Investigation of Metoprolol Concentrations in Plasma Using Automated Sample Preparation and LC-MS/MS Detection

Abstract

Metoprolol (MTP), a selective beta-1 adrenergic blocker, is commonly administered in the form of succinate or tartrate salts, depending on the pharmaceutical formulation. It is typically prescribed in oral forms as either immediate-release or extended-release tablets. This study describes a chromatographic method using automated sample clean-up and elution via a reversed-phase mechanism. A TurboFlow approach was applied with a Cyclone P column, and the elution was performed isocratically using a mobile phase of water and acetonitrile (0.1% v/v formic acid) within 4.5 min. Quantification of MTP was achieved using triple quadrupole mass spectrometry, with the transition m/z 268.1 → m/z 130.96 for metoprolol, while bisoprolol fumarate, the internal standard, was detected at m/z 326.3 → m/z 116.2. The method was validated according to bioequivalence guidelines. Selectivity was assessed by checking for potential interferences from blank samples or related compounds formed during sample preparation. Precision and accuracy were evaluated both within and between runs, with a maximum coefficient of variation (CV%) of 10.28 and a maximum relative error (ER%) of 5.38. Linearity was demonstrated over the range of 5 ng/L to 1000 ng/L, with a lower limit of quantification at 0.042 ng/L, made possible by injecting larger sample volumes. A matrix effect of 89% was considered acceptable when compared to standard solutions. Plasma concentrations of MTP were monitored in patients administered either 50 mg or 100 mg doses. For the 50 mg dose, plasma levels reached up to 34 μg/L, while the 100 mg dose produced concentrations ranging from 3.56 to 50.81 μg/L. Although the higher dose generally resulted in elevated plasma levels, significant variability was observed. A strong correlation (r = 0.992) was found between the administered dose and plasma concentration, though variations in absorption rates and patient demographics likely contributed to the observed variability. This method provides a reliable analytical approach suitable for pharmacokinetic and clinical studies involving metoprolol.

1. Introduction

Metoprolol (MTP) is a selective beta-1 adrenergic receptor blocker with minimal effects on beta-2 receptors. It reduces cardiac output by exerting both chronotropic and inotropic effects, without exhibiting the membrane-stabilizing properties typical of sympathomimetic drugs [1]. MTP is commonly administered in either succinate or tartrate salt forms, with the choice depending on whether immediate-release or extended-release formulations are required [2].

The preference for a particular formulation is due to differences in bioavailability. The succinate form, which has lower bioavailability, is often used for extended-release products to achieve a sustained therapeutic effect. MTP is widely prescribed for the management of heart conditions and is frequently used in combination with treatments for related cardiovascular diseases [3].

Metoprolol is primarily metabolized by cytochrome P450 enzymes, including CYP3A4, CYP2B6, and CYP2C9, which are responsible for its α-hydroxylation. This metabolism results in a decrease in MTP levels and an increase in the plasma concentration of its active metabolite, α-hydroxymetoprolol (HMT). While several CYP enzymes contribute to this process, CYP2D6 plays the most significant role, although it does not substantially affect the overall MTP/HMT metabolic pathway [4,5].

In a study involving six subjects, a single oral dose of 100 mg of MTP produced peak plasma concentrations ranging from 0.03 to 0.28 mg/L, with a mean concentration of 0.13 mg/L reached approximately 2.5 h post-administration [6]. Due to extensive first-pass metabolism, MTP exhibits low bioavailability. The drug can cross both the blood-brain barrier and the placental barrier, and it is also excreted in breast milk [7].

Despite its low bioavailability, a significant portion of orally administered MTP reaches systemic circulation [8]. MTP’s limited ability to cross the blood-brain barrier reduces its central nervous system effects, while placental transfer results in lower fetal blood concentrations compared to maternal levels [9].

MTP concentrations in plasma or urine have been detected with a limit of 240 μg/L using diode-array detection (DAD) [10]. The enantiomers of MTP have also been quantified in plasma with a detection limit of 1 μg/L using fluorescence detection [11,12].

