evipedia.ai
← Back

Gou-teng for Health & Longevity

Evidence Review created on 08/25/2026 using AI4L / Opus 5

Also known as: Uncaria rhynchophylla, Uncariae Ramulus cum Uncis, Uncaria Hook, Gouteng, Gou Teng, Chotoko, Gambir Vine Hook, Diao Gou Teng

Motivation

Gou-teng is the hooked stem of a climbing vine, Uncaria rhynchophylla, that has been used in Chinese herbal practice for roughly fifteen centuries. It is the herb Chinese and Japanese practitioners reach for when the presenting picture is agitation, dizziness, tremor, high blood pressure, or convulsions — a cluster traditionally described as internal wind. Its interest to modern researchers rests on a group of plant compounds that relax blood vessels and act on brain receptors governing arousal and aggression.

The herb rarely travels alone. It is the defining ingredient of three widely dispensed formulas: one prescribed across East Asia for high blood pressure, and two that Japanese physicians prescribe under national health insurance for the irritability, hallucinations, and aggression that accompany dementia. That prescribing base has produced placebo-controlled trials and adverse-event registries that most single herbs never accumulate, though almost all of them test the formula rather than the herb by itself.

This review examines what is known about Gou-teng: how its constituents act, what controlled human studies of its formulas have measured, what harms have been recorded, how it is dosed and sourced, and where the evidence stops short of the claims made for it.

Benefits - Risks - Protocol - Conclusion

High-level overviews of Gou-teng’s chemistry, pharmacology, and clinical record, selected for depth on the herb itself rather than on the formulas that contain it.

Note on priority experts: No content was found from Rhonda Patrick, Peter Attia, Andrew Huberman, Chris Kresser, Life Extension Magazine, or Lifespan.io. Gou-teng has essentially no presence in the English-language longevity media, which cover the unrelated species Uncaria tomentosa instead.

Grokipedia

  • Uncaria rhynchophylla

    Covers the botany, East Asian distribution, and traditional and pharmacological uses of the vine, and is the most useful single page for placing Gou-teng within the wider Uncaria genus.

Examine

  • Gou-teng

    Gives the only independent dose estimate available, extrapolated from rodent work, and flags plainly that all oral-ingestion evidence remains preclinical. Its Research Breakdown has been archived rather than updated.

ConsumerLab

No ConsumerLab article exists for Gou-teng. ConsumerLab has not tested Uncaria rhynchophylla products; its only Uncaria coverage concerns the South American species Uncaria tomentosa, sold as cat’s claw, which is chemically distinct.

Systematic Reviews

Pooled analyses of randomised controlled trials (studies in which participants are assigned to treatment or control by chance) of the three multi-herb formulas built around Gou-teng.

Coverage note: The systematic-review literature represents the claimed benefits but not the principal risk. No systematic review or meta-analysis examines liver injury, low potassium, or blood-pressure overshoot with Gou-teng or its formulas as a primary question; harms evidence in this review therefore comes from national adverse-event databases, retrospective cohorts, and case reports rather than from pooled trial data. Nearly all of this literature originates from Chinese and Japanese traditional-medicine institutions, and the yokukansan trials were largely conducted with product supplied by its sole manufacturer, Tsumura & Co.

Mechanism of Action

Gou-teng’s activity is attributed to oxindole and indole alkaloids — nitrogen-containing plant compounds — concentrated in the hooks: rhynchophylline, isorhynchophylline, corynoxeine, isocorynoxeine, hirsutine, hirsuteine, and geissoschizine methyl ether (GM).

In blood vessels, rhynchophylline blocks L-type calcium channels, opens potassium channels in smooth muscle, and promotes nitric-oxide-mediated relaxation of the vessel lining — the same three levers used by conventional calcium channel blockers (drugs such as amlodipine that widen arteries) and nitrates.

In the brain, GM is the most potent of the alkaloids at the serotonin 5-HT1A receptor, a switch that damps arousal and aggression. Rhynchophylline and isorhynchophylline block NMDA (N-methyl-D-aspartate) receptors, the glutamate-gated channels whose over-activation kills neurons through calcium overload. Alkaloid extracts suppress TLR4/NF-κB/NLRP3 signalling (an innate-immune inflammation cascade) and activate Nrf2/HO-1 (a cellular antioxidant defence pathway), while corynoxine B induces autophagy, the cell’s protein-clearance system.

A competing reading holds that these mechanisms cannot account for clinical effects, because oral bioavailability is very low — 23.4% for rhynchophylline and 4.0% for isorhynchophylline in rats — so achievable brain concentrations may sit below those used in cell experiments; on this view flavonoids such as catechin, or gut-derived metabolites, carry more of the effect.

