Sorbent-Based Microextraction Combined with GC-MS: A Valuable Tool in Bioanalysis

Abstract

Sample preparation is broadly recognized as the most critical, time-consuming, and error-prone step of a bioanalytical workflow. Over the years, the development of pretreatment methods aimed at the isolation and preconcentration of the target analytes from sample matrices has been an ongoing effort. Recent innovations have aimed at miniaturizing sample preparation to streamline laboratory processes and enhance analytical performance. Sorbent-based microextraction techniques, including solid-phase microextraction, microextraction by packed sorbent, bar adsorptive microextraction, capsule phase microextraction, etc., have recently gained attention as effective sample preparation tools prior to gas chromatography-mass spectrometric analysis. This article provides an overview of the bioanalytical GC-MS applications of sorbent-based techniques published in the last decade (2014–2024) that enable the efficient and sensitive determination of various compounds in biological samples.

1. Introduction

Bioanalysis involves the quantification of exogenous or endogenous substances, such as drugs, metabolites, macromolecules, and environmental pollutants in biological samples (i.e., serum, plasma, whole blood, urine, saliva, exhaled breath, and tissues). This discipline is essential for various applications, including disease diagnosis and treatment, bioequivalence studies, clinical and forensic toxicology, drug development, and therapeutic drug monitoring. Biological matrices are typically complex, containing diverse components, such as salts, proteins, peptides, lipids, and organic acids, necessitating advanced and efficient analytical techniques [1,2].

Despite significant advancements and evolution in the bioanalytical field over the years, chromatography continues to be the foundation of all analytical techniques. Among others, gas chromatography combined with mass spectrometry (GC-MS) is a distinctive analytical technique widely used in bioanalysis, offering highly selective and sensitive determinations [3].

Sample preparation is a crucial and often time-intensive step of bioanalytical workflows. It has a substantial impact on the sensitivity, precision, accuracy, and recovery of the subsequent analysis. Following the introduction of the 12 principles of Green Analytical Chemistry (GAC), there has been a growing movement to replace traditional sample preparation methods with microextraction techniques to minimize sample and solvent use [4]. The solid-phase extraction (SPE) technique has played a crucial role in advancing and developing more efficient, eco-friendly sample preparation protocols. Modern approaches require less time and effort compared to traditional SPE, making them more effective [5]. In this context, miniaturized techniques, such as solid-phase microextraction (SPME), microextraction by packed sorbent (MEPS), magnetic SPE (MSPE), pipette-tip solid-phase microextraction (PT-SPE), solid-phase dynamic extraction (SPDE), bar adsorptive microextraction (BAμE), and dispersive micro solid-phase extraction (DμSPE), have gained popularity [5,6,7]. These techniques are well known as sorbent-based microextraction techniques, and they have been extensively used instead of traditional techniques. These procedures provide numerous advantages, such as minimizing the solvents’ consumption, adsorbents, and sample volumes, reducing processing steps, and reducing waste generation. As a result, these approaches pave the way for the development of cost-effective, highly efficient, user-friendly, and eco-friendly extraction and purification processes [8].

The sorbent is considered to be the “heart” of a sorbent-based microextraction procedure. In this context, various materials have been synthesized and utilized in these procedures including molecularly imprinted polymers (MIPs) or membranes (MIMs), carbon-based compounds (CBCs), metal-organic frameworks (MOFs), covalent organic frameworks (COFs), restricted access materials (RAMs), and liquid-supported particles (i.e., ionic liquids and deep eutectic solvents (DES) embedded onto magnetic nanoparticles (MNPs)) [9,10,11,12]. A comprehensive review article on modern materials that are utilized in sorbent-based microextraction techniques has been recently published by F.N. Castaneda et al. [6].

Several review articles have been published in the literature covering topics such as green approaches [7], SPME [1], and various microextraction techniques [8]. However, sorbent-based microextraction technologies in bioanalysis have received relatively little attention. The number of published bioanalytical applications using sorbent-based microextraction prior to GC-MS analysis is shown in Figure 1. The present review summarizes the latest GC-MS applications using sorbent-based microextraction technologies covering a timeframe of the last decade. The latest trends and innovative strategies are thoroughly discussed.

2. Solid-Phase Microextraction

Solid-phase microextraction (SPME) is a fast, solvent-free sample preparation method that combines sampling, isolation, and enrichment in a single step [13]. This technique uses a small amount of an extraction phase, either immobilized or chemically bonded to a solid support, to “capture” analytes, whether volatile (via headspace) or non-volatile (via direct immersion). The process relies on the partitioning equilibrium of analytes between the sample matrix and the extraction phase. SPME is typically utilized prior to separation techniques (GC, LC, CE), but can also be directly coupled with mass spectrometry (MS), eliminating the need for chromatographic separation [14]. This technique offers an efficient, eco-friendly, cost-effective, automated, and continuously evolving approach to the extraction process.

SPME extraction using fiber can be carried out in three main modes: (i) direct immersion (DI), (ii) headspace (HS) extraction above the sample, and (iii) a hollow-fiber membrane-protected approach (HF). Additionally, SPME integrates effortlessly with identification techniques such as GC-MS and LC-MS. Automated SPME injection systems combined with GC or GC-MS are the most widely used setups. When coupled with GC, SPME enables solvent-free sample preparation, particularly effective for determining volatile or semi-volatile organic compounds in complex sample matrices [15].