A common approach to analyzing MTP involves liquid-liquid extraction followed by LC-MS/MS. In one method, MTP, its two major metabolites, and the internal standard chlorpropamide were extracted from plasma (50 μL) using ethyl acetate. Chromatographic separation was performed on a Moon CN column using isocratic elution with a mobile phase of methanol containing 0.1% formic acid (60:40, v/v) in water, at a flow rate of 0.3 mL/min. Detection was achieved through mass spectrometry, employing positive electrospray ionization (ESI) in selected reaction monitoring mode [13].

Another method for quantifying MTP, HMT, and O-desmethylmetoprolol (DMT) in rat plasma involved protein precipitation followed by separation on a C18 column (4.6 × 250 mm, 5 μm) at a flow rate of 1.0 mL/min. A post-column split (1:4) reduced the flow to 0.2 mL/min for MS detection [14].

Automated sample preparation techniques are commonly used to speed up sample clean-up and enhance throughput in analytical procedures [15]. These include on-line solid-phase extraction (SPE) [16], automated off-line SPE methods [17], and TurboFlow technologies [18]. These methods employ specialized columns with unique configurations to effectively isolate analytes from biological matrices.

This study aims to present a reliable method for determining metoprolol concentrations in human plasma, utilizing automated sample preparation, chromatographic separation, and mass spectrometric detection. This method provides robust results suitable for pharmacokinetic and clinical studies.

2. Materials and Methods

2.1. Chemicals and Reagents

In this study, metoprolol tartrate (MTP) was used as the working standard and was obtained from Sigma-Aldrich (Burlington, VT, USA). Bisoprolol fumarate (BSP) served as the internal standard. Methanol, acetonitrile, and formic acid, all HPLC grade, were sourced from Merck (Darmstadt, Germany). Water was purified in-house using a GenPure UV-TOC system Thermo Scientific (Waltham, MA, USA). Certified biological plasma for sample preparation was purchased from TCS Biosciences (Buckingam, UK). Human plasma samples were collected from the Clinical Hospital of Emergency “Sf. Spiridon” (Iasi, Romania), with ethics committee approval. All patients provided informed consent for the use of their biological samples in this research project.

2.2. Instrumentation

Chromatographic determinations were carried out using a Transcend TLX high-performance liquid chromatography (HPLC) system from Thermo Scientific—(Waltham, MA, USA). The system was equipped with two LC pumps (UltiMate HPG-3200RS and UltiMate LPG-3400RS), each capable of operating at pressures up to 1000 bar. Additional components included an open autosampler (PAL2 HTS-xt) with valve modules, a cooled sample tray, and a column thermostat (TCC-3000RS). A valve interface module (Thermo Scientific UltiMate 3000 Transcend2) was coupled with a TSQ Quantum Access Max Mass Spectrometer for detection.

Chromatographic separation and on-line sample clean-up were performed using a Thermo Gold C18 column (50 × 2.1 mm, 1.9 µm particle size) (C1) and a TurboFlow Cyclone-P column 50 × 0.5 mm (C2). The chromatographic system was controlled using Aria software, while Xcalibur software ver. 4.2. SP1 was used for MS system tuning and chromatographic data processing.

2.3. Chromatographic Conditions

The mobile phase for the analytical column consisted of 0.1% (v/v) formic acid in water (FMA(C1)) and acetonitrile containing 0.1% (v/v) formic acid (FMB(C1)). For on-line sample clean-up, a mixture of 0.1% (v/v) formic acid in water (FMA(C2)) and methanol (FMB(C2)) was used.

The Cyclone P column was operated at room temperature, while the analytical column was maintained at 50 °C during analysis. The injection volume was set to 100 µL to achieve a lower limit of quantification. The analytical column temperature was controlled at 40 °C.

The flow rate and gradient for the mobile phases were established, as shown in

Table 1. Elution program for the TurboFlow column (C2) and analytical column (C1).

Step	Duration (s)	Flow mL/ min	Grad	MFA (C2) %A	MFB (C2) %B	Tee	Loop	Flow mL/ min	Grad	MFA (C1) %A	MFB (C1) %B
1	60	1.5	Step	100		== 1	out 3	0.3	Step 5	50	50
2	30	0.1	Step	40	60	T 2	in 4	0.1	Step	50	50
3	60	2.0	Step	-	100	==	out	0.3	Step	50	50
4	60	2.0	Step	60	40	==	in	0.3	Step	50	50
5	60	1.5	Step	100		==	out	0.3	Step	50	50

¹ ==: TurboFlow column and analysis column are independent. ² T: TurboFlow column connected with the analysis column. ³ out: Loop is not connected to any other pipeline. ⁴ in: Loop access system is located before the TurboFlow column. ⁵ Step: instant modifications of the composition of mobile phase.

