Urinary antimony as a biomarker for differentiating heroin and Brown mixture use

Abstract

Heroin is classified as a Schedule I controlled substance with severe legal consequences for its use, whereas liquid Brown Mixture (BM) is a legally available and commonly prescribed cough remedy in Taiwan. However, numerous cases of BM ingestion have been misinterpreted as heroin use because BM contains opium tincture, leading to similar urine test results. Owing to the rapid metabolism of heroin, detection of the parent drug and its primary metabolite, 6-acetylmorphine, in urine is difficult; thus, interpretation often relies on morphine concentration and the morphine-to-codeine (M/C) ratio, which can be inconclusive. In our BM analysis, the M/C ratio was 3.3 ± 0.4, and among 5094 morphine-positive urine samples, 279 (5.5%) had M/C ratios between 2 and 4, suggesting that a substantial number of individuals could be misidentified. Given the lack of reliable urinary biomarkers to distinguish BM from heroin use, this study aimed to identify BM-specific indicators. BM contains antimony (Sb) potassium tartrate, an emetic agent absent in heroin formulations. Although urinary Sb levels are low, they can be quantified using highly sensitive inductively coupled plasma mass spectrometry (ICP–MS). A pharmacokinetic study involving single and multiple BM doses revealed that urinary Sb increased from ≤0.11 ng/mL pre-dose to 0.4–98.9 ng/mL post-dose. Morphine and codeine were determined using gas chromatography–mass spectrometry (GC–MS), and M/C ratios ranged from 2.0 to 14.9, showing a secondary rise at later stages. These findings suggest that urinary Sb may serve as a useful supplementary biomarker for BM ingestion, thereby enhancing the accuracy of forensic toxicological interpretation and reducing false identification of heroin use.

1. Introduction

Opiate drug testing is widely implemented in many countries as part of workplace safety protocols and legal regulations [1]. In the United States, individuals who test positive for opiates may invoke the “poppy-seed defense,” attributing the presence of morphine and codeine in their urine to the consumption of poppy-seed-containing foods [2–4]. A similar issue has emerged in Taiwan, where individuals who test positive for opiates frequently claim that their results stem from the ingestion of BM, a prescription medication commonly used for cough and cold relief. BM, also known as Licorice Cough Syrup or BM Cough Syrup, commonly contains opium tincture. BM typically contains opium tincture, which contains approximately 0.9–1.1% morphine and codeine, whereas camphorated opium tincture contains approximately 0.045–0.055%. BM ingestion may result in positive urine concentrations (300 ng/mL) of morphine and/or codeine in drug urine testing.

The use of BM as a cough suppressant remains prevalent in Taiwan, whereas it has been extremely rare in other countries for many years. Consequently, there is a significant lack of literature on urine testing related to BM ingestion. According to Liu’s study in 2006 [5], morphine concentrations in urine samples from individuals who ingested BM solution remained below 4000 ng/mL. In these cases, the observed M/C ratios in urine were all less than 3. Nonetheless, the M/C ratio might be insufficient as a discriminative marker between heroin consumption and BM ingestion. Variations in the M/C ratio may arise due to differences between batches of natural opium tincture employed by the manufacturer.

Crucially, ultrarapid metabolism of codeine may alter the M/C ratio [6–8]. Codeine can be partially metabolized to morphine in the liver via the cytochrome P450 2D6 (CYP2D6) enzyme. In cases of codeine-only ingestion, a higher proportion of codeine is typically retained, resulting in an M/C ratio of less than 1. Unfortunately, individuals with abnormal CYP2D6 enzyme activity—due to genetic polymorphisms leading to ultra-rapid or intermediate metabolizer phenotypes—may exhibit enhanced conversion of codeine to morphine [9–11]. This could result in elevated M/C ratios, typically ranging from 1 to 2 following codeine-derived opiate intake, and exceeding 3—potentially reaching up to 6—in cases of BM ingestion.