The alkaloids are absorbed and cleared rapidly, with a rhynchophylline half-life near 4.8 hours in rats, distribute across brain regions after crossing the blood–brain barrier, and are metabolised chiefly by cytochrome P450 enzymes — the liver’s main drug-processing family — including CYP3A4, CYP1A2, and CYP2C9, with P-glycoprotein (a cell-membrane drug-efflux pump) limiting absorption.

Historical Context & Evolution

Gou-teng entered the written record in the Mingyi Bielu, compiled by Tao Hongjing around 500 CE, where the hooked stem is indicated for convulsions in children. Its classical role was narrow and symptomatic: to settle tremor, spasm, dizziness, and irritability, states grouped under “liver wind” and “liver yang rising”. It was never a tonic herb, and traditional practice attached no longevity claim to it.

Three formulas carried it forward. Choto-san appears in Xu Shuwei’s Ben Shi Fang (1132) and was absorbed into Japan’s Kampo system — the regulated national canon of traditional herbal formulas. Yokukansan appears in Xue Kai’s Bao Ying Cuo Yao (1555), again for childhood irritability. Tian Ma Gou Teng Yin is modern, composed by Hu Guangci in 1958 explicitly for hypertension.

The pivot toward health optimisation came from two directions. Japanese groups ran placebo-controlled dementia trials from the 1990s onward, first reporting improved cognitive scores in vascular dementia and later in agitation and hallucinations. Independently, Chinese pharmacologists traced blood-pressure and heart-rate effects in animals to calcium- and potassium-channel actions of the isolated alkaloids.

Scientific opinion has not settled. The Cochrane assessment of the hypertension formula found no trial meeting its inclusion criteria, and Chinese-language trials reporting benefit are of low methodological quality. That gap has only partly closed: one placebo-controlled trial published since found a small blood-pressure reduction in masked hypertension — pressure raised outside the clinic only — leaving the size of the effect in ordinary hypertension unsettled.

Expected Benefits

High 🟩 🟩 🟩

Reduced Agitation, Aggression, and Psychotic Symptoms in Dementia

The most consistently replicated effect, though it belongs to the yokukansan formula rather than to Gou-teng alone. The proposed driver is GM’s action at the serotonin 5-HT1A receptor, which reduces aggression and social reactivity in animals. The evidence base is a meta-analysis of five randomised placebo-controlled trials with no measurable disagreement between studies, plus a separate pooled analysis reaching the same conclusion. Two caveats limit it: the benefit did not hold in Alzheimer’s-only populations, and general cognitive scores were unchanged.

Magnitude: Total behavioural-symptom scores fell by a standardised mean difference (a pooled effect size expressed in standard-deviation units) of −0.32, 95% CI (confidence interval, the range in which the true value most plausibly lies) −0.53 to −0.11; the agitation/aggression subscale fell −0.37 (−0.60 to −0.15), delusions −0.51 (−0.98 to −0.04), and hallucinations −0.54 (−0.96 to −0.12).

Medium 🟩 🟩

Blood Pressure Reduction ⚠️ Conflicted

Gou-teng’s alkaloids widen arteries through calcium-channel blockade and nitric-oxide release, and the Tian Ma Gou Teng formula is prescribed across East Asia on that basis. The evidence is directly conflicted: a meta-analysis of 15 Chinese randomised trials reports a substantial added reduction when the granules are layered on conventional drugs, while a placebo-controlled trial in masked hypertension found only a small one. The add-on trials are of low methodological quality from institutions invested in the tradition, so the larger figure is plausible but unverified.

Magnitude: Added to standard antihypertensive drugs, the granules lowered systolic pressure by a further 10.24 mmHg (95% CI 6.95 to 13.54) and diastolic pressure by 5.33 mmHg (3.45 to 7.21) across 1,508 patients; against placebo, the one controlled trial found a daytime reduction of 2.52/1.79 mmHg in 251 people.

Low 🟩

Improved Cognitive Scores in Vascular Cognitive Impairment

Choto-san improved continuous cognitive measures in pooled placebo-controlled trials, plausibly through improved cerebral perfusion and reduced excitotoxic injury — nerve-cell death from over-stimulation. The pooled analysis rates its own evidence low quality and warns the estimate may be inflated.

Magnitude: Direction is favourable on continuous cognitive scores across three trials in 219 people, but absent on the yes/no judgement of global improvement, and the effect holds only in vascular rather than Alzheimer’s dementia; disagreement between studies was high and the review reports no pooled effect figure for the continuous outcome.

Improved Sleep Continuity and Reduced Anxiety

Attributed to the same 5-HT1A activation that reduces agitation. An open-label trial in dementia with Lewy bodies — a dementia marked by prominent hallucinations — reported improved sleep and fewer hallucinations, but had no control group.

Magnitude: Direction is toward improvement where baseline sleep disturbance is present; no placebo-controlled trial isolates sleep as an outcome, and the available reports give no outcome figure for it.

