LC-MS/MS Analysis of THC Isomers & Metabolites in Whole Blood and Urine
Abstract
As the complexity of THC metabolite testing increases, efficient methods are needed to resolve isomeric compounds for accurate reporting. In this work, an approach to test for Δ8-THC and Δ9-THC and metabolites was developed for whole blood and urine matrices.
Introduction
The testing of blood and urine for tetrahydrocannabinol (Δ9-THC) consumption has been around for decades. In the body, Δ9-THC is metabolized into 11-nor-9-carboxy-Δ9-THC (Δ9-THC-COOH), which can then be detected in blood and urine. Historically, Δ9-THC and Δ9-THC-COOH were the only two compounds routinely monitored to determine THC use [1]. As more isomers of Δ9-THC become available on the market, testing becomes more complicated, and novel methods are needed to achieve isomeric resolution. One such isomer, Δ8-THC, emerged on the market after the passing of the 2018 Farm Bill, which appeared to legalize the isomer as a hemp-derived product. This compound forms its own metabolite, 11-nor-9-carboxy-Δ8-THC (Δ8-THC-COOH), which must be resolved from Δ9-THC-COOH. The resolution of these two metabolites is key in reporting accurate specimen findings, and poor resolution can result in invalid data, especially when one isomer is present in much greater abundance.
In addition, clinical and toxicology labs are interested in the ability to test blood samples after an incident to determine if the individual was under the influence of cannabis. However, this information cannot be gained as easily as other types of routine, time-of-incident testing. Once administered, the concentration of Δ9-THC in blood decreases, but metabolism in the body is nonlinear, thus making assumption on time of exposure difficult [1]. Additionally, some studies have detected THC after 33 days of last use in blood and Δ9-THC-COOH up to approximately 30 days of last use in urine of heavy users, indicating that these compounds are not viable candidates for determining if a suspect is currently under the influence when a sample is collected [1,2]. When Δ9-THC is introduced into the body, it is first metabolized by oxidation to 11-hydroxy-Δ9-tetrahydrocannabinol (11-OH-Δ9-THC), which is a short-lived intermediate, before being further metabolized to Δ9-THC-COOH. While the presence of 11-OH-Δ9-THC alone is not enough to definitively determine the exact time of cannabis use/exposure, by monitoring for this metabolite, labs may still gain insight into whether an individual has recently ingested cannabis.
In this work, a method was developed to monitor both Δ8-THC and Δ9-THC carboxy metabolites (Δ9-THC-COOH/Δ8-THC-COOH) in urine. The method was developed to ensure ample resolution between the two isomers is achieved, so the quantitative data is rugged and robust, even when one of the isomers is present in very high concentrations. An additional method was developed for whole blood to monitor both the parent compounds (Δ8-THC/Δ9-THC); hydroxy metabolites (11-OH-Δ9-THC/11-OH-Δ8-THC); and carboxy metabolites (Δ9-THC-COOH/Δ8-THC-COOH).
Experimental
Chromatographic Method
Urine
The method conditions were as follows for detecting Δ8-THC-COOH and Δ9-THC-COOH in urine.