Since heroin metabolizes rapidly in the body, it is challenging to detect the parent drug and its primary metabolite, 6-acetylmorphine (6-AM), in urine. While the presence of 6-AM serves as a specific marker for heroin use, its short half-life limits its diagnostic value. The absence of 6-AM does not rule out heroin consumption [12–15]. Typically, only morphine and codeine can be detected in urine testing. The interpretation relies on the concentration of morphine in urine and the M/C ratio. Consequently, the precise determination of heroin use continues to be a complex and unresolved issue. Acetylated-thebaine-4-metabolite glucuronide (ATM4G) was proposed as a potential marker for heroin use [16,17]. Unfortunately, its rare detection rate in Taiwanese urine samples, lack of commercial standards availability, has limited its practical application in routine testing [18,19]. Given these limitations, particularly in forensic contexts where heroin is classified as a Schedule I controlled substance and carries significant legal implications, the identification of reliable biomarkers for heroin use remains exceedingly challenging.

BM is a liquid pharmaceutical preparation that contains opium tincture and glycyrrhiza. In addition to these components, BM also includes an emetic agent, Sb potassium tartrate, which is not present in heroin products. Historically, Sb potassium tartrate was used as an emetic, expectorant, and dye mordant. This compound has a sweet, metallic taste and is known to be toxic. Owing to its potential adverse health effects, its medical use has largely been discontinued.

Although the concentration of Sb potassium tartrate in BM is relatively low, the urinary Sb concentration can be quantified using ICP–MS, a highly sensitive analytical technique capable of detecting trace elements at the parts-per-trillion (ppt) level [20–22]. Therefore, urinary Sb has been proposed as a potential biomarker for BM consumption, offering a possible means to differentiate BM ingestion from heroin use. To explore the feasibility of urinary Sb as a reliable biomarker for BM ingestion, this study employed a urinary pharmacokinetic design involving single and multiple doses of BM ingestion. Urine samples were collected at multiple time points and analyzed using GC–MS for morphine and codeine, and ICP–MS for Sb. This approach aimed to establish urinary Sb as a useful supplementary biomarker for supporting BM ingestion in morphine-positive cases, thereby enhancing the accuracy and reliability of forensic toxicological assessments in both clinical and legal contexts.

2. Materials and methods

2.1. Reagents and chemicals

Methanol, acetonitrile, formic acid, glacial acetic acid, hydrochloric acid, sodium hydroxide, and ammonium acetate were all purchased from MERCK (Darmstadt, Germany). Deionized water (DIW) was produced by the Millipore system (Milli-Q), and Triton X-100 was obtained from Sigma–Aldrich (MO, U.S.A.).

Standard solutions of morphine, codeine, morphine-d6, and codeine-d3 isotopic internal standards were purchased from Cerilliant (TX, U.S.A.). The Sb ICP–MS standard was obtained from HIGH-PURITY STANDARD (SC, U.S.A.). The ICP mixed internal standard (ISTD), containing Bi, Ge, In, Li, Lu, Rh, Sc, and Tb at concentrations of 100 μg/mL each, was purchased from Agilent (CA, U.S.A.). BM was manufactured by Center Laboratories, Inc. (Taipei, Taiwan).

2.2. Instruments and equipment

2.2.1. Gas chromatography-mass spectrometry

The gas chromatography-mass spectrometry (GC–MS) system used in this study was an Agilent 7890 GC/5977B MS (Agilent, CA, U.S.A.). The capillary column used for gas chromatography was a 30-m HP-5MS (0.25 mm i.d., 0.25 μm film thickness). The injector temperature was set to 230 °C, with a splitless injection of 1 μL. Helium was used as the carrier gas at a flow rate of 1.3 mL/min. The temperature program started at 150 °C (held for 1 min), ramping up at 20 °C/min to 280 °C (held for 2 min). The GC–MS transfer line temperature was set to 280 °C. The ionization voltage of the mass spectrometer was 70 electron volts, with an ion source temperature of 230 °C and a quadrupole temperature of 150 °C.

For codeine quantification, the target ion was m/z 371, and the qualifier ions were m/z 343 and m/z 243. For the internal standard codeine-d3, the quantification ion was m/z 374, and the qualifier ion was m/z 346. For morphine quantification, the target ion was m/z 429, and the qualifier ions were m/z 401 and m/z 414. For the internal standard morphine-d6, the quantification ion was m/z 435, and the qualifier ion was m/z 420.

