Rapid determination of five common toxic alkaloids in blood by UPLC–MRM–IDA–EPI: Application to poisoning case

Toxic alkaloids are typically found in herbal medicines and have strong pharmacological effects and a broad therapeutic spectrum. On the other hand, toxic alkaloids exert toxicological activities in vivo; as such they have a narrow therapeutic window and can induce poisoning due to incorrect dose or misuse. In this view, there is an urgent need to develop a rapid and sensitive assay to detect these toxic alkaloids. This study developed a method for determining five common toxic alkaloids in blood, including brucine, strychnine, aconitine, mesaconitine, and hypaconitine using ultra-high liquid chromatography-tandem quadrupole/linear ion trap mass spectrometry (QTRAP UPLC–MS/MS). The analytes in this investigation were extracted with ether and detected using multiple reaction monitoring (MRM)-information-dependent acquisition (IDA)-enhanced product ion (EPI) scanning modes. SKF_525A served as the internal standard (IS). The approach demonstrated excellent linearity, with a correlation coefficient (R) > 0.9964, and satisfactory sensitivity, with the limit of detection (LOD) of 0.31 ∼ 3.26 ng/mL and a limit of quantification (LOQ) of 1.13 ∼ 11.52 ng/mL. The extraction recovery (ER) was 78.8 ∼ 116.2%, the matrix effect (ME) was −12.3 ∼ 21.2%, and the method accuracy was 0.8 ∼ 12.8%. In addition, the intra-day precision and the inter-day precision (RSD) were 0.7% ∼ 7.4% and 0.4% ∼ 13.5%, respectively. The developed approach is sensitive and efficient, and offer significant application prospect in clinical monitoring and forensic detection of poisoning.     

An UPLC-MS/MS method for the determination of valproic acid in blood of a fatal intoxication case

Valproic acid (VPA) has been used as an anticonvulsant for the treatment of epilepsy. The authors present a fatal case involving a 45-year-old female, found dead lying in bed with empty tablets of Diplexil^® next to her. She was a chronic alcoholic and epileptic who had been under psychiatric treatment, having repeatedly demonstrated intent to commit suicide.A rapid method was developed and validated to determine VPA in blood by ultra-performance liquid chromatography (UPLC) coupled with tandem mass spectrometry (MS/MS) with electrospray ionization source in negative ion mode.The method involved sample treatment with phosphoric acid followed by solid-phase extraction. Chromatographic separation was achieved using an Acquity UPLC^® BEH (2.1 × 50 mm id, 1.7 μm) column and a mobile phase containing ammonium acetate and acetonitrile, at a 0.5 mL/min flow rate. Detection and quantification of VPA was achieved using multiple reaction monitoring (MRM). The MS/MS transitions used for monitoring were m/z 143.1–143.1 for valproic acid and m/z 296.1–205.0 for hydrochlorothiazide used as an internal standard (IS).The limit of quantification (LOQ) was 0.5 μg/mL and the method was linear in the concentration range of 0.5–100 μg/mL. The coefficients of variation obtained for accuracy and precision were less than 10% and the mean recovery was 95% for the three concentrations levels studied (5 μg/mL, 10 μg/mL and 50 μg/mL). Toxicological results showed high concentration of VPA (556 μg/mL) and therapeutic concentrations of tiapride, mirtazapine, oxazepam and nordiazepam. Blood sample analysis also revealed the presence of ethanol at a concentration of 1.34 g/L.A specific, selective and sensitive method for the determination of VPA in blood was developed and can be used in routine forensic investigation. Toxicological results led the pathologist to rule that death was due to an intoxication caused by the simultaneous ingestion of high VPA concentrations and alcohol, with a suicidal legal-medical etiology.     

Simultaneous quantification by LC/ESI–MS/MS of chlorinated tyrosine derivatives in the autopsy sample of a victim of chlorine exposure