. The sample was loaded onto the TurboFlow column using FMA(C2) at a flow rate of 1.5 mL/min via the loading pump. The target compounds were retained at the front of the TurboFlow column for 60 s to remove matrix components and concentrate the analytes (Step 1). The elution flow rate was then gradually reduced to 0.1 mL/min over 30 s. The mobile phase composition was adjusted to 40% FMA(C2) and 60% FMB(C2), which transferred the mobile phase from the loop and eluted the target compounds from the TurboFlow column to the analytical column within 30 s (Step 2) by switching the Tee valve to the inline position (T).

Next, the elution pump separated the target compounds on the UHPLC column using an isocratic method, with a mobile phase consisting of 50% FMA(C1) and 50% FMB(C1) (Step 3). Simultaneously, the loading pump reconditioned the TurboFlow column with 100% FMB(C2) for 60 s (Step 3). Finally, the loop was filled with the solution in preparation for the next sample (Step 4), followed by re-equilibration of the system (Step 5).

2.4. Mass Spectrometry Conditions

The mass spectrometer was a triple-quadrupole system that utilized multiple reaction monitoring (MRM) for detecting the target compound and the internal standard. The detection was carried out under the following conditions: ionization voltage of 4 kV, sheath gas pressure of 20 psi, auxiliary gas pressure of 30 psi, and collision gas pressure of 1.5 mTOR.

The heated electrospray ionization (HESI) probe was set to 200 °C, and the capillary temperature was 330 °C. The quantification transitions for detection were m/z 268.1 → m/z 130.96 for MTP and m/z 326.3 → m/z 116.2 for the internal standard. Quantification transitions used were m/z 268.1 → m/z 115.6 for MTP and m/z 326.3 → m/z 222.5 for BSP.

2.5. Preparation of Standards, Quality Control and Sample Solutions

Standards for the calibration curve were prepared by serial dilution of the initial stock solution in certified plasma, with concentrations calculated for both MTP and BSP. To ensure stability, these solutions were stored at 2–8 °C for a maximum of five days.

A stock solution of BSP at a concentration of 1.0 ppm was prepared, and 100 µL was used in sample preparation, resulting in a final concentration of 100 ng/L. Similarly, 100 mg of MTP was dissolved in water and diluted to 100 mL with the same solvent. A 1 mL aliquot of this solution was further diluted to 100 mL using the mobile phase. For method validation, 0.5 mL of certified plasma was spiked with varying volumes of the standard stock solution and brought to a total volume of 1 mL with water. For the internal standard, 0.5 mL of certified plasma was mixed with 0.1 mL of the stock internal standard solution and diluted to 1 mL with water.

Calibration standards were prepared with concentrations ranging from 5 to 1000 ng/L by diluting 50 µL of various stock solutions in a 50% certified plasma and 50% water mixture (v/v). The resulting concentrations were 5.0, 25.0, 100, 250, 500, and 1000 ng/L. Quality control (QC) samples were prepared similarly to the calibration standards, with concentrations of 25, 500, and 1000 ng/L. Different sets of QC samples, zero solutions, and blank solutions were used for method validation.

Human plasma samples were stored at −20 °C to preserve stability. Prior to analysis, aliquots were thawed and conditioned at room temperature. Samples were centrifuged at 14,000 rpm to remove any coagulated fibrin. A 0.5 mL aliquot of each sample was spiked with 0.1 mL of internal standard solution and diluted to 1 mL with water. The samples were then filtered through a 0.22 μm PTFE filter and transferred to injection vials for analysis.

2.6. Method Validation

Method validation parameters—selectivity, carryover, LLOQ, linearity, accuracy, precision, matrix effect, and stability—were assessed per FDA [19] and EMA [20] guidelines.