Speculative 🟨

Clearance of Aggregated Proteins in Neurodegeneration

Corynoxine B and related alkaloids induce autophagy and accelerate removal of α-synuclein and tau — the proteins that build up in Parkinson’s and Alzheimer’s disease — in cell and rodent models. No controlled human study exists.

Endothelial Protection and Reduced Platelet Aggregation

Rhynchophylline improves function of the vessel lining and isorhynchophylline inhibits platelet clumping in rodent work. No human cardiovascular endpoint has been measured; the basis is preclinical only.

Raised Seizure Threshold

The herb’s oldest recorded indication. Animal seizure models show anticonvulsant activity, attributed to NMDA-receptor blockade and reduced brain inflammation, but no controlled human epilepsy trial has been conducted.

Benefit-Modifying Factors

  • CYP3A4, CYP1A2, and CYP2C9 activity: These liver enzymes break down Gou-teng’s alkaloids. Rapid-metaboliser variants, or enzyme-inducing drugs, lower systemic exposure; inhibitors raise it. Genotype-guided dosing has never been tested for this herb.

  • Dementia subtype at baseline: Benefit is measurable in vascular dementia and in mixed behavioural symptoms, but pooled trials found no effect in Alzheimer’s-only populations — the single strongest predictor of response.

  • Baseline blood pressure and symptom severity: Add-on trials recruited people with uncontrolled hypertension, and behavioural trials recruited people with active agitation. Those already at target, or asymptomatic, have no measured headroom for benefit.

  • Baseline potassium and albumin: Low starting values shorten the time to a potassium-lowering adverse event in licorice-containing formulas, forcing dose reduction or discontinuation before any benefit accrues.

  • Sex: Metabolism of geissoschizine methyl ether by liver enzymes is sex-dependent in rats but not in humans, and no trial has reported sex-stratified efficacy. No sex-based difference in benefit is established.

  • Age: Every positive trial recruited older adults, most above 65, and the completed monotherapy trial required an age of 65 or over. Efficacy in adults under 50 is entirely untested.

Potential Risks & Side Effects

High 🟥 🟥 🟥

Pseudoaldosteronism and Low Potassium from Licorice-Containing Formulas

The dominant real-world harm of Gou-teng products, though the causal agent is the co-formulated licorice, not Gou-teng itself. Licorice’s glycyrrhizic acid blocks 11β-HSD2, the kidney enzyme that normally shields salt-and-water regulation from cortisol, producing pseudoaldosteronism — a state of raised blood pressure, potassium loss, and muscle weakness. Evidence comes from Japan’s national adverse-event database and a retrospective cohort of psychiatric inpatients. It is reversible on withdrawal but has caused severe hypokalemia (dangerously low blood potassium) and, rarely, dangerous heart rhythms.

Magnitude: 17.4% of 304 inpatients given yokukansan developed low potassium; in the national database yokukansan carried a reporting odds ratio (a signal strength comparing how often an event is reported for one drug versus others) of 2.4, 95% CI 1.9 to 2.8, for pseudoaldosteronism versus other licorice-containing formulas, with doses of 7.5 g/day or more, dementia, age above 70, and body weight below 50 kg raising risk.

Medium 🟥 🟥

Drug-Induced Liver Injury ⚠️ Conflicted

Liver injury attributed to Uncaria hook has been reported repeatedly in the Chinese safety-monitoring literature, prompting a computational screen that nominated six constituents — cadambine, rhynchophylline, corynoxeine, isocorynoxeine, strictosamide, and mitraphylline — acting through interleukin-17, an inflammatory signalling molecule. A rodent study found the opposite, the extract protecting against chemically induced liver damage, so direction may depend on dose and species. The only placebo-controlled monotherapy trial excluded anyone with known liver impairment.

Magnitude: Not quantified in available studies. No controlled trial has measured incidence; the signal rests on case reports and a computational screen, neither of which supports a rate.

Low 🟥

Excessive Blood Pressure Drop and Slowed Heart Rate

The intended vascular effect, overshooting. Calcium-channel blockade and potassium-channel opening lower both pressure and heart rate, and the classic animal pharmacology documents dose-dependent hypotension (low blood pressure) and bradycardia (an abnormally slow heartbeat). Risk concentrates in people already on antihypertensives, since the effect is additive rather than replacing.

Magnitude: Direction is downward, with the added systolic reduction of roughly 10 mmHg from add-on trials setting the expected scale; symptomatic low blood pressure and slow heart rate are not separately reported in any trial, so no incidence figure exists.

Sedation and Daytime Drowsiness

A predictable consequence of 5-HT1A activation and the alkaloids’ sedative profile, and the reason the herb was classically given for restlessness. Pooled randomised trials found no excess of adverse events or of withdrawal for adverse events against placebo, so any sedation is mild at trial doses.