Direct detection and accurate quantification of chlorine in autopsy samples are difficult because of the volatility and rapid metabolism of chlorine. Here, we developed and validated a method for quantitative analysis of 3-chloro-l-tyrosine (Cl-Tyr) and 3,5-dichloro-l-tyrosine (DiCl-Tyr) as stable markers of chlorine exposure. Chemical derivatization followed by liquid chromatography coupled with electrospray ionization–tandem mass spectrometry (LC/ESI–MS/MS) enabled us to simultaneously analyze both Cl-Tyr and DiCl-Tyr in an autopsy sample from the victim of chlorine exposure. Cl-Tyr was detected in the heart blood (53.6 ng/mL), urine (9.5 ng/mL), and lung tissue (211.1 ng/g); however, DiCl-Tyr was detected only in the lung tissue (10.3 ng/g). In contrast, in autopsy samples obtained from cases without exposure to chlorine, DiCl-Tyr was not detected in any matrixes. Our result suggested that the simultaneous detection of Cl-Tyr and DiCl-Ty may provide a better appreciation of chlorine exposure. To our knowledge, this is the first time Cl-Tyr and DiCl-Tyr have been determined simultaneously in a real human autopsy sample from a victim of chlorine exposure.     

Development of an LC–MS/MS method for quantification of 3-chloro-L-tyrosine as a candidate marker of chlorine poisoning

A simple and sensitive liquid chromatography coupled with electrospray ionization-tandem mass spectrometry (LC/ESI–MS/MS) method for the determination of 3-chloro-L-tyrosine (Cl-Tyr) was developed and validated. For sample preparation, 50 μL of the body fluids or tissue extracts were processed by protein precipitation followed by the derivatization with dansyl chloride. The calibration curve was linear over the concentration range of 2.0–200 ng/mL blood or 4.0–400 ng/g tissue. Our method allowed the reproducible and accurate quantification. That is, the intra- and inter-assay coefficients of variation were below 7.73 and 6.94%, respectively in both the blood and lung. We applied the developed method to the analysis of Cl-Tyr in the human autopsy samples, which were suspected of chlorine poisoning, and detected 55.2 ng/mL and 206.6 ng/g Cl-Tyr in left heart blood and lung, respectively. Furthermore, in more than 20 autopsy samples, which were obtained from other causes of death including burn, drowning, hanging, internal disease, trauma and drug poisoning, Cl-Tyr was almost not detected in their both body fluids and organ tissues. In conclusion, the data here reported demonstrate that the LC/ESI–MS/MS method allows the Cl-Tyr in the autopsy samples and that chlorine exposure strongly affects its level, providing a basis for novel identification tool of chlorine poisoning.     

Analytical method for detection and quantification of new emerging drug etaqualone in human blood and urine by gas chromatography tandem mass spectrometry

Analytical procedure for detection and quantification of etaqualone in human blood and urine using GC–MS/MS was established and applied to authentic human samples obtained from volunteers. A liquid–liquid extraction method was employed. Each 1.0 mL of blood or urine was alkalized and extracted with diethyl ether. The solvent layer was evaporated to dryness and reconstituted with methanol then analyzed by GC–MS/MS. linear relationships within the concentration range of 1–100 ng/mL were obtained in calibrators for both blood and urine, demonstrating correlation coefficients values being>0.999. For blood and urine samples, the intra-day assay precision and accuracy values are each less than 3.65%, 7.13%, and 6.02%, 9.12%; those values of the inter-day assay are each less than 1.82%, 6.74%, and 3.99%, 7.41%. The extraction recovery rates for etaqualone ranged from 98.7% to 106%. The lower limit of quantifications was 1.0 ng/mL in both blood and urine. Stabilities of etaqualone in blood and urine were satisfactory under various temperatures within 15 days. 8.51 and 2.06 ng/mL of etaqualone in blood and urine were detected at 4 h later oral ingestion; 6.91 and 3.94 ng/mL of etaqualone were also detected 30 min and 2 h later smoking from blood and urine.     

Quantitative analysis of 3,4-dimethylmethcathinone in blood and urine by liquid chromatography–tandem mass spectrometry in a fatal case

We report here the quantitative analysis of cathinone-type designer drug 3,4-dimethylmethcathinone (3,4-DMMC) in blood and urine using liquid chromatography–tandem mass spectrometry (LC–MS/MS) in a fatal case. Abuse of 3,4-DMMC is widespread and a global issue. However, to date, there have been no reports of 3,4-DMMC-related deaths. We encountered a death in which 3,4-DMMC was thought to play a causative role, and successfully identified this designer drug from biological samples by using LC–MS/MS and QuEChERS (quick, easy, cheap, effective, rugged and safe) extraction method. For standard samples, detection of 3,4-DMMC in human blood and urine samples in the calibration range (5–400 ng/mL) was successful with recoveries of 85.9−89.4% (blood) and 95.8−101% (urine), limits of detection of 1.03 (blood) and 1.37 ng/mL (urine) and limits of quantification of 5.00 (blood) and 5.38 ng/mL (urine). The concentrations of 3,4-DMMC in blood (external iliac vein) and urine in the case were 27 mg/L and 7.6 mg/L, respectively. Some metabolites, including 3,4-dimethylcathione (DMC) and β-ketone reduced metabolites (β-OH-DMMC and β-OH-DMC), were detected in both blood and urine.     