Figure 1: Fifty Nanograms per Milliliter of Δ8-THC-COOH and Δ9-THC-COOH Metabolites in Urine with Method Conditions
| Peaks | tR (min) | Precursor | Product 1 | Product 2 | |
|---|---|---|---|---|---|
| 1. | Δ8-THC-COOH | 4.44 | 345.1 | 327.0 | 299.2 |
| 2. | Δ9-THC-COOH | 5.12 | 345.1 | 327.0 | 299.2 |
| Column | Raptor FluoroPhenyl (cat.# 9319A1E) | ||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dimensions: | 100 mm x 3 mm ID | ||||||||||||||||||||||||||||
| Particle Size: | 2.7 µm | ||||||||||||||||||||||||||||
| Pore Size: | 90 Å | ||||||||||||||||||||||||||||
| Guard Column: | Raptor FluoroPhenyl EXP guard column cartridge 5 mm, 3 mm ID, 2.7 µm (cat.# 9319A0253) | ||||||||||||||||||||||||||||
| Temp.: | 40 °C | ||||||||||||||||||||||||||||
| Standard/Sample | |||||||||||||||||||||||||||||
| (±)11-nor-9-carboxy-Δ-9-THC (Δ9-THC-COOH) (cat.# 34068) | |||||||||||||||||||||||||||||
| Other compounds obtained separately. | |||||||||||||||||||||||||||||
| Diluent: | 40:60 Water:Methanol, both with 0.1% formic acid (v/v) | ||||||||||||||||||||||||||||
| Conc.: | 50 ng/mL | ||||||||||||||||||||||||||||
| Inj. Vol.: | 1 µL | ||||||||||||||||||||||||||||
| Mobile Phase | |||||||||||||||||||||||||||||
| A: | Water, 0.1% formic acid | ||||||||||||||||||||||||||||
| B: | Methanol, 0.1% formic acid | ||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||
| Max Pressure: | 385 bar |
| Detector | Shimadzu 8045 MS/MS |
|---|---|
| Ion Mode: | ESI+ |
| Instrument | Shimadzu Nexera X2 |
| Sample Preparation | Blank urine was spiked across the calibration range. Five hundred microliters of sample was added to a glass test tube. Fifty microliters of internal standard was added to each sample and vortexed. Prior to extraction, alkaline hydrolysis was performed on the samples by adding 40 μL of 10 N NaOH to each sample. Samples were capped, vortexed, and heated at 60 °C for 20 minutes. After cooling, 25 μL of glacial acetic acid was added to neutralize the pH of the samples. After hydrolysis, the samples were extracted by LLE. Five hundred microliters of HPLC grade water was added to each tube and vortexed. One hundred microliters of 10% acetic acid was added to each tube and vortexed. Two-and-a-half milliliters of 80:20 hexanes:ethyl acetate was added to each tube, capped, and vortexed until visibly combined. Samples were centrifuged at 2800 rpm for 15 minutes or until the two layers had completely separated. The supernatant was transferred to a clean test tube. Samples were dried down under nitrogen. Samples were reconstituted in 100 μL of 40:60 water:methanol, both with 0.1% formic acid, and vortexed. Samples were transferred to 2 mL short-cap, screw-thread vials (cat.# 21143) with glass vial inserts (cat.# 21776) and capped with short-cap, screw-thread closures (cat.# 24498). |
| Notes | The column was stored in 100% acetonitrile when not in use. |
Blood
The method conditions were as follows for detecting Δ9-THC/Δ8-THC, Δ9-THC-COOH/Δ8-THC-COOH, and 11-OH-Δ9-THC/11-OH-Δ8-THC in whole blood.
Figure 2: Fifty Nanograms per Milliliter of Compounds Δ8-THC, Δ9-THC, 11-OH-Δ9-THC, 11-OH-Δ8-THC, and 250 ng/mL of Δ8-THC-COOH and Δ9-THC-COOH in Whole Blood with Method Conditions
| Peaks | tR (min) | Conc. (ng/mL) | Precursor | Product 1 | Product 2 | |
|---|---|---|---|---|---|---|
| 1. | 11-OH-Δ8-THC | 4.66 | 50 | 331.0 | 313.0 | 201.1 |
| 2. | 11-OH-Δ9-THC | 5.00 | 50 | 331.0 | 313.0 | 201.1 |
| 3. | Δ8-THC-COOH | 5.04 | 250 | 345.1 | 327.0 | 299.2 |
| 4. | Δ9-THC-COOH | 5.85 | 250 | 345.1 | 327.0 | 299.2 |
| 5. | Δ8-THC | 10.88 | 50 | 315.0 | 193.0 | 123.2 |
| 6. | Δ9-THC | 11.26 | 50 | 315.0 | 193.0 | 123.2 |
| Column | Raptor FluoroPhenyl (cat.# 9319A1E) | ||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dimensions: | 100 mm x 3 mm ID | ||||||||||||||||||||||||||||||||||||
| Particle Size: | 2.7 µm | ||||||||||||||||||||||||||||||||||||
| Pore Size: | 90 Å | ||||||||||||||||||||||||||||||||||||
| Guard Column: | Raptor FluoroPhenyl EXP guard column cartridge 5 mm, 3 mm ID, 2.7 µm (cat.# 9319A0253) | ||||||||||||||||||||||||||||||||||||
| Temp.: | 40 °C | ||||||||||||||||||||||||||||||||||||
| Standard/Sample | |||||||||||||||||||||||||||||||||||||