An interesting green SPME application has been reported by I. Pachero-Fernandez et al. for the determination of polycyclic aromatic hydrocarbons (PAHs) in various matrixes, including environmental water, urine, and brewed coffee [16]. The SPME fiber coating was prepared by CIM-80(Al) MOF material using an in situ growth method. A comprehensive review on the usage of MOFs as sorbents in SPME has been recently published [17]. A variety of techniques (i.e., X-ray powder diffraction (PXRD), scanning electron microscopy (SEM), and thermogravimetric analysis (TGA)) have been used for the surface characterization of the SPME fiber. The manufactured SPME was tested both in direct immersion and headspace sampling modes. The separation of the PAHs was accomplished by using a FactorFour™ VF-5 ms capillary column, and their detection was performed in selected ion monitoring (SIM) mode. After optimization of the parameters of both sampling approaches, the optimum extraction time was 60 min, while a sample incubation temperature of 75 °C was selected for the HS-SPME extractions. A schematic representation of the proposed approach is shown in Figure 2. The authors also compared the performance of the MOF-based SPME fiber with the commercial PDMS one. Similar LOQs were obtained using both fibers, except for acenaphthylene and acenaphthene with the PDMS fiber, which were slightly lower (2.0 ng/L) than those obtained with the MOF phase (2.5 ng/L).

Five HS-SPME applications have been reported for the determination of urinary volatile organic compounds (VOCs) [18], amphetamine analogs [19,20], cyanide and thiocyanate [21], and hexanal and heptanal [22]. Ten years ago, R. Cozzolino et al. developed an interesting HS-SPME method for the determination of VOCs in autistic children [18]. The purpose of this study was to identify the VOCs as potential biomarkers for autism spectrum disorder. After screening four different commercially available SPME sorbents, the divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) fiber was selected as optimum. The urine samples were thermostated at 40 °C and analyzed under both acid (pH 1–2) and alkaline medium (pH 12–14) in order to profile a range of urinary components with different physicochemical properties. Higher temperatures (i.e., 60 °C) led to fiber degradation. Metabolomic information patterns were obtained using GC-MS in scan mode at 30–300 m/z. According to the authors, 2-methylpyrazine, 2,3-dimethylpyrazine, and isoxazole existed in higher levels in the urine of autistic children compared to the controls. On the other hand, autistic children exhibited lower concentrations of methoxy-phenyl-oxime, 3-ethyl-pyridine, acetophenone, and 2-heptanone compared to the control group. In 2016, a simple glass device was constructed for the extraction of cyanide and thiocyanide ions from environmental and biological fluids [21]. The system consisted of a glass tube equipped with two valves and a movable glass slide positioned at its center (Figure 3). It incorporates an acceptor phase comprising a polyurethane foam treated with a Hg(II) dithizonate complex, fixed in a cylindrical tube. The principle of the method is based on the interaction of the acceptor phase with the headspace, followed by measuring the absorbance of mercury (II) dithizonate recovered from the polyurethane foam sorbent. The complex was formed in acidic conditions using 1.5 M H₂SO₄. The method was applied for the determination of both ions in saliva samples obtained from smoker and non-smoker volunteers. One of the disadvantages of the method is the utilization of the highly toxic mercury.

A polyamide/graphene oxide/polypyrrole nanofiber was synthesized using electrospinning and utilized for the HS-SPME of methamphetamine in urine samples [20]. The sorbent was characterized by Fourier transform infrared spectroscopy (FT-IR) and field emission-scanning electron microscopy (FESEM). The experimental parameters were optimized using a rotatable central composite design. It was concluded that the sample pH of 11.5 containing 750 mg NaCl was sufficient to obtain an LOD of 0.9 μg/L. After sampling at 40 °C, the analytes were thermally desorpted in the GC inlet at 240 °C for 3 min. The fiber-to-fiber reproducibility was found to be less than 12.3%, while the accuracy of the method ranged between 5.8 and 13.5% expressed as relative error. According to the authors, the manufactured nanofibers offered a high surface-to-volume ratio and sufficient mechanical, thermal, and chemical stability. Amphetamine-like stimulants have been determined in human urine using GC-MS after microextraction [19]. The analytes were extracted using acid-oxidized multiwalled carbon nanotubes coated onto a stainless-steel wire. The attachment of sorbent on the wire was performed using a simple physical adhesion approach. The SPME was thermally conditioned at 250 °C, while the urine sample was incubated at 80 °C for 10–60 min. An HP-5MS capillary column was used for the separation of the analytes using a thermal gradient program up to 175 °C. The quantitation of the compounds was carried out in selected ion monitoring (SIM) mode with one and two quantification and confirmation ions, respectively. The linearity of the method ranged between 1 and 1000 μg/L with adequate sensitivity (LODs: 0.2–1.3 μg/L) for this type of analysis. The method was applied to the analysis of urine samples obtained from young persons who were suspected of the consumption of amphetamine-type stimulant drugs.

The research group of Alberto Chisvert developed a magnetic HS adsorptive microextraction for lung cancer diagnosis [22]. Two endogenous aldehydes, namely hexanal and heptanal, were identified as lung cancer biomarkers in saliva. The method is based on a modification of magnetic headspace adsorptive microextraction followed by GC-MS analysis. A neodymium magnet is employed to hold the magnetic sorbent (containing magnetic nanoparticles) in the headspace vial. The saliva samples were collected using Salivette swabs and centrifuged at 6000 rpm to remove phlegm and food leftovers. The acetonitrile used for the analytes desorption resulted in an extraction recovery of 61–97% following a simplex-centroid mixture design. Enrichment factors of 12–15 were obtained with a LOQ of 0.75 and 0.85 ng/mL. The authors found that the analytes were below LOD in healthy volunteers, while higher values were recorded in lung cancer patients. When compared to other approaches in terms of green characteristics, the method demonstrated comparable metrics using AGREEprep software. An overview of SPME applications is tabulated in

Table 1. SPME applications combined with GC-MS for the analysis of biological samples.