Determination of 3-chloro-l-tyrosine as a novel indicator of chlorine poisoning utilizing gas chromatography-mass spectrometric analysis

Chlorine gas exposure occurs in chemical warfare, industrial and household accidents. In forensic science, the generation of chlorine gas by mixing sodium hypochlorite detergent and strong acid detergent cannot be overlooked because of the possibility of suicide method (NaClO + 2HCl → NaCl + H₂O + Cl₂). Though typical autopsy findings are obtained in chlorine exposure, such as pulmonary edema, useful biomarkers don’t exist. In this research, we developed an analytical method of 3-chloro-l-tyrosine (Cl-Tyr) in blood as a novel marker of chlorine poisoning utilizing gas chromatography–mass spectrometry (GC–MS). Cl-Tyr was purified using protein precipitation and cation-exchange solid phase extraction, derivatized by the silylation agent and subjected to GC–MS. The quantification range was 10–200 ng/mL and good reproducibility was obtained. We applied the developed method to analyze Cl-Tyr in autopsy sample, which is suspected of chlorine poisoning, and detected 59.7 ng/mL Cl-Tyr in left heart blood. To our knowledge, this is the first report of determination of the chlorinated biomolecule in the human autopsy sample from chlorine poisoning.     

Diagnostic approach to drug-screening tests for fatal diabetic ketoacidosis: Forensic autopsy of a methamphetamine abuser

To diagnose the cause of death in autopsy cases, systematic examinations, such as macroscopic, pathological, biochemical, and toxicological are important. In this case report, drug examinations also gave very useful information to diagnose the cause of death, fatal diabetic ketoacidosis (DKA). A female methamphetamine abuser in her forties was found dead lying on a hotel bed. Diagnosing her cause of death was difficult only from the macroscopic findings because there was no fatal and/or serious injury or disease. On toxicological examination, acetone was detected at a high concentration (682 μg/mL in blood, 887 μg/mL in urine) using gas chromatography (GC). Using gas chromatography–mass spectrometry (GC–MS), methamphetamine was detected in the blood, urine, hair, and visceral organs; however, these concentrations were low. At the same time, GC–MS examination revealed a high glucose peak. From the results of the biochemical examination of urine, acetoacetic acid was 1940 μmol/L, β-hydroxybutyric acid was 14,720 μmol/L, and glucose was 4620 mg/dL. Histologically, Langerhans’ islets in the pancreas were fibrotic and atrophic, and no insulin-immunoreactive cells were observed. The subsequent police investigation also revealed that she had contracted diabetes mellitus type 1; therefore, we concluded that her cause of death was DKA, due to a lack of insulin injection.     

Simultaneous quantification of urea,uric acid,and creatinine in human urine by liquid chromatography/mass spectrometry

In forensic analysis, the identification of urine or human urine among unknown liquids plays an important role. Urea, uric acid, and creatinine are major organic compounds found in human urine. Previous studies have reported that the concentration quotients of these three compounds can be used as an index for the identification of human urine. Here we describe a method for the simultaneous quantification of urea, uric acid, and creatinine in human urine by liquid chromatography/mass spectrometry (LC/MS), with the aim of forensic identification of human urine. Separation of the three analytes was achieved by hydrophilic interaction chromatography, using a TSK gel Amide-80 column with a mobile phase composed of acetonitrile and ammonium acetate aqueous solution, coupled with detection using a mass spectrometer. For quantification, melamine and violuric acid were used as internal standards. Human urine samples were pretreated for LC/MS analysis by dilution with LC mobile phase, followed by centrifugation and filtration. The analytes and internal standards were separated within 9 min. The linear ranges were 2.0–40.0, 0.10–1.60, and 0.13–2.00 mg/mL for urea, uric acid, and creatinine, respectively, with correlation coefficients > 0.99. The intra- and inter-day accuracies of the analytes were − 10.6% to 7.4%, and the precision was within 7.6%. For all analytes, no significant matrix effects were observed and recoveries ranged from 95.4% to 104.6%. Quantitative results of 3 analytes were obtained within their linear range from 10 human urine samples and the quotients, UA/UN × 20 and UA/Cre, were calculated based on previous reports.     