| Δ8-Tetrahydrocannabinol (Δ8-THC) (cat.# 34090) | |||||||||||||||||||||||||||||||||||||
| Δ9-Tetrahydrocannabinol (Δ9-THC) (cat.# 34067) | |||||||||||||||||||||||||||||||||||||
| (±)11-nor-9-carboxy-Δ-9-THC (Δ9-THC-COOH) (cat.# 34068) | |||||||||||||||||||||||||||||||||||||
| Other compounds obtained separately. | |||||||||||||||||||||||||||||||||||||
| Diluent: | 40:60 Water:methanol, both with 0.1% formic acid (v/v) | ||||||||||||||||||||||||||||||||||||
| Inj. Vol.: | 5 µL | ||||||||||||||||||||||||||||||||||||
| Mobile Phase | |||||||||||||||||||||||||||||||||||||
| A: | Water, 0.1% formic acid | ||||||||||||||||||||||||||||||||||||
| B: | Methanol, 0.1% formic acid | ||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||
| Max Pressure: | 390 bar |
| Detector | Shimadzu 8045 MS/MS |
|---|---|
| Ion Mode: | ESI+ |
| Instrument | Shimadzu Nexera X2 |
| Sample Preparation | Blank blood was spiked across the calibration range. Five hundred microliters of sample was added to a glass test tube. Fifty microliters of internal standard was added to each sample and vortexed. Samples were extracted by LLE. Five hundred microliters of HPLC grade water was added to each tube and vortexed. One hundred microliters of 10% acetic acid was added to each tube and vortexed. Two-and-a-half milliliters of 80:20 hexanes:ethyl acetate was added to each tube, capped, and vortexed until visibly combined. Samples were centrifuged at 2800 rpm for 15 minutes or until the two layers had completely separated. The supernatant was transferred to a clean test tube. Samples were dried down under nitrogen. Samples were reconstituted in 100 μL of 40:60 water:methanol, both with 0.1% formic acid, and vortexed. Samples were transferred to 2 mL short-cap, screw-thread vials (cat.# 21143) with glass vial inserts (cat.# 21776) and capped with short-cap, screw-thread closures (cat.# 24498). |
| Notes | The column was stored in 100% acetonitrile when not in use. |
Sample Preparation
Controls
Calibrators, QC standards, and internal standards were prepared in methanol.
Urine-Calibrators and QC Sample Preparation
Blank human urine was spiked across the calibration range. Five-hundred microliters of sample was added to a glass test tube. Fifty microliters of internal standard (1000 ng/mL of Δ9-THC-COOH-D9) was added to each sample and vortexed.
Urine-Hydrolysis QC Sample Preparation
A glucuronide positive control (Δ9-THC-COOH-glucuronide) was prepared at a concentration of 750 ng/mL (when liberated) in blank urine. Five-hundred microliters of the sample was added to a glass test tube. Fifty microliters of internal standard (1000 ng/mL of Δ9-THC-COOH-D9) was added to the sample and vortexed.
Urine-Patient Sample Preparation
Five-hundred microliters of patient sample was added to a glass test tube. Fifty microliters of internal standard (1000 ng/mL of Δ9-THC-COOH-D9) was added to each sample and vortexed.
Urine-Alkaline Hydrolysis Procedure
Since glucuronidation occurs during the metabolism of THC, urine specimens need to undergo an alkaline hydrolysis step prior to extraction. Forty microliters of 10 N NaOH was added to 500 µL of sample. The samples were capped, vortexed, and incubated at 60 °C for 20 minutes. Samples were removed and allowed to cool. After cooling, 25 µL of glacial acetic acid was added to the samples to neutralize the pH. The hydrolyzed samples were then transferred to 12 mL glass test tubes and extracted using the following LLE procedure.
Urine-LLE Procedure
Five-hundred microliters of HPLC grade water was added to each tube and vortexed. One-hundred microliters of 10% acetic acid was added to each tube and vortexed. Two-and-a-half milliliters of 80:20 hexanes:ethyl acetate was added to each tube, capped, and vortexed until visibly combined. Samples were centrifuged at 2800 rpm for 15 minutes or until the two layers had completely separated. The supernatant was pulled off the top of each sample and transferred to a clean test tube. Samples were dried down under nitrogen. Samples were reconstituted in 100 µL of 40:60 water:methanol, both with 0.1% formic acid, and vortexed. Samples were transferred to an LC vial with an insert prior to LC-MS/MS analysis.
