Investigators purchased 46 chewable or unit-dose products labeled as cat's claw. The purchases came from online retailers and three gas stations in Mississippi during September 2026, and the set represented 17 distinct brands. Labels did not declare kratom alkaloids or related synthetic derivatives on any product, yet every sample yielded at least one kratom-associated component when analyzed. The study team reported the sampling was cross-sectional; the manuscript did not report batch-to-batch variability beyond these items.
The authors extracted products with methanol and ran ultra-high-performance liquid chromatography with photodiode array detection, comparing chromatograms to a methanol extract of botanically verified Uncaria tomentosa bark. They then used liquid chromatography quadrupole time-of-flight mass spectrometry to identify additional compounds and quantified identified analytes against authentic reference standards. The paper provided exact (Clopper–Pearson) 95% confidence intervals for proportions. The source did not report any bioassay or clinical testing of effects in humans.
Chemical data indicated high prevalence of kratom-derived alkaloids despite absence from labels. Mitragynine was quantified in 41 of 46 products (89%; 95% CI, 76% to 96%). 7-hydroxymitragynine (7-OH) appeared in 35 products (76%; 95% CI, 61% to 87%), with measured per-unit amounts spanning a wide range—from 0.01 mg up to 95.4 mg per unit. MGM-15 (dihydro-7-hydroxymitragynine) was present in 24 products (52%; 95% CI, 37% to 67%), and mitragynine pseudoindoxyl (MP) was identified in 22 products (48%; 95% CI, 33% to 63%). Combining MP and MGM-15, 39 products (85%; 95% CI, 71% to 94%) contained one or both compounds; the authors noted that MP, MGM-15, and MGM-16 were temporarily scheduled by the Drug Enforcement Administration as schedule I agents in August 2026.
The analysis also detected tianeptine—an antidepressant not approved for use in the United States—in five products (11%; 95% CI, 4% to 24%). Quantified amounts ranged from 27.9 mg to 40.2 mg per unit. The source paper did not provide clinical outcome data linked to these specific doses; it reported detection and quantification only. The presence of tianeptine in over-the-counter–market items raises safety questions because consumers cannot identify its inclusion from the label.
Eight products carried explicit claims such as “NO KRATOM and No 7-OH” or “NO 7-OH.” The laboratory findings contradicted those statements: these products nonetheless contained mitragynine, 7-OH, and 3-dehydromitragynine. Labels therefore offered no reliable guide to chemical exposure. The authors emphasized that consumers and clinicians cannot infer presence or absence of opioid-active alkaloids from product branding or marketing claims alone.
The study team recommended that clinicians assessing unexplained opioid toxicity after use of a product labeled as cat's claw consider exposure to kratom-derived opioids and tianeptine. Given the quantified ranges—particularly the high end of 7-OH per unit—the potential for opioid effects, toxicity, or interaction with other central nervous system depressants exists. The source did not offer treatment protocols or case reports, and it explicitly stated the preprint had not undergone peer review and should not be used as a standalone clinical practice guide.
Authors noted that regulatory action focused only on scheduling individual compounds may be insufficient; they suggested product testing and label enforcement as additional strategies. The manuscript reported that all study data are available on reasonable request. Limitations documented in the source include the cross-sectional sampling design and the geographic concentration of purchases (Mississippi), which may limit generalizability. The paper did not report longitudinal surveillance, adverse-event linkage to specific product lots, or consumer use patterns beyond product labeling. The preprint also declared no competing interests and acknowledged funding support from the United States Food and Drug Administration (5U01FD004246).
Clinicians and regulators seeking more detail should consult the full preprint for analytical methods and exact confidence intervals; the authors provided supplementary material and data-access information with the report. The article was posted October 22, 2026, and the authors cautioned that the findings have not been certified by peer review.