Analyte	Sample	SPME Sorbent/ Extraction Conditions	Sample Volume (mL)	LOD (ng/mL)	% RR 1	Reference
PAHs	Environmental water, urine, brewed coffee	CIM-80(Al) MOF/ET 2: 60 min at 75 °C (for HS); ET: 60 min at 50 °C (for direct immersion)	10 mL (for HS) 19 mL (for direct immersion)	0.0003–0.0015	80.1–107 (for urine) 74.6–107 (for coffee)	[16]
VOCs	Human urine	DVB/CAR/PDMS/ET: 30 min at 40 °C (HS)	4	–	–	[18]
Amphetamine-type stimulants	Human urine	MWCNTs-COOH/ET: 20 min at 80 °C (HS)	5	0.2–1.3	88–107	[19]
Methamphetamine	Human urine	Polyamide/graphene oxide/polypyrrole/ET: 20 min at 40 °C (HS)	5	0.9	89.5–95.3	[20]
Cyanide, Thiocyanate	Environmental and biological samples	PUF treated with Hg(II) dithizonate complex/ET: 5 min at 60 °C	5	0.34 (μmol/L)	98.7–101 (wastewater) 99.9–100.5 (saliva)	[21]
Hexanal, Heptanal	Saliva	CoFe2O4 magnetic nanoparticles embedded into a reversed-phase polymer/ET: 40 min at 30 °C	1	0.22–0.26	75–135 (Hexanal) 82–133 (Heptanal)	[22]

¹ RR: Relative recovery. ² ET: Extraction time.

.

3. Microextraction by Packed Sorbent

Microextraction by packed sorbent (MEPS) is a miniaturized sample preparation technique first introduced by Abdel-Rahim in 2004 [23,24]. Its principles are based on solid-phase extraction (SPE) [25]. MEPS setup is composed of a liquid handling gas-tight syringe, as a plug or as a cartridge between the removable needle and the barrel in which the solid sorbent bed is packed in a few milligrams (1–4 mg), as illustrated in Figure 4 [26]. The principle of operation of MEPS briefly involves the conditioning of the sorbent, sample loading by consecutive draw-eject cycles to ensure the retention of the analytes on the sorbent, a washing step for interferent matrix component removal, a drying step (optional), and the elution of the analytes using an organic solvent. The basic parameters that affect the MEPS performance are the type and amount of the packed sorbent, the conditioning of the sorbent, sample pretreatment and loading, sorbent washing, analyte elution, and sorbent reconditioning. These parameters need to be extensively studied and optimized during the development of MEPS bioanalytical protocols to achieve high sensitivity and good recoveries of the analytes. Among various microextraction-based techniques, MEPS has recently emerged as an advanced method in bioanalysis for the preconcentration of analytes from biological fluids, offering numerous advantages.

Commercially available MEPS sorbents include ethyl (C₂), octyl (C₈), and octadecyl (C₁₈) silica-based materials, a C₈/strong cation exchanger (SCX), pure silica, polystyrene polymer, mixed-mode C₈/SCX, molecular imprinted polymers, and organic monolithic materials have also been successfully used [27]. An overview of MEPS applications is summarized in

Table 2. MEPS applications combined with GC-MS for the analysis of biological samples.

Analyte	Sample	MEPS Sorbent	Extraction Time (min)	Sorbent Reusability (Extraction Cycles)	Sample Volume (μL)	Linear Range	LOD (ng/mL)	% RR 1	Reference
Antidepressants	Oral fluid, Plasma	M1 (4 mg; 80% C8—20% SCX)	–	40–50 (oral fluids) 20–30 (plasma)	150	10–500 ng/mL	2–50	12–93%, 28–101%	[28]
Opiates	Urine	M1 (4 mg; 80% C8—20% SCX)	–	–	250	1–1000 ng/mL	1–10	17–107.8%	[29]
Opiates	Whole blood	M1 (4 mg; 80% C8—20% SCX)	–	–	250	5–1000 ng/mL	5	6.06–22.5%	[30]
Amphetamine-type stimulants	Urine	C18	<3	100	200	25–1000 ng/mL	–	19–71%	[31]
Ketamine— Norketamine	Urine, Plasma	M1 (4 mg; 80% C8—20% SCX)	–	–	250	1–250 ng/mL 10–500 ng/mL	5 ng/mL	63–101%	[32]
Ketamine— Norketamine	Hair	M1 (4 mg; 80% C8—20% SCX)	–	–	150	0.05–10 ng/mg	0.1 ng/mg (KET) 0.05 ng/mg (NK)	32–61%	[33]
Cannabinoids	Urine	M1 (4 mg; 80% C8—20% SCX)	–	100	250	1–400 ng/mL	1–10 ng/mL	26–85%	[34]
Tetrahydrocannabinol and metabolites	Plasma	M1 (4 mg; 80% C8—20% SCX)	3	>200	250	0.1–30 ng/mL	0.2 ng/mL	53–78%	[35]
Polyamines	Saliva	C18	–	110	100	0–7.5 mg/L	1.84–34 ng/mL	89–130%	[36]
Polyamines	Urine	Silica-C18	–	–	500	0–400 ng mL	0.18–2.7 ng/mL	90–113%	[37]
PAHs	Urine	C18	9	80	300	1.5–375 ng/mL	0.5–19.4 ng/mL	88–110%	[38]
PAHs	Saliva	C18	21	–	1500	10–842 ng/mL	4.6–79 pg/mL	78–123%	[39]
Nitro explosives	Blood, Urine	Silica-C18	10	60	50	1–250 ng/mL	0.014–0.828 ng/mL	>88%	[40]
Polychlorinated biphenyls	Bovine serum	C18	5	–	100	2–2000 ng/mL	0.06–0.53 ng/mL	60–91.4%	[41]
Tryptophan metabolites	Human serum, cerebrospinal fluid	C18	–	–	40	0.4–10 μM (serum) 0.4–7 μM (cerebrospinal fluid)	0.2–0.4 μM	40–60%, 40–80%	[42]
Aromatic microbial metabolites	Aqueous solutions	C18	21	–	50	–	2–3 μM	21–102%	[43]
Pyrethroid metabolites	Urine	C18	–	70	500	0.05–25 ng/mL	–	92–124%	[44]
Organophosphorus pesticides	Whole blood	C18	–	–	150	0.5–50 μg/mL	0.5–2.5 μg/mL	61–77%	[45]

¹ RR: Relative recovery.