Magnitude: Direction is toward sedation, concentrated at daily doses of 7.5 g or above and in people already taking central nervous system depressants; pooled trials report no separate incidence figure for drowsiness.

Gastrointestinal Upset and Reduced Appetite

The common tolerability complaint with concentrated herbal granules, generally appearing in the first weeks and resolving on dose reduction or with food. Pooled trial data show no excess of adverse events overall against control, indicating the effect is mild where it occurs.

Magnitude: Direction is toward mild upper-gastrointestinal complaints early in treatment; pooled analyses report adverse events only in aggregate and give no separate figure for gastrointestinal complaints.

Speculative 🟨

Increased Bleeding Risk with Antiplatelet or Anticoagulant Therapy

Isorhynchophylline inhibits platelet activation and clot formation in mice. No human bleeding event has been reported, but the completed monotherapy trial excluded anyone taking warfarin or any anticoagulant.

Herb–Drug Interaction via Liver Enzymes and Efflux Pumps

The alkaloids are handled by cytochrome P450 enzymes and P-glycoprotein, and a rat study showed a co-administered herbal alkaloid nearly doubling rhynchophylline exposure. No human interaction study exists.

Risk-Modifying Factors

  • 11β-HSD2 activity: Reduced activity of this potassium-protecting kidney enzyme, whether inherited or caused by licorice itself, is the direct mechanism of pseudoaldosteronism. Genotyping is not clinically available, so low potassium serves as the practical readout.

  • Baseline potassium and albumin: Both low values predicted earlier onset of hypokalemia in the inpatient cohort. Low albumin also raises the free fraction of protein-bound alkaloids, plausibly amplifying pharmacological effect at the same dose.

  • Sex: No sex-specific risk has been demonstrated. Body weight below 50 kg more than doubled pseudoaldosteronism risk, and because that threshold is crossed more often by women, the exposure burden falls unequally at a fixed dose.

  • Pre-existing liver, kidney, or heart-rhythm disease: Impaired liver clearance raises alkaloid exposure; impaired kidney function worsens potassium loss; existing conduction disease compounds the herb’s heart-rate-slowing effect.

  • Age above 70: Independently raised pseudoaldosteronism risk almost sixfold, compounded by lower body weight, multiple concurrent medications, and reduced kidney reserve at the older end of the target range.

Key Interactions & Contraindications

  • Antihypertensives — caution, dose separation: Calcium channel blockers (amlodipine), ACE inhibitors (drugs such as lisinopril that block a blood-pressure-raising enzyme), ARBs (angiotensin receptor blockers such as losartan), and beta-blockers (heart-slowing drugs such as bisoprolol) combine additively. Consequence: symptomatic low blood pressure, dizziness, falls.

  • Rate-slowing cardiac drugs — caution, monitor: Verapamil, diltiazem, bisoprolol, and digoxin all slow cardiac conduction, as do the Uncaria alkaloids. Consequence: symptomatic bradycardia. Monitor resting heart rate weekly during titration.

  • Diuretics and corticosteroids — caution, monitor potassium: Loop diuretics (urine-increasing drugs such as furosemide), thiazides (hydrochlorothiazide), and corticosteroids (steroid anti-inflammatories such as prednisolone) add to licorice-driven potassium loss. Consequence: severe hypokalemia and muscle weakness. Check potassium every 4 weeks.

  • Digoxin — absolute contraindication with licorice-containing formulas: Potassium depletion sharply increases digoxin’s toxicity at unchanged blood levels. Consequence: life-threatening arrhythmia (irregular heartbeat). Single-herb Gou-teng without licorice carries no such interaction.

  • Anticoagulants and antiplatelets — caution, avoid combination: Blood thinners (warfarin, apixaban, rivaroxaban) and clot-preventing drugs (aspirin, clopidogrel). Consequence: theoretical additive bleeding risk from platelet inhibition by isorhynchophylline. The only monotherapy trial excluded all anticoagulant users.

  • CYP3A4 substrates with narrow safety margins — caution, monitor drug levels: Tacrolimus, ciclosporin, simvastatin. Consequence: unpredictable exposure, since the alkaloids share the enzyme and inhibit P-glycoprotein. Separate dosing by 3–4 hours and monitor levels where assays exist.

  • Central nervous system depressants — caution, reduce dose: Sedating medicines — benzodiazepines (lorazepam), Z-drugs (sleep agents such as zolpidem), antipsychotics (quetiapine), opioid painkillers, and sedating antihistamines (diphenhydramine). Consequence: excess sedation, confusion, and fall risk in older adults.

  • Over-the-counter analgesics — caution, monitor blood pressure: NSAIDs (non-steroidal anti-inflammatory drugs such as ibuprofen and naproxen) raise blood pressure and impair kidney potassium handling. Consequence: loss of the intended vascular effect plus compounded potassium loss.