Selectivity was confirmed using six sources, and the calibration curve was linear between 5.0–1000 ng/L. LLOQ was determined with five replicates, showing a signal-to-noise ratio five times higher than blank samples. Dilution integrity was tested at three levels (2×, 5×, and 10×), and precision/accuracy was validated with six replicates, ensuring results within 15% of nominal values and a CV within ±15%. The matrix effect was evaluated by comparing extracted samples to standard solutions, with acceptable CV levels. Stability was confirmed under various conditions, ensuring MTP remained stable during freeze, freeze and thaw, post-preparative conditions, and standard solutions for 48 h at 4 °C.

3. Results

3.1. Method Development

The loading step involved selecting the optimal TurboFlow column based on its chemistry and polymeric stationary phase, which provided the best retention time for the target compounds to coincide with the transfer step. The mobile phase that yielded the strongest peak during transfer, the smallest peak during loading, and the least carryover was evaluated. Three pH options were tested: pH 3 (0.1% formic acid in water), pH 6 (10 mM ammonium acetate in water), and pH 8 (10 mM ammonium acetate in water, adjusted with ammonium hydroxide). The pH 3 provided the best retention, while at pH 6 and 8, analytes exhibited shorter retention times, consistent with MTP’s first dissociation constant of 9.56.

For the transfer step, the optimal solvent concentration was determined to be 60% acetonitrile with 0.1% formic acid. This concentration ensured complete transfer, as lower concentrations resulted in incomplete transfer, and higher concentrations reduced peak intensity for both MTP and BSP. The ideal flow rate combination was 0.1 mL/min for both the loading and eluting pumps, which provided complete transfer with clear peaks during the elution step and no peaks during the transfer step. Testing a higher flow rate of 0.5 mL/min over 10 s caused excessive pressure and did not improve peak shapes.

In the final step, loop reconditioning was optimized by adjusting the organic solvent content. Based on the results from Step 2, a solvent concentration of 40% acetonitrile provided the best performance, ensuring no carryover and maximum peak intensities when mixed with the 60% acetonitrile from the previous step.

Compared to other off-line and on-line methods, the protocol offers several advantages. In terms of analysis time, the method has a shorter duration of 4.5 min compared to on-line SPE methods [21]. On-line SPE methods have a longer reconditioning period due to their different operating principles, similar particle sizes, and laminar flow in the case of SPE columns. In contrast, the TurboFlow method operates with turbulent flow, which has an increased flow section and ensures efficient mass transfer.

3.2. Method Selectivity

Selectivity was studied against 6 sources of biological samples. The analytical method was both selective and sensitive enough to evaluate MTP and the internal standard with sufficient specificity. In this study, some residual signals were present on blank samples. As a result, several other possible metabolites of the MTP were monitored to investigate the extent of any interference caused by metabolites, interference from degradation products formed during sample preparation, and interference from possible co-administered medications.

In the case of the analyzed samples, HMT and DMT were considered interfering metabolites for the determination. According to other studies [12], specific mass spectrometry conditions were used. The samples were analyzed in terms of stability and lack of interferences. For HMT transition m/z 284.1 → m/z 116.1 was used, while transition of DMT was m/z 254.1 → m/z 116.1 and these were monitored during MTP elution. In the case of BSP, no degradation or metabolization products were produced under the method conditions.

Chromatograms indicate the presence of signals that may be associated with possible metabolites, but their levels reveal a concentration of less than 20% of the quantification limit. As stated, some peaks were found for the HMT; however, the level of interference was at most 0.23% in relation to MTP at the lower limit of quantification. Specific chromatograms are presented in Figure 1.

3.3. Linearity

Linearity was evaluated to demonstrate the method’s ability to produce results directly proportional to the analyte concentration within a specified range. For the quantification of MTP in biological samples, the linear range was defined as being between 5.0 and 1000 ng/L.

The internal standard (BSP) was added to each solution at a concentration of 100 ng/L. The correlogram showed a linear relationship between sample concentration and the peak area ratios, with a correlation coefficient (R²) of 0.999, based on the average values of the five slopes included in the study. The mean regression equation was Y = 564.79x + 0.439. An analysis of variance (ANOVA) applied to the slope and intercept confirmed a significant correlation between the variables.

This statistical model was used to verify that the correlation accurately approximated the data for MTP. In Fisher’s statistics, the p-value was less than 0.05, indicating that the regression slope provided a reliable approximation of the data. Moreover, no influence of the independent variable on the dependent variable was observed.