Blood-Calibrators and QC Sample Preparation
Blank human blood (containing K2∙EDTA) was spiked across the calibration range. Five-hundred microliters of the sample was added to a glass test tube. 50 µL of internal standard (1000 ng/mL of Δ9-THC-D3, 11-OH-Δ9-THC-D3, and Δ9-THC-COOH-D3) was added to each tube and vortexed.
Blood-Patient Sample Preparation
Five-hundred microliterspatient sample was added to a glass test tube. Fifty microliters of internal standard (1000 ng/mL of Δ9-THC-D3, 11-OH-Δ9-THC-D3, and Δ9-THC-COOH-D3) was added to each tube and vortexed.
Blood-LLE Procedure
Five-hundred microliters of HPLC grade water was added to each tube and vortexed. One-hundred microliters of 10% acetic acid was added to each tube and vortexed. Two-and-a-half milliliters of 80:20 hexanes:ethyl acetate was added to each tube, capped, and vortexed until visibly combined. Samples were centrifuged at 2800 rpm for 15 minutes or until the two layers had completely separated. The supernatant was pulled off the top of each sample and transferred to a clean test tube. Samples were dried down under nitrogen and reconstituted in 100 µL of 40:60 water: methanol, both with 0.1% formic acid, and vortexed. Samples were transferred to an LC vial with an insert prior to LC-MS/MS analysis.
Results and Discussion
Urine
Chromatographic Performance
The two isomers were well separated in a 6.5-minute isocratic gradient (8.5-minute total analysis time) on a Raptor FluoroPhenyl 2.7 µm, 100 x 3.0 mm column. The resolution achieved by this method was sufficient to prevent quantitative interferences between the two analytes, even at extreme isomer ratios sometimes observed in urine samples (Figures 3 and 4).
Figure 3: Δ8-THC-COOH and Δ9-THC-COOH are still well resolved from each other even at extreme isomer ratios (1 ng/mL Δ8-THC-COOH: 1000 ng/mL Δ9-THC-COOH).
Figure 4: Δ8-THC-COOH and Δ9-THC-COOH are still well resolved from each other even at extreme isomer ratios (1 ng/mL Δ9-THC-COOH: 1000 ng/mL Δ8-THC-COOH).
Linearity
Linearity was demonstrated using a 1/x weighted linear regression, and all analytes showed acceptable r2 values of ≥0.99. Calibration was performed from 5-1000 ng/mL for Δ8-THC-COOH and Δ9-THC-COOH. The calibration range encompasses typical concentration levels for these analytes in urine specimens. To identify any potential matrix interferences, three blank urine samples were prepped and analyzed for any signals that were not attributed to the analytes of interest. No matrix interferences were observed. Carryover was assessed by injecting a methanol blank immediately after the highest calibrator (1000 ng/mL), and no carryover was observed.
Precision and Accuracy
Precision and accuracy were assessed at four different concentrations (LLOQ, Low QC, Medium QC, High QC) and evaluated within a day and as an average of three days (N=9). Method accuracy is defined as the percentage of the measured concentration relative to the known concentration. The interday precision of the method is presented as relative standard deviation (%RSD). These results demonstrate that the method is accurate and precise over the range studied (Tables I and II).
Table I: Interday Method Precision Results (%RSD) in Urine
| Analyte | LLOQ (5 ng/mL) | Low QC (25 ng/mL) | Med QC (250 ng/mL) | High QC (750 ng/mL) |
| Δ8-THC-COOH | 2.14% | 3.81% | 1.87% | 1.76% |
| Δ9-THC-COOH | 1.04% | 2.13% | 1.56% | 0.79% |
Table II: Method Accuracy Results (%) in Urine
| Analyte | LLOQ (5 ng/mL) | Low QC (25 ng/mL) | Med QC (250 ng/mL) | High QC (750 ng/mL) |
| Δ8-THC-COOH | 83.0% | 94.7% | 93.6% | 96.3% |
| Δ9-THC-COOH | 80.6% | 98.1% | 90.5% | 95.2% |
Blood
Chromatographic Performance
The three sets of isomers (6 total analytes) were well separated in a 13-minute gradient (16-minute total analysis time) on a Raptor FluoroPhenyl 2.7 µm, 100 x 3.0 mm column (cat.# 9319A1E).