2.2.2. Inductively coupled plasma mass spectrometry

The inductively coupled plasma mass spectrometer used in this study was an Agilent 7700 ICP–MS (Agilent, CA, U.S.A.). Urinary Sb (¹²¹Sb) analysis was performed using the high-purity helium collision mode.

2.3. Sample analysis

2.3.1. Morphine and codeine analysis

For the analysis, 1 mL of urine was taken, and 100 μL of internal standard working solution, along with 250 μL of hydrochloric acid (37%) solution, was added. The mixture was vortexed for 10 s and then placed in an oven at 120 °C for an 80-min hydrolysis reaction. After cooling, 250 μL of 10 M NaOH solution was added and vortexed for 10 s, followed by the addition of 1 mL of saturated sodium bicarbonate buffer solution and vortexing again for 10 s. The pH was then adjusted to 8.8–9.0 using 10 M, 1 M NaOH, and 1.0 N, 0.1 N HCl solutions. The sample was centrifuged at 3500 rpm for 5 min.

For solid-phase extraction (SPE), a SPEC-DAU extraction cartridge was placed on the solid-phase extraction device. First, 200 μL of methanol was added and left to stand for 1 min to wet the column. The urine sample was then poured into the SPEC-DAU extraction cartridge (allowing at least 2 min for the urine to pass through the adsorption tube). Subsequently, 500 μL of deionized water was added and drawn through slowly. Then, 500 μL of 0.1 M acetic acid solution was added and drawn through slowly. Afterward, 500 μL of methanol was added and drawn through slowly (for at least 3 min). The SPEC-DAU extraction cartridge was then placed in a glass test tube. Finally, the collected solution was dried using a nitrogen evaporator at 55 °C under nitrogen gas.

For derivatization, 50 μL of MSTFA (N-Methyl-N-(trimethylsilyl)trifluoroacetamide) and 50 μL of ethyl acetate were added and vortexed for 10 s. The mixture was then placed in a 70 °C oven for 20 min. After the reaction, the tube was removed from the oven, cooled, and transferred to a micro-vial for GC–MS analysis.

2.3.2. Urinary creatinine and Sb analysis

Initially, a 1 mL aliquot of each urine sample was collected for creatinine measurement using the enzyme-multiplied immunoassay technique. Urinary Sb concentrations were measured using inductively coupled plasma mass spectrometry. Prior to ICP–MS measurement, 10 mL of thawed urine was centrifuged at 3000 rpm for 5 min at room temperature. From the resulting supernatant, 1 mL was taken and mixed with 1 mL ICP mixed internal standard working solution (5 ng/mL, used as an internal standard) and 8 mL of a 1% (v/v) mixture of nitric acid and hydrochloric acid (in a 1:1 ratio).

The prepared solution was analyzed by ICP–MS under the following conditions: argon plasma, nebulizer, and spray chamber. The target isotope for quantification was ¹²¹Sb, with ¹⁰³Rh as the internal standard. Calibration curves were generated from standard Sb solutions across a range of known concentrations. The limit of detection (LOD) of Sb was 0.01 μg/L, and the limit of quantification (LOQ) was 0.03 μg/L. Seronorm™ Trace Elements Urine L-2 (Seronorm Pharmaca, Billingstad, Norway) was used as the external reference material for quality control, yielding a recovery rate of 90.2%.

2.4. Study design

2.4.1. Statistics of morphine-positive urine samples

Statistics on morphine-positive urine samples were collected from the drug testing laboratory between 2014 and 2023. A total of 5094 subjects were analyzed using the M/C ratio. The study was approved by the Institutional Review Board of Chung Shan Medical University Hospital (IRB No. CS1-22153).

2.4.2. Quantification of BM

The BM samples were diluted 1000-fold prior to analysis. Morphine and codeine concentrations and their ratios were determined using GC–MS. The concentration of Sb was measured using ICP–MS.