  • Blood-pressure-lowering supplements — caution, stagger introduction: Beetroot or dietary nitrate, hibiscus, garlic extract, magnesium, potassium salts, and taurine act additively with Gou-teng. Consequence: overshoot and orthostatic hypotension (dizziness on standing from a blood-pressure drop).

  • Sedating supplements — caution, reduce dose: Valerian, kava, passionflower, melatonin, glycine, and high-dose magnesium potentiate drowsiness through overlapping central mechanisms. Consequence: excess daytime sedation and impaired driving performance.

  • Licorice root supplements — caution, avoid stacking: Taken alongside yokukansan or Choto-san, which already contain licorice, the combined glycyrrhizic acid load compounds directly. Consequence: pseudoaldosteronism at lower formula doses than expected.

  • Other Chinese herbal formulas — caution, avoid concurrent use: A rat study showed a co-administered alkaloid nearly doubling rhynchophylline exposure and halving its clearance. Consequence: unpredictable dose escalation. Migraine and cognition trials both excluded concurrent herbal medicines.

Populations who should avoid Gou-teng:

  • Anyone with serum potassium below 3.5 mmol/L, or on a potassium-depleting drug without monitoring
  • Anyone with known liver impairment or unexplained liver-enzyme elevation above twice the upper reference limit
  • Anyone with Child-Pugh Class B or C cirrhosis (moderate to severe liver-function failure)
  • Anyone with an estimated glomerular filtration rate below 30 mL/min/1.73 m² (severely reduced kidney filtration)
  • Anyone with resting heart rate below 50 beats per minute, second- or third-degree heart block (impaired electrical conduction in the heart), or an unpaced sick sinus syndrome (a failing natural pacemaker)
  • Anyone with systolic blood pressure below 100 mmHg or a history of symptomatic orthostatic hypotension
  • Anyone taking warfarin or a direct oral anticoagulant
  • Anyone taking digoxin, when the preparation contains licorice
  • Pregnant or breastfeeding women, and anyone under 18, in whom no controlled safety data exist
  • Anyone with a known allergy to Rubiaceae-family plants

Risk Mitigation Strategies

  • Choose a licorice-free preparation where possible: Single-herb Gou-teng granules carry none of the pseudoaldosteronism risk behind the 17.4% hypokalemia rate seen with yokukansan. This removes the largest documented harm outright.

  • Cap licorice-containing formulas below 7.5 g/day: Doses of 7.5 g/day and above marked the threshold at which hypokalemia risk rose. Staying at 5 g/day or less keeps exposure under that inflection point.

  • Check potassium at baseline, week 4, and quarterly: Low potassium precedes symptoms. Anyone starting below 4.0 mmol/L, or on a diuretic or corticosteroid, moves to 4-weekly checks for the first 3 months to catch pseudoaldosteronism early.

  • Check liver enzymes at baseline and week 8: Uncaria hook has repeated case reports of liver injury with no established incidence. Baseline and early follow-up liver-enzyme measurement converts an unquantified risk into a detectable one.

  • Start at half the target dose for 2 weeks: Beginning at 1.2 g/day of granule and titrating after 14 days limits the initial blood-pressure drop and surfaces sedation or gastrointestinal upset before full exposure.

  • Record home blood pressure and resting heart rate for 7 days before starting: Establishes the baseline against which overshoot is judged, and identifies pre-existing bradycardia below 55 beats per minute that would contraindicate the herb.

  • Take the first dose in the evening: Shifts any sedation and the peak vascular effect into sleep, reducing daytime drowsiness and the fall risk that follows an unanticipated blood-pressure drop.

  • Stop 7 days before any surgery or dental extraction: Removes the theoretical additive bleeding risk from platelet inhibition and the interaction with anaesthetic agents that lower blood pressure.

Therapeutic Protocol

  • Crude herb decoction: The classical form. Chinese pharmacopoeial dosing is 3–12 g of dried hooked stem daily, decocted in water and taken in two divided portions, most often within a multi-herb formula rather than alone.

  • Late addition to the decoction: Traditional practice adds Gou-teng only in the final 10–20 minutes of boiling, on the reasoning that prolonged heat degrades the alkaloids. This is the single best-known preparation rule for the herb.

  • Concentrated granule, single herb: The only regimen tested against placebo as monotherapy used 1.2 g of granule twice daily for 16 weeks in adults aged 65 and over with mild cognitive impairment (measurable memory decline short of dementia).

  • Concentrated granule, formula: Yokukansan is dosed at 7.5 g/day in three portions, supplying roughly 3 g of Uncaria hook; Gou-Teng-San is under trial at 5 g twice daily for migraine.