Linearity
Linearity was demonstrated using a 1/x weighted linear regression, and all analytes showed acceptable r2 values of ≥0.99 for all analytes. Calibration was performed from 0.5-100 ng/mL for 11-OH-Δ8-THC, 11-OH-Δ9-THC, Δ8-THC, and Δ9-THC and from 2.5-500 ng/mL for Δ8-THC-COOH and Δ9-THC-COOH. The calibration range encompasses typical concentration levels for these analytes in blood specimens. To identify any potential matrix interferences, three blank blood samples were prepped and analyzed for any signals that were not attributed to the analytes of interest. No matrix interferences were observed. Carryover was assessed by injecting a methanol blank immediately after the highest calibrator (100/500 ng/mL), and no carryover was observed.
Precision and Accuracy
Precision and accuracy were assessed at four different concentrations (LLOQ, Low QC, Medium QC, High QC) and evaluated within a day and as an average of three days (N=9). The interday precision of the method is presented as relative standard deviation (%RSD). These results demonstrate that the method is accurate and precise over the range studied (Tables III and IV).
Table III: Interday Method Precision Results (%RSD) in Whole Blood
| Analyte | LLOQ (0.5/2.5 ng/mL) | Low QC (5/25 ng/mL) | Med QC (45/225 ng/mL) | High QC (75/375 ng/mL) |
| 11-OH-Δ8-THC | 6.73% | 6.12% | 3.65% | 3.44% |
| 11-OH-Δ9-THC | 9.90% | 4.13% | 3.61% | 3.37% |
| Δ8-THC-COOH* | 4.63% | 4.15% | 2.20% | 3.33% |
| Δ9-THC-COOH* | 4.04% | 2.13% | 1.78% | 2.74% |
| Δ8-THC | 11.4% | 7.77% | 8.89% | 6.95% |
| Δ9-THC | 9.32% | 4.00% | 1.56% | 2.23% |
*Note: Target QC concentrations were 5X higher for Δ-8/9-THC-COOH.
Table IV: Method Accuracy Results (%) in Whole Blood
| Analyte | LLOQ (0.5/2.5 ng/mL) | Low QC (5/25 ng/mL) | Med QC (45/225 ng/mL) | High QC (75/375 ng/mL) |
| 11-OH-Δ8-THC | 86.2% | 92.6% | 97.3% | 96.7% |
| 11-OH-Δ9-THC | 91.1% | 94.8% | 97.2% | 97.6% |
| Δ8-THC-COOH* | 96.8% | 96.0% | 97.1% | 97.2% |
| Δ9-THC-COOH* | 93.6% | 95.9% | 98.4% | 97.9% |
| Δ8-THC | 90.9% | 93.4% | 92.4% | 94.4% |
| Δ9-THC | 91.1% | 93.1% | 98.8% | 98.2% |
*Note: Target QC concentrations were 5X higher for Δ-8/9-THC-COOH.
Cross-Analyte Interference
Potential interference from 12 other commonly encountered isobaric or structurally similar cannabinoids were also investigated. A sample containing cannabidivarin (CBDV); cannabidiol (CBD); cannabigerol (CBG); tetrahydrocannabivarin (THCV); exo-tetrahydrocannabinol (exo-THC); 9(S)-hexahydrocannabinol (9(S)-HHC); cannabicyclol (CBL); 9(R)-hexahydrocannabinol (9(R)-HHC) cannabinol (CBN); 9(S)-Δ6a,10a-THC (Δ10-THC); cannabichromene (CBC); and Tetrahydrocannabinolic Acid A (THCA-A) was made and analyzed by the developed method. It was determined that all compounds monitored were well resolved from the analytes of interest. CBL elutes in the same MRM window as Δ9-THC but has a notable difference in retention time (11.16 min vs. 11.40 min) so no misidentification should occur. Based on these results, no cross-analyte interference is expected (Figure 5).