The t-statistic for the slope was greater than 2.086, which corresponds to 4 degrees of freedom and a 95% confidence level. Individual relative deviations for MTP reached a maximum of 7.77%, within the range of ±20%. The t-test (p < 0.05) confirmed that the linear model was appropriate, establishing a linear relationship between sample concentration and peak area.

The SS and MS parameters in

Table 2. Statistical evaluation of slope used in the linearity of the method.

Coefficients of Regression		Residual Output
Parameter	Value	Conc (ng/L)	Predicted (ng/L)	Residuals	Percentile
Slope	564.79	5	5.24	−0.24	4.59
Standard error	2.94	25	23.32	1.68	7.22
p values	4.69267 × 10−8	100	92.79	7.21	7.77
t Stat	0.099	250	263.66	−13.6	5.18
F	11,304.13	500	495.81	4.19	0.84
Min. 95%–Max. 95%	550.04 ÷ 579.54	1000	999.18	0.82	0.08

had close values, so the F ratio is 11,304.13, which is higher than the critical value, suggesting the independent values (concentration of substance) influence the variation of the dependent parameters (peak area).

3.4. Lower Limit of Quantification (LLOQ)

The lower limit of quantification (LLOQ) is defined as the minimal concentration that can be quantified with acceptable accuracy and precision. The LLOQ was determined by injecting a plasma sample with a concentration of 5 ng/L. For the calculation of both the lower limit of detection (LLOD) and the LLOQ from calibration standards, a regulatory approach was followed, as outlined in the relevant guidelines [22]. Based on the standard deviation of residuals (1.53) and the slope of the calibration curve, the LLOD was calculated as 0.012 ng/L, and the LLOQ was determined to be 0.042 ng/L.

Other studies have reported an LLOQ of 0.1 ng/L for MTP in bioequivalence studies [23,24]. A key advantage of the TurboFlow system is its ability to handle high-volume loads on the Cyclone column without compromising the peak shape of the analytical column. During the method transfer, only the compounds of interest are concentrated at the head of the analytical column, preserving their integrity.

The signal-to-noise ratio at the lowest calibration point was 453, confirming that the method is highly sensitive for detecting MTP (Figure 1a). Additionally, the LLOQ met the regulatory requirement of being within 5% of the upper limit of quantification (ULOQ).

3.5. Dilution Integrity

Dilution integrity was tested on two- (2 × QC), five- (5 × QC), and ten-fold (10 × QC) dilutions. The samples were subjected to analysis of accuracy and precision and the recovered values were within the range of ±15%. Recoveries were 91.25% for 2 × QC dilution, 87.5% for 4 × QC dilution, and 110.1% for 10 × QC dilution. Also, the CV% were 3.64% (2 × QC), 5.28% (4 × QC), and 6.25% (10 × QC), respectively.

3.6. Precision and Accuracy

Precision, expressed as the recovery of MTP, and accuracy, which reflects the closeness of determined values to the nominal concentrations, were assessed using quality control (QC) samples with known amounts of MTP. The QC samples were analyzed by comparison to the calibration curve, and the calculated concentrations were compared to their nominal values.

For precision, the mean concentrations did not deviate by more than 15% from the nominal values of the QC samples. Specifically, recovery ranged from 102%, for the lowest concentration on the first day, to 110.28% for the second series of analyses. For the medium concentration, recovery was 99.26% in the first analysis series and 109.15% in the second. At the highest concentration of 1000 ng/L, which is at the upper limit of quantification, recovery values were 104.95% for MTP in the first analysis and 106.43% in the reanalysis.

The coefficients of variation (CV) for precision did not exceed 15% across all QC samples, with a maximum CV of 10.58% for MTP. Regarding accuracy between analytical determinations, the CV also did not exceed 15%. Specifically, the CV values for the 25 ng/L level were 5.6% and 10.58%. For the medium concentration level, the CV values were 10.28% and 2.85%, while for the highest concentration, the CV values were 9.26% and 8.53%. In all cases, the coefficient of variation was independent of concentration.

The average inter-day and intra-day CV values were 8.09% for QC1, 6.56% for QC2, and 8.89% for QC3 (see

Table 3. Accuracy and precision of determinations for QC samples. (QC1—25 ng/L, QC2—500 ng/L and QC3—1000 ng/L).