Quantification of the benzimidazole opioid analog isotonitazene in human hair using liquid chromatography-tandem mass spectrometry

Benzimidazole opioids were originally developed from the late 1950s to 1970s as analgesics for medical use, although a lot of them could not be approved as licit medicines because of their severe side effects and physical dependence. Such benzimidazole opioid analogs as abused drug, however, have recently been found in illicit drug markets throughout the world. Isotonitazene is one such benzimidazole opioids, whose analgesic potency can be as much as 500 times greater than that of morphine, according to previous animal studies. In line with this potency, a couple of hundred fatalities related to it were reported to date. In this study, a well validated method for the quantification of isotonitazene in human hair samples using liquid chromatography (LC)–tandem mass spectrometry (MS/MS) was established, and could be applied to authentic samples which were seized by the police security bureau. Isotonitazene concentrations in the seized hair averaged 6.11 pg/mg. The LLOQ and LOD of this method were 1.25 and 2.5 pg/mg, respectively; the calibration curve of the substance in hair samples showed a good linearity in the concentration range of 2.5–250 pg/mg (r > 0.999); the extraction recovery rates were 87.3–105% in the tested range; the inter- and intra-day precisions and accuracies (%biases) were not greater than 9.09% for each determination. Isotonitazene in human hair showed good stability at room temperature and under dark storage conditions for 30 days. As for matrix effect in hair samples, moderate ion suppression of target substances could be found. This is the first report for the analysis of isotonitazene in human hair samples.     

Blood concentrations of clobazam and norclobazam in a lethal case involving clobazam,meprobamate and clorazepate

Clobazam is a benzodiazepine with anti-anxiety and anticonvulsant properties marketed in several countries. Norclobazam, a metabolite of clobazam, has similar pharmacological activity but weaker sedative and tranquilizing effect. The two drugs were detected by GC–MS and determined by HPLC-DAD in the samples from a postmortem case. The femoral blood concentrations of clobazam and norclobazam were 0.72 and 36 μg/mL, respectively. The concentration of the active norclobazam was very high. The sum of both clobazam and norclobazam blood concentration (36.72 μg/mL) was clearly toxic, but was not necessarily fatal. Other associated drugs concentrations were within their therapeutic ranges. Interactions due to drug association were discussed.     

Detection and quantification of psychotropic drug etaqualone in human hair using GC–MS/MS

In this study, sensitive analytical procedure for detection and quantification of etaqualone in human hair samples using gas chromatography tandem mass spectrometry (GC–MS/MS) was newly established, and applied it to authentic human samples obtained from an abuser. In this method, the hair samples were treated with hydrochloric acid and then extracted with ethyl ether. The ether layer was dried in a warm water bath, and the residue was reconstituted in ethyl acetate, followed by GC–MS/MS analysis. Multiple reaction monitoring (MRM) mode was used for data collection, and quantitative analysis was performed using internal standard method. Good linear relationship within the concentration range of 1–100 pg/mg were obtained in calibrators for the hair samples showing its correlation coefficient value was 0.9993. The lower limit of quantitation in this study was 1 pg/mg and the recovery rate examined ranged from 100.4% to 108.5%. The intra-day precision and accuracy were less than 5.0% and 5.8%, respectively. The inter-day precision and accuracy were lower than 6.4% and 4.6%, respectively. Using this established method, etaqualone could be detected in the hair sample obtained from a suspected user to be level of 65.2 pg/mg. It should be expected that the method established in this study would contribute to rapid detection and identification of psychotropic drug etaqualone among multiple fields including forensic investigation, clinical application and of course public health matters.     