.

MEPS is considered an outstanding approach for sample pretreatment, and it is particularly advantageous for small sample volumes that can be as small as 10 μL [46]. MEPS syringes offer significant advantages in terms of reusability, as they can be reconditioned and reused multiple times (10 to 300 uses), unlike conventional single-use SPE cartridges. Additionally, MEPS supports full automation due to its compatibility with autosamplers, enabling the direct online coupling of sample preparation techniques with analytical instruments such as GC-MS and LC-MS/MS [25]. Regarding the MEPS as a sample preparation approach in drug analysis, more information can be found in a critical review article by G. Alves et al. [27].

Gallardo and her research group developed several analytical methodologies by employing MEPS as the sample pretreatment technique for the determination of drug substances and related metabolites in both conventional [28,29,30,31,34,35,45] and non-common biological samples [33]. A recent application published in 2024 involved the development of two separate protocols for the extraction of five antidepressant drugs and one metabolite (namely fluoxetine, venlafaxine, o-desmethylvenlafaxine, citalopram, sertraline, and paroxetine) from oral fluid and plasma samples by a mixed-mode C₈/strong cation exchanger sorbent cartridge and their quantification by GC-MS/MS [28]. The main parameters affecting the extraction procedure, such as the number of draw-eject sample cycles, washing, and elution solvents, were cautiously studied and optimized. Although the whole sample preparation procedure was described as time-consuming and labor-intensive, adequate LLOQ values for all analytes were obtained, and the method could successfully be applied for the quantification of the antidepressants in the context of therapeutic monitoring purposes.

In a similar manner, the same research group integrated the MEPS procedure with gas chromatography-tandem mass spectrometry for the determination of several opiates in the urine [29] and whole blood [30] samples of opiate users. In both studies, a C₈/SCX MEPS sorbent in combination with the use of acid solvents facilitated the sufficient retention of the analytes in the sorbent bed. Taking into consideration their pKa values, the acidic conditions enhanced the positive charge of the determined compounds, while the alkaline elution solvent enabled their desorption from the sorbent. In the first study [30], the authors highlighted low recoveries as the primary disadvantage of the method. This was attributed to the high amounts of present interfering compounds in the blood samples and the low amount of the MEPS sorbent available from the manufacturer, which may impede the interaction of the analytes with the sorbent. Despite this, satisfactory precision and accuracy were observed by using only 250 μL of sample. In the second study [29], the LOQ was 1 ng/mL for all analytes, except one, and LODs were in the range of 1–10 ng/mL, compared to similar published methods, making the developed method suitable for forensic routine analysis.

Two another interesting studies from Gallardo’s research group involved the development and validation of analytical methods for the identification and quantification of ketamine and its main metabolite norketamine in urine and plasma samples [32] and in hair [33] using MEPS and GC-MS/MS analysis. The first study outlined a simple and rapid protocol that requires only a small sample volume (250 μL) and eliminates the need for a derivatization step before analysis. This method achieves low detection limits (5 ng/mL) and high recovery rates, ranging from 73% to 101% in urine and 63% to 89% in plasma, for both compounds in biological specimens. In contrast to the first study, in the second, hair samples were used for the determination of the analytes [33]. Hair is considered a non-conventional biological sample, which requires a previous wash and decontamination step before the extraction procedure. Although this step may lead to a laborious and time-consuming sample preparation process, the hair matrix provides a historical record of exposure to drugs and other substances, and thus it can be used for long-term drug detection [47]. Moreover, it is less prone to external conditions (high temperature, rain, drought, etc.), is not subject to rapid metabolism or excretion, unlike urine and plasma, and allows analyte detection within a wider time window (from weeks to years, depending on the hair length). These facts render hair samples extremely valuable when it comes to forensic investigations and drug testing, such as the ketamine case studies. Within this context, in the second study, lower LLOQs (0.05 ng/mg) and LODs (0.01 ng/mg and 0.05 ng/mg for ketamine and norketamine, respectively) were comfortably achieved, despite the lower obtained recoveries (<50%) [33]. Hence, the second GC-MS/MS method seems to be more sensitive than the first one.

Another category of analytes, amphetamine-type stimulants, attracted the interest of Gallardo et al. [31]. Amphetamines are chemically synthesized psychoactive compounds, like several neurotransmitters, that are widely consumed by drug abusers. Consequently, their determination in biological fluids remains a crucial issue in toxicological analysis [48]. Gallardo and her colleagues developed and fully validated a MEPS-GC-MS method for amphetamine determination in urine samples [31]. By using only 200 μL of a sample, satisfactory recoveries and adequate LODs and LOQs were observed, with a larger number of compounds being analyzed compared to other reported methods [49,50,51,52]. The overall bioanalytical protocol was revealed to be simple, rapid (extraction process < 3 min), sensitive, selective, and accurate.