  • Competing approach — whole formula: East Asian practitioners overwhelmingly prescribe Gou-teng inside a formula, arguing the co-herbs are integral. Choto-san descends from Xu Shuwei; Tian Ma Gou Teng Yin from Hu Guangci.

  • Competing approach — single herb: The Hong Kong group led by Zhixiu Lin at the Chinese University of Hong Kong ran the only isolated-herb trial, on the view that formula testing cannot attribute effect and complicates dose control.

  • Best time of day: Evening dosing suits the sedative and blood-pressure-lowering profile; the migraine and cognition trials both used split morning and evening dosing, so a split regimen is the tested default.

  • Half-life and dose splitting: Rhynchophylline’s half-life is roughly 4.8 hours in rats and the alkaloids clear rapidly, which is why every human protocol splits the daily dose into two or three portions.

  • Genetic polymorphisms: No pharmacogenetic testing is validated for Gou-teng. CYP3A4, CYP1A2, and CYP2C9 variants plausibly alter alkaloid exposure, and reduced 11β-HSD2 activity raises licorice sensitivity, but neither has been tested prospectively.

  • Sex-based differences: Liver metabolism of geissoschizine methyl ether differs by sex in rats but not in humans, and no trial reports sex-stratified dosing. Lower average body weight argues for the lower end of the range.

  • Age-related considerations: All efficacy trials recruited older adults, but age above 70 independently raised pseudoaldosteronism risk almost sixfold, so the oldest start at the lowest dose with 4-weekly potassium checks.

  • Baseline biomarkers: Starting potassium, liver enzymes, kidney filtration, blood pressure, and resting heart rate determine both eligibility and starting dose. Potassium below 4.0 mmol/L argues for a licorice-free preparation.

  • Pre-existing conditions: Vascular cognitive impairment and active behavioural symptoms are the settings with measured benefit. Controlled hypertension, Alzheimer’s-only dementia, liver disease, and conduction disease each argue against expecting benefit or tolerating the herb.

Discontinuation & Cycling

  • Intended duration: Short-to-medium term, not lifelong. The tested regimens ran 4 weeks for migraine, 12 weeks for behavioural symptoms, and 16 weeks for cognition; no trial has run beyond 6 months, so open-ended use is unstudied.

  • Withdrawal effects: None reported. No trial has documented rebound agitation, rebound hypertension, or any withdrawal syndrome, and the alkaloids’ short half-life makes physical dependence mechanistically implausible.

  • Tapering: Not required pharmacologically. The exception is blood-pressure use: abrupt discontinuation returns pressure to its untreated level within days, so conventional antihypertensive dosing is normally reviewed before stopping.

  • Stopping for adverse events: Pseudoaldosteronism reverses on withdrawal, with potassium typically recovering over 1–2 weeks. Immediate discontinuation is the primary treatment, ahead of potassium supplementation.

  • Cycling: No efficacy-maintenance rationale exists, as tolerance has never been demonstrated. A pragmatic pattern is 12–16 weeks on, then reassessment against the pre-treatment baseline, with continuation only where a measured change occurred.

  • Reassessment trigger: Where the tracked outcome — blood pressure, agitation score, or cognitive screen — is unchanged at 16 weeks, continuing carries the accumulating potassium and liver risk without a demonstrated return.

Sourcing and Quality

  • Botanical species: Five Uncaria species are official sources — U. rhynchophylla, U. macrophylla, U. sinensis, U. hirsuta, and U. sessilifructus — with materially different alkaloid profiles. A certificate of analysis naming the species is the single most important document.

  • Plant part: Only the hook-bearing stem qualifies as Uncariae Ramulus cum Uncis. Hookless stem is a cheaper, lower-alkaloid substitution; visible hooks on the raw herb are the simplest authenticity check available to a buyer.

  • Alkaloid standardisation: Look for quantified rhynchophylline and isorhynchophylline content on the label. Most crude herb is sold unstandardised, which makes dose reproducibility across batches and suppliers effectively impossible.

  • Third-party testing: Chinese crude botanicals carry documented heavy-metal and pesticide burdens. Independent certification for identity, heavy metals, pesticides, and microbial load — from testers such as USP, NSF, or Eurofins — is the practical minimum.

  • Pharmaceutical-grade formulas: Japanese Kampo granules are made to pharmaceutical manufacturing standards and are the most consistent products available. Their sole manufacturer, Tsumura & Co., also funds and authors much of the supporting research.

  • Avoid species confusion: Uncaria tomentosa and Uncaria guianensis, sold as cat’s claw, are South American species with entirely different constituents. Products labelled only “Uncaria” or “cat’s claw” are not Gou-teng.

Practical Considerations

  • Time to effect: Blood-pressure change appears within 2–4 weeks. Behavioural symptoms improved over 4–12 weeks in trials; the cognition trial measured its primary outcome at 16 weeks, which is the realistic horizon for any cognitive signal.