Specific interpretation of hair concentrations in 2 fatal metformin intoxication cases

Hair analysis is very useful for toxicological investigations since, by providing a wider detection window, it gives the possibility to perform a retrospective study on the historical consumption of a substance. Unfortunately, there are no data available for hair concentrations in metformin-related deaths. In this study, the authors present 2 cases of fatal metformin intoxication in which, for the first time, hair analysis was performed using a specific GC–MS/MS method. Metformin was tested positive in femoral blood (112.3 mg/L and 64.7 mg/L respectively) and cardiac blood (226.9 and 203.2 mg/L) of the two subjects. For case 1, other samples were also tested positive, including vitreous humor (31.1 mg/L) and gastric contents (773.5 mg/L). In case 2, metformin was measured at 844.9 mg/L in urine. Metformin hair concentrations were 28.3–44.8 and 22.5 ng/mg for both cases, respectively. The concentrations found in the 2 fatal cases are clearly higher than those obtained in a previous study with subjects under treatment (0.3–3.8 ng/mg) or those found in 3 post-mortem cases where metformin death was excluded (0.6–1.4 ng/mg). Excessive sweating during the agonal phase due to fatal hypoglycemia could explain these elevated concentrations as sweat can have contaminated the hair.     

Assessment of blood 2-aminothiazoline-4-carboxylic acid concentrations: Age and sex differences,and correlation with carboxyhemoglobin in fire victims

Recently, 2-aminothiazoline-4-carboxylic acid (ATCA), a cyanide (CN) metabolite, has been proposed as a stable diagnostic marker of CN poisoning. In this study, liquid chromatography coupled with electrospray ionization – tandem mass spectrometry was used to quantify ATCA concentrations in human postmortem blood samples, and differences in ATCA concentrations according to age and sex were determined. Both age and sex had significant effects on blood ATCA concentrations. Although ATCA concentrations exhibited an inverted U shape with increasing age in men, in women ATCA concentrations plateaued at around 40–59 years of age. There were significant differences between the sexes in ATCA concentrations for the 20–39 and 40–59 year age groups (P < 0.05 and P < 0.01, respectively). Correlations between ATCA concentrations and carboxyhemoglobin (CO-Hb) saturation were also examined in fire victims. ATCA concentrations increased significantly with increasing CO-Hb saturation (r = 0.382, P < 0.01). In addition, ATCA concentrations were also correlated to CN concentrations (r = 0.309, P < 0.05). The results of our study may provide novel information about the contribution of CN poisoning to the cause of death at fire scenes.     

A GC–MS method for the determination of furanylfentanyl and ocfentanil in whole blood with full validation

Purpose

Fentanyl analogues are popular in recent years among drug addicts and have been related to many overdoses and deaths worldwide. Furanylfentanyl, ocfentanil, acetylfentanyl and butyrfentanyl are among the most common of these drugs. Methods for the determination of furanylfentanyl and ocfentanil by gas chromatography–mass spectrometry (GC–MS) in biological samples do not exist, and therefore, their development would be extremely useful for routine toxicological analysis.

Methods

A GC–MS method was developed and fully validated for the determination of furanylfentanyl and ocfentanil in whole blood. This method was also suitable for the determination of acetylfentanyl and butyrfentanyl. The method included solid-phase extraction after protein precipitation using acetonitrile, and it was applied during the toxicological investigation of forensic cases. Methadone-d₃ was used as internal standard for the quantification of the analytes.

Results

The limit of detection and limit of quantification values were 0.30 and 1.0 ng/mL for furanylfentanyl and ocfentanil and 0.15 and 0.50 ng/mL for acetylfentanyl and butyrfentanyl, respectively. The calibration curves were linear (R²?≥?0.993) from 1.00 to 100 ng/mL for furanylfentanyl and ocfentanil and from 0.50 to 50.0 ng/mL for acetylfentanyl and butyrfentanyl. The recoveries were not lower than 85%, while accuracies and precisions were not greater than 6.0% (% error) and 8.0% (% relative standard deviation), respectively, for all four fentanyl analogues.

Conclusions

The developed method is the first one in the literature for the detection of furanylfentanyl and ocfentanil in biological fluids by GC–MS, and it provides very high sensitivity comparable to that by liquid chromatography–tandem mass spectrometry.

     

High blood mirtazapine concentration in a newborn – A case of suspected postpartum infanticide

We report a sudden death of an infant due to mirtazapine poisoning. A 15-day-old newborn boy was found dead when he was sleeping beside his mother who had suffered from panic disorder for approximately 1 year. After giving birth, she complained of palpitations and shaky hands, and was prescribed mirtazapine. The deceased newborn weighed 3,282 g and his height was 55 cm. There were no autopsy findings related to the death. The mirtazapine concentration as quantitated by liquid chromatography-tandem mass spectrometry analysis was 620 ng/mL in right heart blood, and was approximately 10 times higher than the therapeutic level in adults. Because transfer of mirtazapine into breast milk is low, mirtazapine was likely administered intentionally to the newborn. Based on the newborn’s immature renal, liver, and blood–brain barrier function, the cause of death was attributed to mirtazapine poisoning. Poison-related homicide in the infant is rare. We report the first case of intentional mirtazapine poisoning case in a newborn.     