Nominal Conc. (ng/L)	Intra-Day (n = 3)			Inter-Day (n = 9)
	Found (ng/L) (Mean ± SD)	RSD (%)	Accuracy (%)	Found (ng/L) (Mean ± SD)	RSD (%)	Accuracy (%)
25	27.57 ± 1.54	5.6	110.28	25.53 ± 2.7	10.58	102.12
500	496.3 ± 51	10.28	99.26	545.76 ± 15.55	2.85	109.15
1000	1049.56 ± 97.22	9.26	104.95	1064.29 ± 90.82	8.53	106.43

).

3.7. Matrix Effect

The matrix effect (MF) on MTP analysis was evaluated using QC samples prepared from six different plasma sources and compared with standard solutions (SS1 and SS2) at equivalent concentrations (25 ng/L and 1000 ng/L), which were prepared using water as the solvent. The matrix effect was calculated as the ratio of the peak area in the presence of the matrix (QC samples) to the peak area in the absence of the matrix (standard solutions).

In addition, the normalized matrix factor (NMF) was calculated by dividing the MF of the analyte by the MF of the internal standard. The MF and NMF values for MTP and the internal standard are shown in

Table 4. Matrix factor calculation based on matrix effect for concentrations of 25 ng/L and 1000 ng/L.

	QC1 (25 ng/L) (n = 6)		QC2 (1000 ng/L) (n = 6)
Target compound	MTP	IS (BSP)	MTP	IS (BSP)
Matrix factor (MF)	0.942	0.807	0.913	0.855
CV (NM) (%)	5.59	8.40	5.95	3.52
Normalized MF (NMF)	1.095		1.069
CV (NMF) (%)	8.11		6.71

. The coefficients of variation (CV%) for the concentrations were 5.95% for MTP (QC2) and 8.40% for the internal standard (QC1). The overall CV for NMF was 10.73%, which complies with the bioanalytical method validation guidelines [21].

3.8. Stability Assessment

Stability was ensured for each stage of the analytical method, which means that the conditions applied to the stability tests were similar to those used for the actual study samples.

An evaluation of certified plasma samples spiked with 25.0, 500.0, and 1000 ng/L of MTP was conducted after being frozen at −20 °C for 7 days to establish a maximum storage time in the freezer. The main recoveries were established to be 108.53 ± 2.44% (QC1), 95.09 ± 3.33% (QC2), and 97.92 ± 1.15%.

Standard stock solutions were evaluated over a 48 h interval, stored at 4 °C. A series of solutions freshly prepared were analyzed and the results were evaluated using a calibration curve. As stated in the matrix effect, higher values were recorded for recovery, namely 29.92 ± 0.41, 526.89 ± 6.43, and 1163 ± 21.44 ng/L, indicating good stability during the time interval.

Determinations were made using QC samples containing high (1000 ng/L), medium (500 ng/L), and low concentrations (25 ng/L). Long-term stability was evaluated elsewhere [25], so we have considered the stability for samples that were processed at room temperature.

In the case of post-preparative monitoring for MTP and the BSP, the stability was verified for a period of up to 24 h in the autosampler. A significant decrease appeared after 24 h, especially for the low concentration samples, where values showed a significant decrease of up to 14.68 ng/L.

The freeze and thaw conditions demonstrated sample stability for two cycles; however, for the third cycle, the concentrations were found to be inconclusive. The main recovery covering all concentrations was 98.81 ± 1.57%.

QC samples were analyzed as compared to a calibration curve obtained from freshly injected calibration standards, and the obtained concentrations were compared to the nominal concentrations and revelant data are presented in

Table 5. Stability evaluation of MTP in certified plasma (50, 500, and 1000 ng/L).