In the same context, cannabis and related products are considered some of the most available (or diffused) and abused illicit drugs worldwide [53]. The determination of some cannabinoids in biological samples can be challenging for various reasons. In particular, in addition to the matrix complexity of body fluids, cannabinoids are extensively metabolized and may disappear quickly from the bloodstream, are highly lipophilic, tend to accumulate in fatty tissues, and remain in low concentrations in aqueous biological fluids. Furthermore, they have a short detection window (for a few hours and up to a day) and individual variability (body fat percentage, frequency of use, etc.) that affects cannabinoid concentration, making it harder to standardize detection thresholds [54]. Despite the aforementioned limitations, several research groups have developed methods for their detection. Gallardo et al. established two bioanalytical protocols for the trace-level determination of cannabinoids and their main metabolites in urine [34] and human plasma samples [35] using MEPS prior to GC-MS and GC-MS/MS, respectively. In both methods, a small sample volume of 250 μL was used. In the first study [34], the method was considered suitable for cannabinoid determination, noting that, for all cannabinoids, lower LLOQs were achieved (1–10 ng/mL) than the recommended cut-off concentrations (50 ng/mL) for laboratory screen tests. In the second study [36], although the sample needed to be diluted 20 times, the method’s sensitivity was not affected, and very low LLOQ values for all analytes were obtained (0.1 ng/mL), taking into consideration the limitations of the cannabinoid’s metabolic profile. The sensitivity and extraction recoveries of the method were also enhanced by employing an increased number of 26 sample strokes, in order to ensure sufficient sorption of the analytes on the sorbent bed.

Finally, Gallardo’s research group designed and validated a method for the quantification of six organophosphorus pesticides in whole blood samples by MEPS and GC-MS/MS analysis [45]. The method was found to be selective, as no signals deriving from other compounds that could be present in authentic samples were detected at the corresponding retention times and transitions of the studied compounds. The recoveries obtained were comparable to those of other methods that used larger sample volumes [45,55,56,57]. In contrast, the described method required only 150 μL of the sample for MEPS. The authors pointed out that a higher number of sample loading cycles by using an automated MEPS system could yield better responses and even higher recoveries. Regarding the LLOQs, although higher values were achieved compared to other published studies [58,59], the method was deemed appropriate for quantitative analysis in cases of acute poisoning where elevated concentrations of the studied compounds are typically observed.

In a similar vein, Pavón and his research group designed and optimized two analytical methodologies for the determination of biologically active polyamines and related compounds in urine [37] and saliva samples [36]. The analytes were derivatized in situ with ethyl chloroformate using both methods prior to their isolation from the matrix using microextraction by packed sorbent and their GC-MS detection. The derivatization step was deemed to be imperative due to the high polarity and poor volatility of polyamine molecules. In the method applied to saliva samples [37], it is notable that the MEPS cartridge is reusable for up to 110 extraction cycles under optimal conditions, with impressive recovery performances for all screened analytes ranging from 89% to 109%. In the method applied to urine [38], it is significant to highlight the method’s reliability, expressed as %RSD, which was assessed in terms of repeatability and reproducibility. In all cases, the %RSD values did not exceed 15%, which was considered highly satisfactory as they reflect variations across all procedures, including derivatization, extraction, and separation. In addition, LODs and LLOQs were evaluated and found to be in the low range of μg/L in both methods. All of these mentioned features render each of the two methods appropriate for monitoring the studied polyamines in healthy individuals and cancer-diagnosed patients, given that bioactive polyamines have been classified as viable cancer biomarkers [60].

Pavón et al. also worked on the determination of polycyclic aromatic hydrocarbons (PAHs) in human urine by using a fully automated microextraction using a packed sorbent coupled to GC-MS analytical instrumentation [38]. The method involved the easy pretreatment procedures of urine samples, requiring a simple hydrolysis step and no derivatization prior to analysis. The hydrolysis step was included to convert PAHs into OH-PAHs, as the latter are considered reliable biomarkers for assessing PAH exposure levels and are more harmful to human health than their parent compounds. The automation of the method necessitated minimal sample manipulation, thus reducing errors associated with sample preparation and saving time, compared to the more sensitive LC-MS/MS previously reported methods that involve a manual sample pretreatment [61,62,63]. Moreover, the developed method provides the advantage of shorter durations for both the extraction process and chromatographic separation, totaling 17.5 min, compared to a previously published method [64]. The automated MEPS procedure streamlines cartridge washing, sorbent reconditioning, and subsequent sample loading during the preceding chromatographic separation run, enabling high sample throughput of up to three samples per hour. The proposed methodology was applied to urine samples from both smokers and non-smokers, and the results aligned with those of the previously described methods.

Moving forward, Cordero et al. developed a sensitive and automatic method for the determination of unmetabolized PAHs in saliva before their biotransformation in the body, based on MEPS and GC-MS analyses [39]. The hydrolysis step could be omitted, and saliva samples were easily collected and handled during the pretreatment procedure. The main advantages highlighted by the authors include the method’s full automation, where the entire analytical process (extraction, separation, and detection) occurs online, minimizing associated errors. Additionally, the MEPS workflow is notably short, requiring only 21 min per sample. Additionally, the use of more common analytical instrumentation, such as a single quadrupole mass spectrometer, makes the developed method more accessible compared to other reported methods [65,66] that employ advanced analytical equipment, thus making it suitable for most laboratories.

Other published works regarding biological samples that involve MEPS as the sample preparation approach and GC-MS instrumentation for analysis are the determination of pyrethroid metabolites in human urine samples [44], the determination of tryptophan metabolites in serum and cerebrospinal fluid samples [42], the microextraction of aromatic microbial metabolites [43], the quantification of nitro explosives in environmental aqueous and fluidic biological samples [40], and the screening of polychlorinated biphenyls residues in bovine serum samples [41].

Synthetic pyrethroids are some of the most used insecticides and are being quickly metabolized in humans. They are predominantly excreted as metabolites in urine. These metabolites can act as biomarkers of both occupational and non-occupational exposure. Hence, human biomonitoring of them is essential for public health. One study [44] describes the development of a novel, fast, and eco-friendly analytical method using offline microextraction using a packed sorbent and solid support derivatization followed by GC-MS detection for the qualitative and quantitative determination of five pyrethroid metabolites in human urine samples. The MEPS procedure was conducted offline using a manually operated syringe (eVol), with several parameters influencing the thoroughly studied and optimized process. The analytes were extracted from enzymatically hydrolyzed urine using a C18 sorbent, followed by simultaneous derivatization and elution in a single step. This approach eliminates the vaporization step, making the method even more straightforward. Overall, the proposed method is rapid, exhibits adequate LLOQs for all analytes, and utilizes small sample and organic solvent volumes. As a result, it has a low environmental impact, and it can be effectively applied in biomonitoring studies.