Estimation of chlorpyrifos distribution in forensic visceral samples and body fluids using LCMS method

The ingestion of the insecticide chlorpyrifos leads to fatal intoxication in suicidal cases, and its distribution can be assessed only after post-mortem. This study attempted to investigate the distribution of chlorpyrifos in forensic visceral tissue samples like stomach, liver, kidney, heart, brain, lung, spleen, muscle and body fluids like blood and urine by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) with the liquid-liquid extraction method. An analysis time of 8.77 min with 1 μl as an injection volume was chosen. The results obtained are analysed using Labsolutions insight LCMS software, considering Multiple Reaction Monitoring (MRM) and Retention Time (RT). Our findings depict that the stomach tissue has very high concentrations of chlorpyrifos, which suggests the possibility of high distribution and aspiration in the stomach tissue. The concentration of chlorpyrifos in stomach tissue ranged from 816 to 901 μg/g in male cadavers and from 443 to 612 μg/g in female cadavers. Among all the tissues, the stomach tissue showed the highest concentration, while the lowest concentration was found in muscle. Metabolic distribution of chlorpyrifos from the stomach contents to the other surrounding organs may occur and hence the concentrations of CPF were also seen in other organs like the liver, lung, kidney, brain, and muscle. These toxicological results from autopsy findings, together with LC-MS/MS indicate that stomach tissue examination gives an accurate profile of insecticide poisoning in forensic samples with acute poisoning.     

Simultaneous quantification of batrachotoxin and epibatidine in plasma by ultra-performance liquid chromatography/tandem mass spectrometry

An ultra-performance liquid chromatography/tandem mass spectrometry (UPLC-MS/MS) method was developed and validated for the simultaneous quantification of batrachotoxin and epibatidine in plasma. Plasma samples were pretreated by liquid-liquid extraction with acetonitrile and methanol. The toxins were separated on a reversed phase C18-column (2.1 mm × 50 mm, 1.7 μm) using a formic acid/acetonitrile gradient elution. Quantification was carried out by mass chromatography with each product ion referenced against midazolam-d₄ as an internal standard (IS). The two toxins and the IS were separated within 2 min. The calibration curves for the two toxins spiked into human plasma showed good linearities in the range from 2.5 to 250 ng/mL. The detection limits were estimated to be 0.5 ng/mL for batrachotoxin and 1 ng/mL for epibatidine with a signal-to-noise ratio of 3:1. Overall recoveries ranged from 69.6% to 98.2%, and no significant matrix effects were observed. The intra- and interday accuracies were 94.7–102.3%, and the precisions were 1.0–10.3%. This method was successfully applied for the quantification of batrachotoxin and epibatidine in rat plasma samples taken after intraperitoneal administration of the toxins. This is the first report to use UPLC-MS/MS to simultaneously quantify batrachotoxin and epibatidine in plasma samples.     

Development for the measurement of serum thiosulfate using LC–MS/MS in forensic diagnosis of H2S poisoning

Thiosulfate measurement is crucial to diagnosis of hydrogen sulfide (H₂S) poisoning in forensic toxicology. Although GC–MS method is currently regarded as a standard thiosulfate measurement, it requires complicated sample preparation prior to analysis. This study presents a simple, rapid, and highly sensitive method for the quantitative analysis of serum thiosulfate by using liquid chromatography–tandem mass spectrometry (LC–MS/MS). This method is based on selected reaction monitoring and has high sensitivity with a lower quantification limit of 0.5 μM. Precision and accuracy of this method meet the basic requirements for quantitative analysis (intra- and inter-day tests have a relative standard deviation of ⩽10.4%; range of analytical recovery is 94.3–102.6%). On the measurements of serum thiosulfate by our developed method, a thiosulfate concentration as 57.5 μM was detected clearly in the H₂S poisoning case comparing to the non poisoning case in which only a trace amount of thiosulfate was observed.