  • Common pitfall — over-boiling: Boiling Gou-teng for the full decoction time degrades its alkaloids. Adding it only in the final 10–20 minutes is the standard corrective, and the most frequently ignored preparation rule.

  • Common pitfall — assuming the herb equals the formula: Almost all clinical evidence tests multi-herb formulas. Taking single-herb Gou-teng and expecting the dementia-agitation result imports a finding the isolated herb has never demonstrated.

  • Common pitfall — stacking licorice: Yokukansan and Choto-san already contain licorice. Adding a licorice supplement, or a second licorice-containing formula, is how most pseudoaldosteronism cases arise.

  • Regulatory status: A dietary supplement in the United States with no approved indication, a pharmacopoeial drug in China, and a prescription Kampo medicine in Japan reimbursed under national health insurance.

  • Structural incentives: Reimbursement makes these formulas markedly cheaper than branded agents for dementia agitation, giving Japanese payers a systematic reason to favour them; in Western systems, where nothing reimburses them, no payer or guideline body has reason to fund comparative trials at all.

  • Organisational interests: The Japan Society for Oriental Medicine and the China Association of Chinese Medicine endorse these formulas, and their memberships derive practice income from prescribing them; symmetrically, the psychiatric and geriatric societies favouring pharmaceutical comparators have members whose income depends on those.

  • Cost and accessibility: Inexpensive and widely available. Crude herb costs a few US dollars per 100 g; Kampo granules cost more but remain far below branded pharmaceuticals, so cost is not a meaningful barrier.

Interaction with Foundational Habits

  • Sleep: Direct and potentiating. Activation of the serotonin 5-HT1A receptor by geissoschizine methyl ether damps arousal, and the herb’s classical indication was restlessness. Evening dosing exploits this; morning dosing risks daytime drowsiness. For anyone already taking a sedating agent the effect is additive, and the other agent is usually reduced first.

  • Nutrition: Indirect. Absorption of the alkaloids is food-sensitive, so dosing is kept consistent relative to meals. Where a licorice-containing formula is used, a high-potassium diet — leafy greens, legumes, avocado — offsets renal potassium loss, while high sodium intake amplifies the pseudoaldosteronism mechanism directly.

  • Exercise: Indirect, potentially blunting. Added blood-pressure and heart-rate reduction narrows cardiovascular reserve, so exertional dizziness and a lower peak heart rate are plausible. Training intensity anchored to heart rate typically needs recalibration after 2 weeks. No evidence suggests any effect on muscle growth or strength adaptation.

  • Stress management: Direct and potentiating. The same receptor action that reduces aggression in dementia trials plausibly lowers stress reactivity, and the herb’s traditional use targeted irritability. No human study has measured cortisol or any stress biomarker, so the interaction is mechanistically expected rather than demonstrated.

Monitoring Protocol & Defining Success

Baseline testing precedes the first dose and settles the two variables that decide whether Gou-teng is tolerable: potassium handling and liver function. Blood pressure and resting heart rate are recorded across a full week beforehand, since the vascular effect is the one most likely to overshoot.

Ongoing monitoring is front-loaded: potassium and blood pressure at 4 weeks, liver enzymes at 8 weeks, then the full panel at 6 months and every 6–12 months after. Anyone on a licorice-containing formula, a diuretic, or a corticosteroid moves to potassium checks every 4 weeks for 3 months. Success is defined against the pre-treatment baseline: a measurable fall in agitation score, blood pressure, or both by 16 weeks, with potassium and liver enzymes unchanged.

Qualitative markers tracked alongside laboratory values:

  • Agitation and irritability — the outcome with the strongest trial support
  • Sleep continuity — night wakings and time to fall asleep
  • Daytime alertness — the earliest sign of over-sedation
  • Muscle weakness or cramping — the first symptom of falling potassium
  • Dizziness on standing — signals blood-pressure overshoot
Biomarker Optimal Functional Range Why Measure It? Context/Notes
Serum potassium 4.2–4.8 mmol/L Detects pseudoaldosteronism before symptoms Conventional range (3.5–5.2 mmol/L) tolerates values that already signal licorice effect. No fasting needed. Baseline, week 4, then quarterly
Blood pressure, home seated Below 120/75 mmHg The intended effect and the likeliest overshoot Average two morning and two evening readings across 7 days. Pair with a standing reading to catch the drop on standing
Resting heart rate 55–70 beats per minute Alkaloids slow cardiac conduction; bradycardia is the dose-limiting cardiac effect Measure lying down on waking, before rising. A wearable trend is sufficient. Below 55 argues against continuing
ALT and AST ALT below 25 U/L in men, below 20 U/L in women; AST below 25 U/L Liver injury is the least quantified serious risk ALT and AST are alanine and aspartate aminotransferase, liver enzymes released by damaged liver cells. Conventional limits near 40–55 U/L miss early injury. Fast 8–12 hours; pair with GGT (gamma-glutamyl transferase, a bile-duct enzyme) and bilirubin
eGFR Above 90 mL/min/1.73 m² Kidney function governs potassium handling and alkaloid clearance eGFR is the estimated glomerular filtration rate, a calculated measure of kidney filtering capacity. Fasting preferred but not required. Below 30 is a contraindication. Repeat with each potassium panel
Serum magnesium 2.0–2.5 mg/dL Magnesium depletion makes potassium loss resistant to correction Conventional range extends down to 1.7 mg/dL, which is too permissive here. Draw with the potassium panel
Plasma renin activity and aldosterone No established target for this herb; track the direction of change from the individual’s own baseline, with both suppressed alongside low potassium confirming pseudoaldosteronism Distinguishes licorice effect from other causes of potassium loss Draw in the morning after 15 minutes seated. Order only if potassium falls below 3.8 mmol/L
Cognitive screen 26/30 or above Anchors any claimed cognitive benefit to a measurement The Montreal Cognitive Assessment is a 30-point office test of memory and executive function. Use the same language version at the same time of day; ceiling effects blunt sensitivity in high performers. Baseline and 16 weeks

Emerging Research

  • Gou-teng monotherapy in mild cognitive impairment: The only placebo-controlled trial of the isolated herb (NCT04532697), 56 adults aged 65 and over, quadruple-blinded, 1.2 g granule twice daily for 16 weeks. Completed October 2023; results have not been posted.

  • Gou-Teng-San for episodic migraine: NCT07496034, a Phase 3 double-blind trial of 40 adults taking 5 g twice daily for 4 weeks, with change in monthly migraine days as the primary endpoint. Recruiting in Taiwan; primary completion December 2026.

  • Gastrodia and Uncaria Drink in masked hypertension — could strengthen the case: NCT02156024, a Phase 4 trial in masked hypertension at the Shanghai Institute of Hypertension, reported a 2.52/1.79 mmHg daytime reduction against placebo in 251 people (Zhang et al., 2020).

  • Bioavailability ceiling — could weaken the case: First-pass breakdown in the liver, differing sharply between the two mirror-image forms, gives rhynchophylline 23.4% and isorhynchophylline 4.0% oral bioavailability in rats (Wang et al., 2017). Comparable human figures would put many cell-culture concentrations out of reach.

  • Liver-injury screening — could weaken the case: A computational screen nominates six Uncaria constituents as liver-injury candidates acting through interleukin-17 signalling (Zhang et al., 2022). Targeted human safety surveillance would either substantiate the signal or dismiss it.

  • Species standardisation — could weaken the case: Comparative chemistry across the five official source species (Cao et al., 2024) shows differing constituent profiles, meaning trial results may not transfer between products sold under the same name.

  • Neuroinflammation control — could strengthen the case: Alkaloid extracts reduce inflammation and oxidative stress in Parkinson’s models via the TLR4/NF-κB/NLRP3 and Nrf2/HO-1 pathways (Zhang et al., 2024). Human dose-ranging work would be the necessary next step.

  • Endothelial repair — could strengthen the case: Rhynchophylline was identified as an activator of sirtuin-3, a mitochondrial maintenance enzyme, restoring energy-producing function in vessel-repair cells (Lin et al., 2025) — a vascular-longevity mechanism distinct from blood-pressure lowering.

Conclusion

Gou-teng is the hooked stem of an East Asian vine, used for fifteen centuries against tremor, dizziness, irritability, and high blood pressure. Its plant compounds widen blood vessels and act on a brain receptor that governs arousal and aggression, and modern research largely confirms both.

The strongest finding is a reduction in agitation, aggression, delusions, and hallucinations in dementia, replicated across randomised trials with unusually consistent results. Added to standard blood-pressure drugs, one formula appears to lower pressure further, but the evidence is contradictory: the trials showing a large effect are of poor quality, and the single trial against a dummy treatment found only a small one. Cognitive improvement in dementia caused by poor blood supply to the brain is weaker still. Claims about slowing the loss of brain cells rest entirely on cells and animals.

The main harm is a licorice-driven loss of potassium affecting a sizeable minority taking the Japanese formulas — caused by a companion herb rather than by Gou-teng itself, and reversible on stopping. Liver injury has been reported often enough to register as a genuine concern, though no rate is known.

Two features weaken the whole evidence base. Almost everything tests multi-herb mixtures rather than the herb alone, and the research is dominated by parties with an interest in the outcome: the sole manufacturer of the Japanese formulas, Chinese institutions whose standing rests on validating the tradition, and professional bodies on both the herbal and conventional sides that earn income from what they endorse.

Top - Benefits - Risks - Protocol