2.4.3. Urinary pharmacokinetic study of BM

A urinary pharmacokinetic study was conducted with healthy volunteers receiving BM under three dosing regimens: single dose, multiple low doses, and multiple high doses. Urine samples were collected at multiple time points post-administration. The concentrations of morphine, codeine, and Sb in urine were analyzed using GC–MS and ICP–MS, respectively. Detailed dosage protocols, sampling schedules, and analytical procedures are provided in Supplementary Table S1 (https://doi.org/10.38212/2224-6614.3603).

The study was approved by the Institutional Review Board of Chung Shan Medical University Hospital (IRB No. CS-17069). Participants with a history of drug abuse, pregnancy, or hepatic/renal/cardiac dysfunction were excluded.

2.4.4. Real case study

Real cases claiming to have taken BM, as well as suspected cases provided by the drug testing laboratory, were collected. Morphine, codeine, and Sb concentrations were analyzed using GC–MS and ICP–MS to assess the actual situation.

3. Results and discussion

3.1. Statistical analysis of morphine-positive urine samples

Urine testing for opioids cannot directly detect heroin. Due to the rapid metabolism of 6-AM, which has a short detection window of only 2–8 h in urine, its practical application in routine testing is limited. Currently, urine testing can only detect morphine (a secondary metabolite of heroin) and codeine (a metabolite of acetylcodeine from impurities). To better understand the M/C ratio and distribution of morphine-positive urine test results, we compiled statistics on morphine-positive urine samples with various M/C ratios over the 10 years and compared them with Liu’s study [5]. Because of the batch variations of opium tincture sourced by the manufacturer, the M/C ratio of 2–4 was an important range for BM ingestion.

In our study (2014–2023), a total of 5094 morphine-positive urine specimens were categorized by M/C ratio (Supplementary Table S2 (https://doi.org/10.38212/2224-6614.3603)). Yearly variations in M/C ratio distributions were observed, likely due to the opium’s origin and heroin production process of codeine impurities. Results with M/C ratios above 10 were the most frequent, comprising 30.4% (1547 samples), followed by M/C values 6–8 at 24.6% (1251 samples), and then 4–6, which accounted for 19.3% (984 samples). Ratio at 8–10 made up 18% (918 samples), while M/C ratios between 2 and 4 were less frequent at 5.5% (279 samples). Ratio below 2 constituted only 2.3% (115 samples). Notably, M/C ratio 2–4 significantly decreased in 2021 and 2022, dropping to only 1.1% (5 samples) and 1.7% (4 samples), respectively.

To achieve a more detailed understanding of the M/C distribution, we examined the statistics yearly by single M/C intervals (Supplementary Table S2 (https://doi.org/10.38212/2224-6614.3603)). Samples with M/C ratio 3–4 (197 samples) were 2.4 times more frequent than ratio 2–3 (82 samples). Although samples with M/C ratio 2–4 only represented 5.5%, these findings suggest that as many as 279 individuals may have been misclassified as heroin users. Of these cases, 209 individuals had morphine concentration below 4000 ng/mL.

In Liu’s study [5] of the 90 urine samples of heroin users, 37.8% (34 samples) had morphine concentrations below 4000 ng/mL. The highest proportion (28.9%, or 26 samples) had M/C ratios at 4–6 (Supplementary Table S2 (https://doi.org/10.38212/2224-6614.3603)). When viewed in finer intervals (Supplementary Table S2 (https://doi.org/10.38212/2224-6614.3603)), the most common group was M/C values above 10, comprising 24.4% (22 samples). M/C ratios 3–4 accounted for 13.3% (12 samples). It is important to note that there were no samples with M/C ratios below 3.

3.2. Quantitative analysis of BM

Although the BM medicine bottle label indicates that each mL contains 0.12 mL of glycyrrhiza fluid extract, 0.012 mL of opium tincture, and 0.24 mg of Sb potassium tartrate, it only indicates approximate values for the main ingredients. However, precise quantification of morphine, codeine, and Sb was required for the pharmacokinetics study.

