and urinary catecholamines, plasma free metanephrines (O-methylated metabolites of catecholamines) provide higher sensitivity and specificity for the diagnosis of these tumors and thus have been recommended as first-line testing along with urinary fractionated metanephrines [3]. ese measurements commonly include metanephrine (MN) and normetanephrine (NMN), the O-methylated metabolite of epinephrine (EPI), and norepinephrine (NE).
Liquid chromatography with electrochemical detection (LC-ECD) is one of the commonly used methods for measuring plasma metanephrines [4,5], yet this approach requires large sample volumes, laborious sample preparation, and long chromatographic time to achieve the sensitivity. Immunoassay-based methods provide an alternative to LC-ECD; however, the nonspecific binding and cross-reactivity cannot be avoided using this approach [6,7]. Over recent years, liquid chromatography with tandem mass spectrometry (LC-MS/MS) has been widely used to measure plasma free metanephrines due to its inherent selectivity and sensitivity [8][9][10][11][12][13][14][15][16]. However, these compounds are poorly ionizable and with high polarity, which presents at very low concentrations in a complex matrix. Hence, an effective, simple, sensitive, and cost-effective sample cleanup for the complex matrix prior to the LC-MS/MS analysis is crucial.
Over the past decades, general solid-phase extraction (SPE) has been often used as the sample preparation step in the analysis of plasma metanephrines [9][10][11][12][13][14][15][16]. Weak cation-exchange SPE sorbents (WCX) have become popular for the isolation of metanephrines from biological samples [10][11][12][13][14][15][16]. Besides, the ion-pairing reagent can also be used to form a complex with metanephrines, which could be retained on the C18 SPE cartridges [9]. In these studies, extract evaporation for concentration or use nonvolatile salts as buffers or a large volume of sample (500 μL [9] or 900 μL [12]) and injection volume (35 μL [11] or 40 μL [9]) were required to achieve the necessary sensitivity and specificity. ese procedures can increase the complexity and time requirements and shorten the life of the analytical column.
Porous graphitic carbon (PGC), which has particular physical and chemical properties, was developed as a stationary chromatographic phase for LC in the early 1980s. e very polar compounds, including ionized molecules, can be retained on PGC without using ion-pairing agents. is unique property comes from its delocalized band of electrons, enabling stabilizing electronic interactions and direct π overlap with diverse analytes [17]. e PGC stationary phases have been exploited to analyze polar pharmaceutical compounds [18], impurities [19], or endogenous metabolites. Currently, the PGC sorbent has become commercially available in SPE and microextraction by packed sorbent (MEPS) formats. MEPS is a miniaturized form of the SPE technique that has been accepted as an attractive option and a powerful sample preparation approach suitable for accomplishing bioanalytical challenges since it emerged in 2004 [20]. e greatest advantage of this method is its clean and straightforward MEPS procedure with a short runtime. It can handle a small sample volume while maintaining sufficient selectivity, precision, and accuracy [21]. However, to the best of our knowledge, only several MEPS methods using the amino-propyl silane (APS) or C18 sorbents have been introduced for the sample cleanup of several biogenic amines [22][23][24]. e aim of this work was to develop and validate an analytical method for the quantitation of metanephrines by employing simple and specific MEPS of minimal plasma sample compatible with hydrophilic interaction chromatography (HILIC) mode that is more suitable for the LC-MS/ MS analysis of hydrophilic compounds [25]. A large number of potential interferences of plasma metanephrines were also investigated. Moreover, this study evaluated its usefulness by comparing it with other metabolites, including urinary free catecholamines, metanephrines, and vanillylmandelic acid (VMA) in a routine clinical setting.

Materials and Chemicals.
MN and NMN were purchased from Toronto Research Chemicals (Ontario, Canada). MN-d3 and NMN-d3 were obtained from Cambridge Isotope Laboratories, Inc. (MA, USA). Endocrine plasma controls and plasma calibration standard metanephrines were acquired from Chromsystems (Grafelfing, Munich, Germany). HPLC-(high-performance liquid chromatography-) grade acetonitrile, methanol, formic acid, acetic acid, and ammonium formate were obtained from ermo Fisher Scientific Inc. (MA, USA). Ultrapure water was obtained using a Milli-Q system (Waters Millipore, MA, USA).