Figure 5: Separation of Δ8/9-THC Isomers and Metabolites from 12 Cannabinoids
| Peaks | tR (min) | Precursor | Product 1 | Product 2 | |
|---|---|---|---|---|---|
| 1. | CBDV | 3.22 | 286.9 | 165.1 | 122.9 |
| 2. | 11-OH-Δ8-THC | 4.60 | 331.0 | 313.0 | 201.1 |
| 3. | 11-OH-Δ8-THC | 4.94 | 331.0 | 313.0 | 201.1 |
| 4. | Δ8-THC-COOH | 4.98 | 345.1 | 327.0 | 299.2 |
| 5. | Δ8-THC-COOH | 5.77 | 345.1 | 327.0 | 299.2 |
| 6. | CBD | 6.20 | 315.0 | 193.0 | 123.2 |
| 7. | THCV | 6.85 | 286.9 | 165.1 | 122.9 |
| 8. | CBG | 6.87 | 317.0 | 193.1 | 123.0 |
| 9. | exo-THC | 9.88 | 315.0 | 193.0 | 123.2 |
| Peaks | tR (min) | Precursor | Product 1 | Product 2 | |
|---|---|---|---|---|---|
| 10. | Δ8-THC | 10.80 | 315.0 | 193.0 | 123.2 |
| 11. | 9(S)-HHC | 11.05 | 317.0 | 193.0 | 123.1 |
| 12. | Δ9-THC | 11.16 | 315.0 | 193.0 | 123.2 |
| 13. | CBL | 11.40 | 315.0 | 193.0 | 123.2 |
| 14. | 9(R)-HHC | 11.85 | 317.0 | 193.0 | 123.1 |
| 15. | CBN | 12.82 | 311.1 | 223.0 | 293.2 |
| 16. | Δ10-THC | 13.05 | 315.0 | 193.0 | 259.2 |
| 17. | CBC | 13.96 | 315.0 | 193.0 | 123.2 |
| 18. | THCA-A | 14.08 | 359.3 | 341.2 | 219.0 |
Benefit of Inert Columns
HPLC columns with inert hardware can offer many benefits over traditional stainless steel. At Restek, these columns have a premium inert coating applied to the stainless-steel surface of the column, which helps to reduce nonspecific binding of chelating analytes. Acidic compounds may be prone to chelation, which can result in poor peak shapes and reduced sensitivity. Analysis using the method described in this work was performed on a Raptor FluoroPhenyl column and a Raptor Inert FluoroPhenyl column to determine if the inert column had an impact on the performance of any of the analytes (Figure 6). While the parent compounds showed little difference between the stainless steel and inert columns, the hydroxy and carboxy compounds showed an increase in peak area and peak height when analyzed on an inert column. For laboratories looking to improve sensitivities for hydroxy and carboxy analytes, an inert column can be beneficial over traditional stainless-steel.
Figure 6: Comparison of Δ9-THC, Δ8-THC, and Metabolites on Raptor FluoroPhenyl Columns vs. Raptor Inert FluoroPhenyl Columns
Conclusion
Two methods were developed for reliable and accurate THC isomer testing in human urine and whole blood. A straightforward liquid-liquid extraction method was implemented for both whole blood and urine matrices [3,4]. An LC-MS/MS method was developed to separate carboxy metabolites in urine with full resolution between isomers for accurate reporting. A second method was developed to monitor parent compounds, hydroxy, and carboxy metabolites in whole blood matrix for time of incident intoxication testing. These methods were demonstrated to be fast, rugged, and sensitive in order to meet reporting requirements and are suitable for clinical and toxicology labs that are interested in implementing accurate reporting of intoxication at time of incident and/or differentiating consumption between Δ8-THC and Δ9-THC.
References
- E.L. Karschner, M.J. Swortwood-Gates, M.A. Huestis, Identifying and Quantifying Cannabinoids in Biological Matrices in the Medical and Legal Cannabis Era. Clinical Chemistry 66 (7) (2020) 888-914. https://academic.oup.com/clinchem/article/66/7/888/5867829?login=false
- Mayo Clinic Laboratories, Marijuana: Comprehensive urine drug testing to confirm and monitor use https://www.mayocliniclabs.com/test-catalog/drug-book/specific-drug-groups/marijuana (Accessed April 1, 2025).
- Virginia Department of Forensic Science, Toxicology Procedures Manual, 2024, https://dfs.virginia.gov/wp-content/uploads/220-D100%20Toxicology%20Procedures%20Manual-2816-67ec03f17b822.pdf
- N.B. Tiscione, R . Miller, X. Shan, J. Sprague, D.T. Yeatman. An efficient, robust method for the determination of cannabinoids in whole blood by LC-MS-MS. J Anal. Toxicol., 40 (2016) (8) 639-648. https://doi.org/10.1093/jat/bkw063
This method has been developed for research use only; it is not suitable for use in diagnostic procedures without further evaluation.