Therefore, GC–MS and ICP–MS were employed in three BM samples (No.1, No.2, No.3), as summarized in Table 1. Sb concentrations were 54.6 μg/mL in No.1, 59.2 μg/mL in No.2, and 57.3 μg/mL in No.3, with a mean Sb concentration of approximately 57.0 ± 2.3 μg/mL.

Table 1.

Quantitative analysis of morphine, codeine, morphine-to-codeine ratio, and antimony concentration in BM samples.

Sample	Morphine (μg/mL)	Codeine (μg/mL)	M/C ratio	Antimony (sb) (μg/mL)
No. 1	138.1	47.0	2.9	54.6
No. 2	161.6	50.7	3.2	59.2
No. 3	168.7	45.5	3.7	57.3
Mean ± SD	156.1 ± 16.0	47.7 ± 2.7	3.3 ± 0.4	57.0 ± 2.3

BM samples were diluted 1000-fold prior to analysis. Morphine and codeine concentrations were determined using gas chromatography–mass spectrometry (GC–MS), whereas antimony (Sb) concentrations were measured using inductively coupled plasma mass spectrometry (ICP–MS). The morphine-to-codeine (M/C) ratio was calculated from measured urinary concentrations. Data are presented as mean ± standard deviation (SD) from three BM samples.

Sample No.1 had a morphine concentration of 138.1 μg/mL and codeine concentration of 47.0 μg/mL, with an M/C ratio of 2.9. Sample No.2 showed a morphine concentration of 161.6 μg/mL and a codeine concentration of 50.7 μg/mL, with an M/C of 3.2. For sample No.3, the morphine concentration was 168.7 μg/mL and the codeine concentration 45.5 μg/mL, yielding an M/C of 3.7. While there was some variation among the three BM samples, the average morphine concentration was 156.1 ± 16.0 μg/mL, the average codeine concentration was 47.7 ± 2.7 μg/mL, and the average M/C was approximately 3.3 ± 0.4. The ratio was higher than the M/C ratio of 2.88 reported in Liu’s study [5], which may be due to batch variations in natural opium tincture sourced by the manufacturer.

3.3. Single dose urinary pharmacokinetics study

To evaluate the urinary concentrations of morphine, codeine, and Sb after a single BM dose, we conducted a pharmacokinetics study (10 mL single dose), as shown in Fig. 1 (a, c, and e) and Supplementary Table S3 (https://doi.org/10.38212/2224-6614.3603).

Fig. 1.

Urinary pharmacokinetic profiles of morphine, codeine, and antimony following single-dose BM administration. Healthy volunteers received a single oral dose of 10 mL BM. Urine samples were collected at multiple time points after administration. Morphine and codeine concentrations were analyzed using GC–MS, whereas antimony (Sb) concentrations were determined using ICP–MS. Panels (a), (c), and (e) show the urinary concentrations of morphine, codeine, Sb, and the morphine-to-codeine (M/C) ratio in Cases 1, 2, and 3, respectively. Panels (b), (d), and (f) show creatinine-corrected urinary concentrations in Cases 1, 2, and 3, respectively. Peak urinary morphine and codeine concentrations were observed approximately 4 h after BM ingestion, whereas Sb showed slower urinary elimination and prolonged detection. The experiment was conducted in three independent subjects, and representative pharmacokinetic profiles are presented.

Prior to BM ingestion, the morphine and codeine concentrations in the urine of the three subjects (Case 1, Case 2, Case 3) were below detection limits. After ingesting 10 mL of BM, morphine concentrations peaked at the 4-h mark with concentrations of 1839 ng/mL, 1373 ng/mL, and 2324 ng/mL for each respective case, followed by a gradual decline. Corresponding codeine concentrations were 674.5 ng/mL, 451.4 ng/mL, and 794.2 ng/mL. M/C ratios were approximately 3.0 ± 0.6 during the non-terminal phase of codeine metabolism. However, as codeine concentrations dropped faster in the terminal phase, M/C values trended upward, surpassing 4 and even reaching 7 in Case 1.