LC-MS/MS Conditions.
e LC-MS/MS system was composed of an API 4000+ triple quadrupole mass spectrometer (AB Sciex Instruments, CA, USA) equipped with a Shimadzu LC-30A HPLC system (Kyoto, Japan). e mass spectrometer was operated in the positive electrospray ionization (ESI) mode with multiple reaction monitoring (MRM). e column oven was kept at 40°C, and the autosampler was set at 4°C. e chromatographic separation was performed on a ZIC-HILIC column (150 mm × 2.1 mm, 3.5 μm) from Merck (Darmstadt, Germany). e analytes were eluted by gradient mobile phases containing acetonitrile (Mobile Phase A) and 40 mM ammonium formate (Mobile Phase B, pH 3.0). e initial LC gradient of 83% A was linearly decreased to 75% within 3 min and held for 1 min. e gradient was then quickly ramped to 60% A in 0.1 min and held for 1 min; finally, the column was equilibrated at 83% for 3 min. e flow rate was 0.35 mL/min, and the cycle time was 8 min. e injection volume was 5 μL.
Analyst software, version 1.6.2, from AB Sciex was used for instrument control and analysis. e ion spray voltage was optimized to 1500 V at 550°C, and the curtain gas was maintained at 45 psi. e nebulizer gas and the heater gas were set at 40 psi, while the collisionally activated dissociation (CAD) gas was kept at 12 psi. e MRM transitions and compound parameters are summarized in Table S1, with 100 ms dwell time for each transition.

Standards and Specimens.
MN and NMN standard stock solutions were prepared in 0.2 M acetic acid at a concentration of 1 mg/mL. Optimization of ion source parameters was performed by infusing a metanephrines solution (500 ng/mL in 50% methanol containing 0.2 M acetic acid) at a flow rate of 5 μL/min. e internal standard (IS) working solution, including MN-d3 (5 ng/mL) and NMN-d3 (40 ng/ mL), was prepared in 0.2 M acetic acid in water.
e Chromsystems plasma calibration standards were reconstituted according to the reagent instruction. A sixcalibrator curve was prepared and stored at −20°C until use.
is material was stable for at most 3 months. For the method validation, quality control (QC) samples were prepared at three concentrations (low: 100 pg/mL for MN, 150 pg/mL for NMN; middle: 500 pg/mL for MN, 750 pg/mL for NMN; high: 2000 pg/mL for MN, 3500 pg/mL for NMN) of plasma samples. For the routine clinical analysis, normal-concentration (60 pg/mL for MN and 100 pg/mL for NMN) and pathological-concentration (1500 pg/mL for MN and 7000 pg/mL for NMN) controls were also prepared by reconstituting Chromsystems normal-and pathologicalrange controls according to the reagent instruction, respectively. Subsequently, pathological-range control was diluted 1 : 2 with blank plasma from Chromsystems to ensure that NMN fell within the linearity range. Both controls were aliquoted and kept at −20°C for storage up to 2 months.

Sample Collection and Preparation.
EDTA plasma specimens were selected from samples submitted for clinical testing. Samples were de-identified and managed according to the guidelines approved by the Institutional Review Board of the Peking University ird Hospital.
A type of MEPS containing PGC sorbents (particle size: 30 μm; pore size: 250Å; ermo Fisher Scientific, Australia) and a hand-held automated analytical syringe (eVol; ermo Fisher Scientific, Australia) was employed for the plasma extraction.
e new sorbent was activated in 3 × 200 μL of methanol and 3 × 200 μL of water before extraction. e first step was protein precipitation with methanol. Briefly, 100 μL of methanol was added to 100 μL of plasma sample containing 10 μL of IS working solution. e precipitated sample was then centrifuged at 15000 rpm for 5 min. e supernatant was diluted with 1000 μL of water, and then it was drawn up and down through syringe (7 × 200 μL) without discarding it. e sorbent was washed with 50 μL of water, and analytes were eluted with 40 μL of mobile phase solution. After each extraction, the sorbent was rinsed with 4 × 100 μL of a mixture of acetonitrile:water (8 : 2, v/v, containing 2% formic acid) followed by 2 × 100 μL of water. e MEPS could reuse the sorbents up to 50 times. e schematic diagram of the operation course of MEPS is shown in Figure S1.