Conditions	Nominal Conc. (ng/L)	Concentration Stability (ng/L)
		Recovery (%)	Recovered Conc. (ng/L)	CV%
Freeze (−20 °C) 7 days	25	108.53	27.13 ± 0.66	2.44
	500	95.09	475.43 ± 15.84	3.33
	1000	97.92	979.2 ± 11.3	1.15
Freeze and thaw (3 cycles) (−20 °C)	25	100.20	25.05 ± 0.85	3.38
	500	96.87	484.37 ± 4.14	0.85
	1000	99.37	993.69 ± 4.62	0.47
Post-preparative (24 h stable, room temperature)	25	103.95	25.99 ± 7.48	2.78
	500	122.24	611.18 ± 52.63	8.61
	1000	114.41	1144.13 ± 38.87	3.4
Standard solutions (48 h, 4 °C)	25	119.67	29.92 ± 0.41	1.38
	500	105.38	526.89 ± 6.43	1.22
	1000	116.30	1163 ± 21.44	1.84

.

3.9. Application of the Method for the Evaluation of Real Biological Samples

This method was tested on 40 individual human samples. Out of these 40 samples, 20 individual cases were patients whose medication consisted of MTP 50 mg and the other patients had a medication of MTP 100 mg. Sampling was performed after administration and within a time interval of 2.5 h from the last dose. For the 50 mg dose, concentrations are generally lower and spread over a wider range. For the 100 mg dose, the values are higher on average, and the concentrations are more varied, with a clearly marked maximum value (P1) (Figure 2a).

All the blood samples were stored in a freezer at −20 °C during a time interval that did not exceed the stability of the substance in storage conditions, namely 30 days. As the samples were arbitrary, evaluations characteristic of bioequivalence were not included; instead, a simple monitoring of active substance concentrations in relation to the administered quantity was performed.

In all the subjects included in the study, MTP was found, and the levels found were up to 34 μg/L for 50 mg MTP. For the patients to whom 100 mg MTP was administered, the concentration range was between 3.56 and 50.81 μg/L. The comparison between concentrations for the two doses (50 mg and 100 mg) highlights significant variations depending on the sample. In some cases, a higher dose does not always result in a significant increase in concentration (e.g., P2 and P11) (Figure 2b), and all data are consistent with other findings [26].

Considering the median values, as well as the minimum and maximum concentrations, the correlation between the quantity administered and plasma levels had a correlation coefficient of 0.992. Expecting a direct proportional relationship between the administered quantities (50 mg and 100 mg) and maximum plasma concentrations, the observed values were 34 μg/L for 50 mg of MTP administered and 50 μg/L for 100 mg of MTP administered. This situation is created by possible different absorption rates of MTPs but also, can be associated with different demographic data. It shows that, in most cases, the concentration for the 100 mg dose is higher than for the 50 mg dose, but there is no direct linear relationship between the two doses. Instead, there is considerable variability (Figure 2c).

4. Conclusions

The validated method for MTP determination included a fast pre-treatment stage, where samples were filtered, spiked, and diluted with water for full analysis.

The chromatographic system employed a two-stage sample transfer from the TurboFlow column to the analytical column, using a mixture with higher isoelutropic strength. The conditions at each stage adhered to standard chromatographic principles. During the loading phase, an aqueous acidic solution was used in a polymeric column, which retained the analytes. For the transfer stage, a mixture containing 60% methanol provided higher elution strength, which was rapid, efficient, and enabled proper elution of analytes onto the analytical column. Proper control of this step was crucial—if not well established, chromatographic peaks for MTP and BSP could be lost or eluted without retention.

While other automated methods for MTP determination, such as on-line SPE, are available and validated, the TurboFlow method demonstrated higher throughput, as the analysis time was considerably shorter. The use of a high injection volume also allowed for superior lower limits of quantification (LLOQ) compared to other on-line and off-line methods.

Selectivity was demonstrated using plasma from different sources, with minimal carryover observed. Two metabolites of MTP were also monitored to confirm that no hydrolysis occurred under the MS conditions. The matrix effect was negligible, indicating that the proposed clean-up method efficiently removed interfering matrix constituents.

The method showed a wide linearity range (5 ng/L to 1000 ng/L), with points close to the calibration curve. Residuals exhibited heteroscedasticity, but no outliers were identified outside the confidence level of the regression. The accuracy and precision values met bioequivalence guidelines, confirming that the method is reliable for use in bioequivalence and clinical studies involving the analysis of MTP, BSP, or their metabolites in biological fluids.

Following validation, the method was applied to biological samples from the patients included in the study. The method successfully identified and quantified metoprolol levels ranging from 0.1 μg/L to 50 μg/L, which correlated well with other findings.