Baseline urinary Sb concentrations in all three subjects were ≤0.11 ng/mL before BM ingestion. After the 10 mL BM dose, Sb was detectable in the urine, peaking between the 4th and 6th hours at 23.60 ng/mL, 33.10 ng/mL, and 29.70 ng/mL, respectively. Compared to morphine and codeine, Sb showed slower urinary elimination kinetics and lower urinary concentrations.

To correct for potential bias, the concentrations of morphine, codeine, and Sb in urine were normalized to creatinine levels, as shown in Fig. 1 (b, d, and f). The results revealed a secondary peak, which may have resulted from redistribution of BM components to various organs, such as the brain, followed by re-entry into systemic circulation.

3.4. Multiple low doses urinary pharmacokinetics study

Typically, patients take multiple low doses of BM per day. Therefore, we conducted a pharmacokinetics study according to the recommended adult dosage (5 mL per dose, 3–4 times daily) and collected urine samples at timed intervals to measure urinary concentrations of morphine, codeine, and Sb, as shown in Fig. 2 (a and c) and Supplementary Table S4 (https://doi.org/10.38212/2224-6614.3603).

Fig. 2.

Urinary pharmacokinetic profiles following multiple low-dose BM administration. Healthy volunteers received repeated oral administration of 5 mL BM every 8 h for a total of five doses. Urine samples were collected at timed intervals throughout the study period. Morphine and codeine concentrations were measured using GC–MS, whereas antimony (Sb) concentrations were analyzed using ICP–MS. Panels (a) and (c) show the urinary concentrations of morphine, codeine, Sb, and the morphine-to-codeine (M/C) ratio in Cases 1 and 2, respectively. Panels (b) and (d) show creatinine-corrected urinary concentrations in Cases 1 and 2, respectively. Repeated BM administration produced cumulative increases in urinary morphine, codeine, and Sb concentrations, with Sb demonstrating relatively slower elimination kinetics. The experiment was conducted in two independent subjects, and representative pharmacokinetic profiles are presented.

Two subjects (Case 1 and Case 2) took 5 mL of BM every 8 h, for a total of five doses. Before BM ingestion, urinary concentrations of morphine and codeine were below detection limits. After the first dose, peak concentrations were observed at 2–4 h, with morphine levels at 1925 ng/mL and 1360 ng/mL, and codeine at 483.9 ng/mL and 394.8 ng/mL, respectively. Due to cumulative dosing effects, concentrations after the second BM dose were higher, with morphine levels reaching 3015 ng/mL and 2275 ng/mL, and codeine levels at 982.2 ng/mL and 715.8 ng/mL. Following the final (fifth) dose, morphine concentrations peaked at 5920 ng/mL and 3509 ng/mL, while codeine reached 1089 ng/mL and 1171 ng/mL, respectively, before gradually declining.

Both subjects had Sb concentrations (≤0.11 ng/mL) prior to BM ingestion. After the first 5 mL dose, Sb concentrations peaked at 6 h, reaching 10.90 ng/mL and 12.10 ng/mL, respectively. Due to cumulative effects, Sb levels after the second BM dose were higher at 20.60 ng/mL and 23.80 ng/mL. After the final (fifth) dose, Sb concentrations peaked at 34.30 ng/mL and 33.40 ng/mL, subsequently decreasing. Sb showed smaller fluctuations in concentration and slower excretion compared to morphine and codeine. To correct for potential bias, the concentrations of morphine, codeine, and Sb in urine were normalized to creatinine concentration, as shown in Fig. 2 (b and d). This showed more pronounced concentration changes, illustrating the impact of cumulative dosing.

3.5. Multiple high dose urinary pharmacokinetics study

To simulate a high-dose intake BM, one subject (Case 1-1, Case 1–2) participated in two trials separated by a three-week interval. The subject took 20 mL of BM every 8 h, totaling four doses, as shown in Fig. 3 (a and c) and Supplementary Table S5 (https://doi.org/10.38212/2224-6614.3603). Prior to BM ingestion, urinary morphine and codeine levels were below detection limits. After the first dose, peak concentrations were reached at 2–4 h, with morphine levels at 2741 ng/mL and 3106 ng/mL, and codeine levels at 921.4 ng/mL and 882.8 ng/mL, respectively. Due to cumulative effects, concentrations after the second BM dose were higher, with morphine at 4460 ng/mL and 4246 ng/mL, and codeine at 1302 ng/mL and 1379 ng/mL. Following the final (fourth) dose, morphine reached maximum levels of 8596 ng/mL and 9517 ng/mL, while codeine peaked at 2543 ng/mL and 3014 ng/mL.