Method Validation.
e method was validated according to the Clinical and Laboratory Standards Institute (CLSI) C62-A and FDA guidelines. Linearity was assessed by analyzing calibrators at six concentrations on three different days. e lower limit of quantification (LLOQ) was defined as the lowest concentration within ±20% of imprecision and accuracy. e limit of detection (LOD) was defined as the signal-to-noise ratio >3. To evaluate imprecision and accuracy in the plasma analysis, five replicates of three control materials were repeated on three consecutive days (a total of 15 measurements at each level of control). Additionally, normal and pathological levels of QC samples from Chromsystems were processed alongside, and the results were compared with the expected range provided by the manufacturer. Duplicates of each control were assayed for 20 days. We prepared two spiked samples using pooled plasma to investigate the MEPS recovery: prespiked and postspiked. e prespiked sample was prepared by spiking analytes into pooled plasma before MEPS, whereas the postspiked sample was prepared by spiking analytes at the same concentrations as the extract obtained after MEPS of pooled plasma. e percent extraction recovery was calculated as the percent response ratio of each analyte in the prespiked sample to those in the postspiked sample. A postcolumn infusion study was performed to evaluate the matrix effect by injecting the plasma matrix after MEPS (no IS added) onto the LC column and simultaneously injecting the IS standard solution at 10 μL/min into the column effluent. To evaluate the carryover, a set of high (H) and low (L) samples was assayed in the following consecutive order: 3L, 2H, 1L, 2H, 4L, 2H, 1L, 2H, 1L, 2H, 1L. Analytical specificity was investigated by analyzing a number of endogenous and exogenous compounds. Individual aliquots of middle QC plasma samples were spiked with solutions of potential interferences. Recommendations from CLSI EP7-A2, Interference Testing in Clinical Chemistry, were used to guide prepared interference concentrations. Plasma sample stability was assessed after 24 h storage at room temperature or in a refrigerator (4°C) for 5 days. e postextracts stored at 4°C in the autosampler for 120 h were also tested. At each storage condition, the determination of samples was repeated in triplicates.

Column Chromatography and MS Condition
Optimization. Poor retention on the traditional C18 columns due to high polarity represents one of the challenges when performing chromatographic separation of metanephrines. Another difficulty is well-resolved chromatography of metanephrines and compounds that are structurally similar to them. Although HILIC columns are specifically suitable to retain metanephrines (mixed mode of adsorption and better ionization efficiency) [26], analytical interferences with the measurement of plasma metanephrines in HILIC mode have been recently reported. For example, cross-contamination may occur due to epinephrine/NMN isobaric analyte pairing with identical MRM transitions, 4-hydroxy-3-methoxymethamphetamine (HMMA), a metabolite of 3, 4-methylenedioxymethamphetamine (MDMA), and NMN share a common pharmacophore resulting in the same product ion after fragmentation [27]. An endogenous analyte found in all plasma samples, 3-O-methyldopa, could not be eliminated by multistage fragmentation (MRM 3 ) [28], which may enable more specific target quantification compared with traditional MRM [29]. erefore, chromatographic separation remains essential for interferences involving unresolvable mass fragmentation. Hence, a variety of HILIC-type columns were screened, including Atlantis HILIC silica (150 mm × 2.1 mm, 5 μm) and XBridge Amide (50 mm × 2.1 mm, 3.5 μm) from Waters (Milford, MA, USA), Luna Diol (100 mm × 2.1 mm, 5 μm) from Phenomenex (Torrance, CA, USA), Cosmosil HILIC (100 mm × 2.0 mm, 5 μm) from Nacalai tesque (Kyoto, Japan), and ZIC-HILIC (150 mm × 2.1 mm, 3.5 μm). e ZIC-HILIC column was optimal column for retaining and separating metanephrines with low interferences (detailed in the Analytical Specificity section). Typical chromatograms from a patient without PPGL and those with histologically confirmed PPGL are shown in Figure 1. e two analytes were completely separated within 5.0 min with a runtime of 8.0 min. MS parameters were optimized using a direct flow injection analysis of 500 ng/mL analyte solution in 0.2 M acetic acid in 50% methanol. As previously reported [24,30], the more stable ion signals of [M-H 2 O + 1] + for MN and NMN for metanephrines were observed compared with the expected mass/charge (m/z) ratios of [M + H] + in the ESI source, which was used as the precursor ions. MRM transitions for all the analytes and their deuterated analogues, along with mass detector settings, are described in Table S1.