In the case of tryptophan metabolites [42], adequate LODs and LLOQs offer satisfactory recoveries for all studied metabolites in both matrices, given a low sample volume of only 40 μL. This parameter is very important for cerebrospinal fluid samples, as it requires an invasive collection process and specialized healthcare training to perform it and thus has limited accessibility and sample volumes. One limitation of the proposed method is the requirement of a 30-min derivatization procedure at high temperature (90 °C) followed by a 30-min cooling step before GC-MS analysis. These steps contribute to the method’s time-consuming nature.

Regarding the nitro explosives [40], a simple, rapid, and accurate MEPS procedure was developed for the quantification of 12 nitroaromatic compounds followed by GC-MS detection in environmental water, human urine, and plasma samples. The main factors affecting recovery were thoroughly investigated and optimized. According to the authors, the developed method reduces sample preparation and analysis time compared to previously reported conventional LLE [67] and SPE [68] methods. Furthermore, the MEPS procedure required a lower sample volume of 500 μL for only 10 extraction cycles, reduced organic solvent consumption, minimized matrix effect, and thus could broaden the method’s applicability.

Another interesting research work published by Ding et al. [41] involves the fabrication of a self-made MEPS device by packing C₁₈ particles in a microinjection needle (50 μL) for the clean-up of seven polychlorinated biphenyls from bovine serum followed by GC-MS. A schematic representation of the MEPS device is illustrated in Figure 5. The overall developed method was considered fast, simple, and environmentally friendly, as it required only two sample loading and elution cycles, with both sample and organic solvent consumption in the μL range. The optimized conditions resulted in a time-saving and user-friendly analytical procedure, demonstrating superior figures of merit.

Finally, another noteworthy research work dealt with the microextraction of aromatic microbial metabolites, namely phenyl carboxylic acids, by MEPS prior to the GC-MS analysis of their silyl derivatives [43]. The conditions selected as optimum for adsorption–desorption, derivatization, and GC determination enabled the detection of various compounds that are established as potential disease markers in cases of sepsis and other serious diseases caused by significant microbial loads. The entire sample preparation time did not exceed 21 min, while the developed analytical methodology was robust, demonstrating adequate recovery values ranging from 20% to 65% for hydroxylated acids and to 100% for more nonpolar acids. Additionally, the authors compared their findings with those from an analogous LLE procedure and highlighted the superior performance of the MEPS method for most of the acids. The method was deemed reliable for monitoring the concentrations of the compounds in blood samples from both healthy donors and patients in the early stages of an infectious process.

4. Dispersive Micro Solid-Phase Extraction

Dispersive micro solid-phase extraction (DμSPE) is a downsized version of dispersive solid-phase extraction (D-SPE). The latter was initially introduced by Anastasiades et al. as an efficient sample preparation technique for pesticide analysis in complex matrices [69]. In DμSPE, a small amount of sorbent, typically a few milligrams, is dispersed into a solution (typically by sonication, stirring, vortex, or shaking) containing the analyte to enable fast, simple, and more cost-efficient extraction. This dispersion enables rapid and uniform interaction between the sorbent and analytes. The sorbent is subsequently separated by filtration or centrifugation, followed by analyte elution using an appropriate organic solvent. It is important to note that μSPE differs from SPME, where the sorbent is coated onto a fiber core and directly immersed in the analyte-containing sample solution [70].

Mohebbi et al. combined a DμSPE protocol with deep eutectic solvent (DES) liquid microextraction for the quantitation of amitriptyline, nortriptyline, desipramine, clomipramine, and imipramine in biological fluids (plasma, urine, etc.) [71]. The principle of the method lay in the dispersion of 75 mg of C₁₈ material in 5 mL of the diluted alkaline sample (pH 10), followed by centrifugation and the collection of C₁₈ particles. The analytes were then eluted using DES (choline chloride/4-chlorophenol) and the sedimented phase was injected into GC-MS for analysis. The procedure is schematically illustrated in Figure 6. Among the tested sorbents, C₁₈ provided the best extraction performance of all of the analytes. As the analytes are basic compounds (pK_a = 9.2–10.47), the sample was basified to enhance their extraction capabilities. The achieved LOD values were between 8–15 ng/L in urine and 32–60 ng/L in plasma, respectively.