Fig. 3.

Urinary pharmacokinetic profiles following multiple high-dose BM administration. One healthy volunteer participated in two independent trials separated by a 3-week interval and received repeated oral administration of 20 mL BM every 8 h for a total of four doses. Urine samples were collected at multiple time points during the study period. Morphine and codeine concentrations were determined using GC–MS, whereas antimony (Sb) concentrations were quantified using ICP–MS. Panels (a) and (c) show the urinary concentrations of morphine, codeine, Sb, and the morphine-to-codeine (M/C) ratio in Cases 1-1 and 1–2, respectively. Panels (b) and (d) show creatinine-corrected urinary concentrations in Cases 1-1 and 1–2, respectively. Repeated high-dose BM ingestion resulted in marked cumulative increases and prolonged urinary excretion of Sb compared with morphine and codeine. Representative pharmacokinetic profiles from two independent trials are presented.

Before BM ingestion in both trials, urinary Sb concentrations were ≤0.11 ng/mL. After the first dose, Sb peaked at 6 h, with concentrations of 45.10 ng/mL and 47.30 ng/mL. Cumulative effects from subsequent doses led to higher levels after the second dose, at 77.10 ng/mL and 84.10 ng/mL. Following the final (fourth) dose, Sb peaked at 93.10 ng/mL and 98.90 ng/mL, then gradually decreased. Sb concentrations showed smaller fluctuations and slower excretion compared to morphine and codeine. To correct for potential bias, the concentrations of morphine, codeine, and Sb in urine were normalized to creatinine concentration, as shown in Fig. 3 (b and d), further accentuating the concentration fluctuations observed with cumulative dosing.

3.6. Case study and study limitations

3.6.1. Real-case study

To further understand real sample cases, we analyzed several urine samples suspected of BM ingestion. Using GC–MS and ICP–MS, we assessed morphine, codeine, and Sb concentrations in urine samples from four individuals (drivers, employees) claiming BM consumption (A1–A4) and seven suspected cases (B1–B7), as shown in Table 2.

Table 2.

Real-case urinary analysis of morphine, codeine, morphine-to-codeine ratio, and antimony concentrations in individuals claiming or suspected of BM ingestion.

Case	Morphine (ng/mL)	Codeine (ng/mL)	M/C ratio	Antimony (sb) (ng/mL)
Driver A1	448	175	2.6	145.51
Employee A2	901	129	7.0	19.62
Driver A3	1070	369	2.9	2.91
Driver A4	1563	407	3.8	47.76
Suspected B1	305	65	4.7	0.04
Suspected B2	393	82	4.8	0.11
Suspected B3	1061	252	4.2	0.05
Suspected B4	1357	362	3.8	0.04
Suspected B5	13050	4950	2.6	0.05
Suspected B6	16750	5750	2.9	0.11
Suspected B7	27200	8600	3.1	0.07

Urinary morphine and codeine concentrations were analyzed using gas chromatography–mass spectrometry (GC–MS), whereas urinary antimony (Sb) concentrations were determined using inductively coupled plasma mass spectrometry (ICP–MS). Cases A1–A4 were individuals claiming BM ingestion, whereas cases B1–B7 were suspected opiate-positive cases submitted for forensic toxicological evaluation. Elevated urinary Sb concentrations supported BM ingestion.

In the four cases claiming BM ingestion (A1–A4), morphine levels exceeded 300 ng/mL. For Cases A1 and A3, the M/C ratios were below 3, which matched Liu’s study that BM ingestion results in morphine concentration not exceeding 4000 ng/mL and the M/C ratio below 3 [5]. Additionally, the concentrations of Sb were 145.5 and 2.910 ng/mL, confirming BM ingestion. While the M/C ratios were above 3 (7.0 and 3.8) in cases A2 and A4, the Sb concentrations were 19.62 and 47.76 ng/mL, also confirming BM ingestion.