Optimization of Sample Preparation.
e choice of a suitable sample preparation technique for the isolation of metanephrines from complex biological matrices was not straightforward. MEPS is a novel miniaturized SPE with several advantages, such as shorter sample preparation time, miniaturization of sample volume, lower consumption of organic solvents, and absence of extract evaporation. In this study, pooled plasma samples were used for the development and IS for quantification due to the analytes' endogenous contribution. Several crucial parameters related to MEPS were screened in this assay.

Choice of Sorbent Type.
Since plasma proteins cause clogging of MEPS cartridges, protein precipitation (PP) before loading samples on the MEPS cartridges is often necessary. Several types (methanol, acetonitrile, and mixture) of organic agents were tested in this study. Although a large number of different sorbents are available, the choice of MEPS cartridge is very difficult due to the low retention of metanephrines from the PP solution. First, cation exchange and C18 (particle size: 40-60 μm; pore size: 60Å; ermo Fisher Scientific, Australia) cartridges of MEPS, mostly used in SPE, were selected for testing. However, none of these enabled sufficient retention of metanephrines on the sorbent even when the dilution of PP supernatant (up to 20×) contained a large amount of organic solvents. e organic solvents in supernatant reduced the retention of metanephrines.
Another suitable sorbent for hydrophilic compounds was PGC, recently introduced in the MEPS format [31]. Methanol was chosen as the suitable PP solvent because only the 5 : 1 (water:precipitated sample) dilution ratio was needed to enable the retention of metanephrines on PGC sorbents, while a larger dilution ratio was necessary for acetonitrile as a deproteinization agent. erefore, the PGC sorbent was selected for the MEPS extraction without a large loss of analytes in the loading sample step.

Loading Sample.
e retention of analytes on the PGC sorbent was affected by the number of extraction cycles. e different number of cycles through which the sample went through efficiently increased the extraction. On the other hand, plasma, a complex matrix, could saturate the sorbent phase and block the access of analytes to the adsorption sites. e result indicated that the seven extraction cycles provided the best conditions for metanephrines retained on the PGC sorbent (Figure 2(a)). An increased number of extraction cycles did not improve the extraction efficiency and might shorten the MEPS sorbent's lifetime.

Washing and Eluting.
e sorbent washing step is essential in removing unwanted interferences of the matrices without significant leakage of analyzed compounds. ree different washing solutions (water, 10% methanol in water, and 2% formic acid in water) were assessed for pooled plasma samples. Among these solvents, water was employed as the most appropriate solvent without sacrificing recovery (data not shown). e subsequent stage involved assaying the elution of metanephrines from the sorbent with minimal solvent volume. e elution efficiency was tested using the mobile phase, a mixture solution of acetonitrile and water (8 : 2) containing 2% formic acid, 2% aqueous ammonia, or no additive as solvents (Figure 2(b)). e highest analytical signals among the applied elution solvents corresponded to the mobile phase, which was also the most suitable elution solvent due to the perfect compatibility with HILIC chromatography.

Carryover of MEPS.
We finally investigated the carryover of MEPS, which was considered as a decelerating step in sample preparation. e small amount used in the MEPS cartridge can be easily and effectively washed between extractions. A mixture of acetonitrile and water containing 2% formic acid was used as a suitable washing solution (4 × 100 μL) to reduce carryover for reusing of the MEPS sorbent (Figure 2(c)). Carryover was <0.2% prior to reuse.

Linearity and Analytical Sensitivity.
e responses of the analytes were linear in the calibration range of 24.7-2717 pg/mL for MN (R 2 > 0.9951) and in the calibration range of 24.5-4010 pg/mL for NMN (R 2 > 0.9942), with a coefficient of variation (CV) of ≤7.3% for the slopes of each analyte on three different days. e F-test for lack-of-fit was also tested for linearity; the results of pure error tests for MN (F * � 0.887, P � 0.487) and NMN (F * � 0.154, P � 0.959) showed that the relationship was linear for MN and NMN. e LLOQ were 24.7 pg/mL for MN with a CV ≤13.0% and the relative error (RE) ≤−3.2% and 24.5 pg/mL for NMN with a CV ≤11.0% and the RE ≤−1.8%. e LOD for MN and NMN were 12.4 pg/mL and 12.3 pg/mL, respectively, by gradually diluting calibrators.