Two different approaches were reported in 2020 for the quantitation of phenazopyridine [72] and four antiepileptic drugs [73] in human urine using GC-MS. In the first method [72], magnetic graphene oxide nanoparticles modified by poly(thiophene-pyrrole) copolymer were utilized as the sorbent in DμSPE for the extraction of phenazopyridine from human urine. After the addition of 40 mg of sorbent in the sample, the analytes were adsorpted on the material’s surface and then eluted with 1 mL methanol. In the second step, the methanolic portion was utilized as the disperser in dispersive liquid-liquid microextraction (DLLME) for further sample cleanup and analyte preconcentration. The salting-out effect in the DLLME procedure was investigated, and the authors found that the NaCl concentration of 15% w/v was optimum in terms of extraction recovery. The accuracy of the method was adequate, as the relative recoveries were within a range of 85–89%. The determination linear range was investigated in the range of 0.5–250 ng/mL (r² = 0.9988), while a significant preconcentration factor of 600 was achieved. The applicability of the method was studied by analyzing spiked human urine samples. M. Parisa et al. published a solvent-assisted DμSPE protocol for the quantification of four antiepileptic drugs (valproic acid, phenobarbital, levetiracetam, and pregabalin) in urine [73]. In this method, the sorbent was dispersed by quickly injecting a mixture of the adsorbent (magnetic Fe₃O₄@SiO₂) and disperser solvent (methanol) into the aqueous sample, which increased the contact surface area, reduced extraction time, and minimized sorbent usage. A central composite experimental design consisting of 30 runs was followed for the optimization of the main parameters. It was found that the simultaneous decrease in the disperser solvent volume and sorbent amount resulted in an enhancement of the sorbent dispersion process. Under SIM mode, each drug was monitored using one quantifier and six qualifier ions. High determination coefficients r² (>0.995) were recorded with LOQs ranges of 0.2–0.4 ng/mL. Acceptable recoveries (92.9–103.2%) with an RSD < 4.7% were noticed. An analogous magnetic dμSPE protocol coupled with GC-MS was developed for the determination of tramadol and fluoxetine in environmental water and biological samples [74]. Graphene oxide-modified Fe₃O₄ was prepared and utilized as the sorbent. The screening of the most critical parameters was performed using the Plackett-Burnam design in 16 experiments. The desorption time, pH, and desorption solvent volume had a negative effect, and the stirring rate had a positive effect on the response. After optimization, the sample pH, desorption time, stirring rate, and desorption solvent volume were 2.2, 7.8 min, 1600 rpm, and 230 μL, respectively. Compared to other published approaches, the method showed better accuracy, within a range of 96.2–103.9%. The method showed a wide linear range of almost three orders of magnitude, which was superior to the reported approaches. The same analytes were determined in urine samples using a DLLME assisted by magnetic carbon composite DμSPE followed by GC-MS analysis [75]. The samples were initially combined with a novel magnetic and porous adsorbent, composed of MOF-derived NH₂-functionalized Fe₃O₄ nanoparticles. The eluent was directly utilized as the disperser solvent alongside CH₂Cl₂ as the extracting solvent for DLLME. The combination of these techniques offered remarkable enrichment factors and extraction recoveries due to the advanced high-surface-area sorbent. Linear ranges of 0.1–30 ng/mL for tramadol and 0.1–35 ng/mL for methadone were achieved, with LODs of 73 and 32 pg/ mL, respectively. The authors compared the results of their method with those obtained from the standard HPLC/MS-MS approach, which were in good agreement.

A. Jouyban et al. published a polymer-based dispersive solid-phase extraction combined with DES-based DLLME for the determination of four hydroxylated PAHs from urine samples [76]. The authors took advantage of using 175 mg of polystyrene dissolved in 1.5 mL of isopropanol rapidly injected into the sample solution. After centrifugation at 1120× g, the adsorbed analytes were eluted using choline chloride/dichloroacetic acid, and the supernatant was further mixed with choline chloride/3,3-dimethyl butyric acid and dispersed in aqueous solution. The final sedimented phase was injected into GC-MS. Increasing the polystyrene amount, the extraction efficiency was also increased up to 175 mg, while no effects were observed by changing the sample pH in the range of 2–10. Analysis of the urine samples collected from tobacco smokers revealed the presence of 1-hydroxynaphthalene and 2-hydroxy-naphthalene, with concentrations ranging from 29 to 53 and 43 to 84 ng/L, respectively.

5. Bar Adsorptive Microextraction

Bar-adsorptive microextraction (BAμE) was firstly introduced by N. R. Neng et al. in 2010 [77]. Theoretical aspects on BAμE were described in the research work of N.R. Neng et al. [78]. It is a significant sorption-based micro-technique designed to address certain limitations of stir-bar sorptive extraction. This technique enables the extraction of polar and non-polar compounds by utilizing alternative materials as sorbent phases. In its most widely recognized configuration, BAμE employs polypropylene cylindrical bars (15 mm in length and 3 mm in diameter), with the sorbent phase affixed using an adhesive film. A key advantage of the BAμE technique is the possibility of using various sorbent phases that enhance the extraction selectivity for specific analytes. Additionally, it supports the use of environmentally friendly sorbent materials with notable physicochemical properties, ensuring efficient extraction performance.

A high-throughput BAμE and microliquid desorption was proposed by S. M. Ahmad for the determination of ketamine and norketamine in urine samples [79]. Since both analytes present a weak basic character, the urine samples were initially basified at pH 11 and then extracted using NVP-DVB sorbent for 30 min at 1800 rpm. The desorption was carried out through cavitation for 15 min using 100 μL methanol. The authors used a similar apparatus to the one described in their previous study [80]. According to the authors, this apparatus presented remarkable effectiveness as a rapid tool for the simultaneous microextraction of up to 100 samples. Matrix-matched calibration curves were constructed in the range of 5–1000 μg/L with r² > 0.999. No matrix effects were observed, as the recoveries were acceptable, being in the range of −9.1–9.0. One year later, the same group of authors utilized the same microextraction technique to determine six tricyclic antidepressants (namely amitriptyline, mianserin, trimipramine, imipramine, mirtazapine, and dosulepin) in urine using large-volume injection GC-MS [81]. One-variable-at-a-time was used for the optimization of the BAμE procedure. Six different polymer coatings were studied, and it was found that C₁₈ and Strata-cyano materials provided the best selectivity. These polymers differ significantly, as the C₁₈ polymer contains a long aliphatic chain that facilitates strong hydrophobic interactions with the aliphatic chains in the molecular structures of the target drugs. On the other hand, a cyano-based polymer enables dipole-dipole interactions via its cyano group and the electronegative elements (N and/or S) within the chemical structures of the six drugs. For the analyte elution, an ultrasonic irradiation (100 W) was applied to the sorbent coatings for 15–30 min, while methanol was employed as the elution solvent. More recently, C.V.P. Almeida et al., inspired by the previous work of M.N. Oliveira [81], developed a BAμE protocol for the isolation of alkylamine stimulants (1,3-dimethylbutylamine, 1,4-dimethylpentylamine, heptaminol, isometheptene, 1,5-dimethylhexylamine, and tuaminoheptane) from human urine matrixes [82]. The identification of the analytes was carried out in SCAN mode in the range of 70–600 m/z. The analytes were derivatized with N-methyl-N-trimethylsilyl-trifluoroacetamide at 80 °C for 10 min in order to increase their volatility. A temperature gradient from 90 °C to 300 °C at a rate of 20 °C/min was followed to separate the analytes. As the sample pH is a crucial parameter that can influence the extraction recovery of the analytes, this value was studied in the range of 2–11. The stability of the target analytes in the final extract prior to GC–MS analysis was examined up to 144 h after preparation. The methodology proved to be an excellent alternative, as it is cost-effective, user- and eco-friendly, and requires only a small volume of urine sample (1 mL).