For the suspected cases (B1–B6), Sb levels were ≤0.11 ng/mL across all samples. Among them, Cases B1–B4 had morphine concentrations below 2000 ng/mL and M/C ratios between 3.8 and 4.8. However, very low Sb concentrations suggest no BM ingestion. Cases B5 and B6, with M/C ratios below 3 (2.6 and 2.9), had morphine concentrations exceeding 1.300 × 10⁴ ng/mL, similar to typical high level BM ingestion; however, the very low Sb concentrations suggest no BM ingestion. In Case B7, the morphine concentration was 2.720 × 10⁴ ng/mL, with an M/C ratio 3.1 and the Sb level of 0.070 ng/mL, supporting our hypothesis and ruling out BM ingestion.

3.6.2. Limitations of this study

In urinary pharmacokinetic experiments with single-dose (10 mL), multiple low dose (5 mL), and multiple high dose (20 mL) administrations, we observed the concentration changes of Sb, morphine, codeine, and M/C ratio following BM ingestion. Before controlled pharmacokinetic experiments, urinary Sb, morphine, and codeine levels were significantly low. Nevertheless, our study has several limitations. First, the small number of participants in the study made it impossible to thoroughly examine the relationship between the excretion concentrations of Sb, morphine, and codeine and other factors, such as age, metabolic capacity, as well as the amount, frequency, and long-term use of BM. Second, in cases where BM is used concurrently with heroin, determining whether heroin has been used becomes less certain. This is because BM ingestion would result in detectable urinary Sb concentrations; however, when not in the terminal stage of metabolism, there may be a tendency toward higher M/C ratios, which can exceed 4 and even reach 6.

The last limitation of this study is the inability to distinguish between BM ingestion and individuals’ occupational exposure in Sb-related industries [23]. Taiwan imports several thousand tons of Sb trioxide each year for use in the manufacturing of lead-acid batteries, PET bottles, glass, and other products. Due to occupational exposure, the workers may have elevated background levels of Sb in their urine. According to the Occupational Safety and Health Administration’s guidelines on Sb exposure toxicity, urinary Sb levels in the general population are typically below 1 ng/mL [24]. Among workers in Sb-related industries, workers average around 100–300 ng/mL, while levels among acute exposure, urinary Sb levels can exceed 2000 ng/mL. Urinary Sb concentrations in Sb industry workers may overlap with those observed in individuals with high-dose or frequent BM ingestion.

Although urinary Sb concentrations can suggest BM ingestion, both BM and heroin are used concurrently, and occupational exposure to Sb could result in elevated background Sb levels, potentially leading to a false inference. In such cases, differentiating BM ingestion from heroin use would ultimately depend on hair analysis to detect 6-AM, a definitive marker of heroin consumption [25].

4. Conclusion

This study demonstrates that urinary Sb, measured by ICP–MS, may serve as a useful supplementary biomarker for identifying BM ingestion in morphine-positive urine samples. Urinary Sb increased from baseline (≤0.11 ng/mL) to 0.4–98.9 ng/mL following BM ingestion, and exhibited a longer detection window than morphine and codeine, supporting its forensic utility.

When interpreted together with morphine concentration and the M/C ratio, urinary Sb may improve differentiation between BM ingestion and heroin exposure, thereby enhancing the accuracy of forensic toxicological assessment.

However, several limitations should be considered. The small sample size may limit generalizability, occupational Sb exposure may confound interpretation, and concurrent BM and heroin use remains difficult to distinguish. Further studies are needed to validate these findings.

In conclusion, urinary Sb is a promising supplementary biomarker for BM ingestion, but should be interpreted cautiously in conjunction with other indicators.

Funding Statement

The authors would like to thank the Ministry of Science and Technology and the Food and Drug Administration, Department of Health, Taiwan, for financial support of this research (MOST 105-2113-M-040-004 and MOHW 108-FDA-B-114-000732)