Precision, Accuracy, and Extraction Recovery.
e intra-and interassay imprecision for the spiked QC and Chromsystems QC samples were ≤12.8% and ≤13.6%, respectively. e intraday accuracy for the spiked QC samples was within the range, from 94.0% to 108.1%, with a mean of 99.3%, whereas the interday accuracy was 88.6% to 108.0% with a mean of 101.0%. e intraday accuracy for the Chromsystems QC samples was 88.0% to 103.0% with a mean of 95.9%, and the interday accuracy was 97.2% to 109.0% with a mean of 102.5%. e results confirmed the assay was precise and accurate.
Extract recoveries were evaluated by the traditional postextraction method. e spiked plasma samples were extracted and analyzed along with the nonspiked plasma samples. e extraction recoveries of analytes were higher than 77.6%.

Matrix Effect and Carryover.
e deuterated IS solutions' signal intensities were monitored during the infusion study and indicated no significant matrix-induced ion suppression effect at the relevant retention time for all of the tested sample types ( Figure S2). e mean (SD) of low MN concentration samples following 60 pg/mL samples was 57.5 (5.1) pg/mL, and the mean (SD) of low concentration samples following 1500 pg/mL samples was 60.7 (7.6) pg/ mL, resulting in a carryover of 3.2 pg/mL for MN. e mean (SD) of low NMN concentration samples following 100 pg/ mL samples was 104.0 (12.1) pg/mL, and the mean (SD) of low concentration samples following 7000 pg/mL samples was 89.2 (15.9) pg/mL, resulting in carryover of −14.8 pg/mL for NMN. No statistically significant carryover was observed for either analyte.

Analytical Specificity.
Potential interferences, test concentrations, and percent deviation of analyte concentrations in test samples versus respective baseline samples are shown in Table S2 for endogenous and exogenous substances. Petteys et al. [11] investigated 28 substances, including structurally related compounds, common drugs, or supplements, and isoproterenol, isoetharine, 3, 4-methylenedioxyamphetamine (MDA), and 3, 4-methylenedioxymethpheamphetamine (MDMA), which may interfere with quantification. In our study, 54 compounds were assessed as potential interferences to ensure the robust performance of the proposed MEPS method, including pseudoephedrine, isoproterenol, isoetharine, labetalol, HMMA, MDA, MDMA, epinephrine, and 3-O-methyldopa interfering with MRM detection of metanephrines [11,[27][28][29], except for endogenous compounds, vitamins, and frequently used over-the-counter drugs. Levodopa, isoetharine, isoproterenol, and 4-methoxymethamphetamine (PMMA) were identified at retention times of 1.20 min, 2.70 min, 1.13 min, and 1.93 min in the mass transition chromatograms of MN, respectively, but did not interfere with quantification. HMMA, 3-O-methyldopa, and epinephrine were identified at 2.46 min, 5.09 min, and 5.51 min in the mass transition chromatograms of NMN, respectively, but did not interfere with quantification. MDA and a mixture solution of amine were visible at 2.06 min and the range of 1.6-2.2 min in mass transitions of metanephrines, respectively, and did not interfere with quantification. In contrast, these compounds in the mixture could not be identified due to the lack of Journal of Analytical Methods in Chemistry individual standard reference. Representative chromatograms of some interferences in human plasma are shown in Figure 3. Additionally, labetalol and pseudoephedrine were not visible in this assay because of the selectivity of PGC sorbent. None of the 54 tested compounds interfered with MN and NMN. Below ±15% difference in the concentrations for the QC samples indicated that the quantitation was not affected by the variety of substances.