6. Other Microextraction Techniques

An interesting stir-bar-supported μSPE was proposed for the quantification of polychlorinated biphenyl congeners in serum samples followed by GC-MS analysis [83]. The proposed configuration involved the placement of a small stir bar inside the porous polypropylene membrane along with the sorbent material. The edges of the membrane were heat-sealed to secure the contents. The authors concluded that the best extraction results were obtained with C₁₈ material. Similarly to [81], sonication greatly affects the desorption process, and therefore a time span of 15 min was adequate for the complete desorption of the analytes. Various solvents with varying polarity indexes (toluene, n-hexane, methanol, and acetonitrile) were tested, and it was found that toluene (at a volume of 250 μL) provided the best extraction recoveries for all compounds. Enrichment factors in the range of 11.7–18.4 were obtained, but the absolute recoveries varied between 59.0 and 86.7% as μSPE is not an exhaustive technique like SPE. The inter-day (3.2–9.1%) and intra-day (3.1–7.2%) RSDs were within acceptable ranges. An analogous ultrasound-assisted μSPE was proposed by the research group of S. Mohammadiazar for the simultaneous determination of methadone and tramadol in serum samples [84]. Various modern sorbents were tested, including carbon nanotubes, oxidized carbon nanotubes, and TiO₂ nanoparticles. Among these materials, oxidized carbon nanotubes provided the best performance due to the strong affinity between the drugs and certain adsorbents. Using a non-polar HP-5MS capillary column, the analytes were chromatographed in less than 10 min and quantified under SIM mode. One milligram of sorbent was sufficient for the extraction of the drugs at a sample volume of 1200 μL and a pH value of 9.0. As expected, the volume of the desorption solvent (N,N-dimethylformamide) decreased, and the preconcentration factor increased; thus, the authors selected the value of 50 μL for this purpose. The method also indicated good repeatability and linearity over a wide concentration range. Low LODs were achieved (0.5 and 0.8 ng/mL), while the relative recoveries were within the range of 96.4–105.0%.

Two sorptive-based extraction protocols were proposed for the determination of synthetic opioids in oral fluid samples [85] and basic drugs in blood [86]. The first approach was based on the utilization of a fabric-phase sorptive extraction (FPSE) for the isolation of fentanyl analogs from saliva prior to GC-MS analysis. The combination of London dispersion forces, hydrogen bonding capability, and dipole-dipole interactions within the sol-gel CW 20M sorbent enabled strong intermolecular interactions between the FPSE membrane and the analytes. More information about the FPSE can be found elsewhere [87]. Cotton swabs were used for sample collection, followed by their extraction using a mixed mode CW 20 M/sol-gel PTHF/sol-gel PDMDPS fabric membrane with dimensions of 1 × 1 cm. Each material was characterized by FT-IR. The target compounds exhibited intermediate-to-low polarity, and ethyl acetate was better suited for releasing the compounds retained in the polar CW20M coating into a less polar solvent. NemrodW^® software was utilized to create an asymmetric screening design for the examination of the most influential factors of the FPSE procedure. The method was validated according to FDA guidelines. A representative total ion chromatogram is illustrated in Figure 7.

The LODs ranged from 1 to 15 ng/mL, except for furanyl fentanyl (25 ng/mL), and the intra-day and inter-day precision did not exceed 8.2% and 8.6% RSD, respectively. A total of 15 samples from volunteers and 8 samples from patients with drug addiction undergoing detoxification treatment were analyzed with the proposed method. In the latter method, the authors took advantage of using cellulose paper material for the extraction of basic compounds from blood samples [86]. Five antidepressants, four benzodiazepines, and ketamine were determined by GC-MS using an SH-Txi-5Sil MS capillary column. A filter paper of 1.5 × 1.5 inches was found to be best for the extraction of the compounds. The most critical parameters were screened using the Plackett-Burman design. The authors concluded that the sample pH, the stirring rate, and the sample volume were the most influential parameters that were then optimized using a central composite design. The precision (intra-day and inter-day) was less than 10%, while the enrichment factors were within the range of 15–20. According to the text, only 2 mL of methanol was used for the elution of the drugs. The method is simple, without the need for additional steps, such as centrifugation, vortex, etc., The method was linear, in the range of 0.05–2 μg/mL for all drugs, with an r² > 0.995.

7. Conclusions

Sorbent-based microextraction techniques are widely utilized in research laboratories worldwide. They continue to revolutionize sample preparation by offering an effective, environmentally friendly, and highly selective approach to analytical chemistry. Their growing popularity stems from years of innovation in developing new strategies and refining established sampling techniques. Key attributes, such as minimal solvent use, high extraction efficiency, and potential for automation, make these techniques a valuable alternative to traditional extraction methods. Advanced materials with exceptional physicochemical properties, such as MOFs, graphene-based magnetic sorbents, and MIPs, have been used in sorbent-based microextraction techniques. Their combination with GC-MS resulted in a powerful bioanalytical tool widely used in both academic research and routine analysis.