Stability.
Aliquots of plasma samples were analyzed in triplicates stored at room temperature for 24 h, and 4°C for 5 days. e results indicated that a ≤14.7% change in all the analytes was obtained under these conditions with satisfactory stability according to the acceptance criterion of ±15% change.
Additionally, a ≤4.5% change was observed for the postextract at 4°C in the autosampler for 120 h, allowing reanalysis in routine without reperforming MEPS. (Table 1)

Comparison of Analytical Performance by Available
Methods. A comparison of the analytical performance of the developed method with published assays is summarized in Table 2.
e table shows that the proposed method offers several advantages over previously published methods. First, MEPS was more cost-effective than conventional SPE (MEPS could be reused more than 50 times for plasma samples compared with conventional SPE column or microplate SPE, which could only be used once). Second, MEPS was more ecofriendly compared with the reported sample cleanup. Only  Journal of Analytical Methods in Chemistry relatively small amounts of solvents were needed in MEPS for the washing (<1 mL) and elution (10-50 μL) of the analytes from the sorbent due to ∼2 mg of the solid-packing sorbent. ird, MEPS requires a reduced total analysis time. e novel PGC MEPS sorbent's choice enables using the mobile phase as the elution solvent compatible with HILIC mode. e amount of sorbent was small; sorbent conditioning, evaporation, and reconstitution were unnecessary in MEPS as they were in SPE.

Plasma Metanephrines in Clinical Specimens.
In this study, the levels of plasma metanephrines from 161 suspected patients in a seated position were measured using the newly developed method. In 23 patients (7 males; 16 females: 14-64 yrs), the diagnosis of PPGL was based on histological confirmation of surgically resected tumors. Diagnosis of inoperable malignant PPGL in 1 patient (females; age 48 yr) was observed based on the evidence of metastatic disease by imaging studies. Other 137 patients (73 males; 64 females: 14-87 yrs) included primary hypertension, secondary hypertension caused by other factors, and nonfunctioning adrenal tumors diagnosed using clinical or pathological findings that excluded PPGL. Figure 4     Although the plasma test's diagnostic accuracy may be higher than for the urinary test, the difference is relatively small. Yet, no single test can provide an absolute diagnosis of PPGL because a single abnormal metabolite is not a specific index based on the normal concentrations of other metabolites, and the measurement of multiple biochemical analytes can be used for probabilistic elimination for the diagnosis of pheochromocytoma. erefore, the diagnostic performance of plasma metanephrines was also compared with other biochemical tests in a group of 104 patients, including 21 patients with PPGL, which were determined in our laboratory [26,32] (Table 3). Urinary free catecholamines and VMA showed a comparable lower sensitivity (71.4%). Combining plasma NMN and MN results demonstrated a comparable diagnostic sensitivity (90.5% versus 90.5%) with urinary free metanephrines but a lower diagnostic specificity (88.0% versus 98.8%). Ideally, samples collected from supine patients compared with seated patients would provide the optimum diagnostic sensitivity and specificity for plasma metanephrines. [3]. However, the inappropriate sampling position is sometimes unavoidable in the clinic, and repeated blood samples tested from a supine position offer a cost-effective approach. e false-positive

Conclusion
e present study described a simple, miniaturized, and costeffective analytical method for determining plasma metanephrines using the MEPS sample preparation technique. To the best of our knowledge, a PGC sorbent was used for the first time as the stationary phase for sample preparation of metanephrines. e choice of the novel PGC MEPS sorbent, suitable for the extraction of hydrophilic compounds, enabled us to use the mobile phase as the elution solvent, which was fully compatible with HILIC mode. erefore, the time needed for the sample processing was significantly reduced without the evaporation and reconstitution steps. e analysis cost was considerably reduced due to the reuse of MEPS cartridges for about 50 extractions, decreased sample volume, solvent, and waste disposal. e reliability and applicability were validated using clinical plasma specimens. is new method may be potentially used as a clinical tool for routine laboratory analysis, where simplicity, rapidity, and small sample volumes are required.  Journal of Analytical Methods in Chemistry 9 e authors thank Dr. Huang Zhang from the Clinical Epidemiology Research Center in thePeking University ird Hospital who provided the help of lack-of-fit statistic. Table S1: MRM transitions and respective fragmentation parameters for tandem mass spectrometric analysis of metanephrines and their respective deuterated internal standards. Table S2: deviations from baseline concentrations of plasma metanephrines for exogenous and endogenous potential interferences. Figure S1: schematic diagram of operation course of MEPS. Figure S2: postcolumn infusion experiment. (Supplementary Materials)