Zhi Cao Wu
StarAconitum kusnezoffii Reichb.
Traditionally used for
- Nerves & recovery
- Pain & joints
Cautions & contraindications
- Pregnancy
- Heart conditions
- Toxic — professional use only
☯ TCM Properties
Dispels Wind-Dampness; Warms the Channels and Disperses Cold; Alleviates Pain; Reduces Swelling and Draws Out Toxins
Traditional Chinese Uses
Zhi Cao Wu (processed aconite lateral root, Aconitum kusnezoffii) is an extremely toxic herb processed to reduce its toxicity before clinical use. Like processed Chuan Wu, it strongly warms the channels, disperses deep cold, and unblocks Qi and Blood in severe Wind-Cold-Damp bi with intractable joint pain and numbness. It is reserved for serious cold-obstruction conditions when other treatments have been inadequate, requiring strict professional supervision.
Relationships
Full-screen graph →Click a node for details · double-click to expand it · Ctrl/⌘ + scroll or two fingers to zoom and pan.
Botanical Description
Aconitum kusnezoffii Reichb. (Ranunculaceae) is a perennial herb native to northeastern China, Korea, eastern Russia, and Mongolia, growing 70-150 cm tall with palmately divided leaves and showy dark purple, helmet-shaped flowers in terminal racemes. It bears a paired tuberous root system of a 'mother' and 'daughter' tuber. Zhi Cao Wu is the processed lateral (daughter) tuber, prepared by prolonged soaking, repeated boiling, and steaming to hydrolyze the highly toxic diester diterpenoid alkaloids (mesaconitine, hypaconitine, deoxyaconitine) into far less toxic monoester or amine forms. Raw Cao Wu is severely cardio- and neurotoxic and is not used internally. In traditional Chinese medicine, processed Zhi Cao Wu is acrid, bitter, hot, and toxic, entering the heart, liver, kidney, and spleen channels; it dispels wind-damp, warms the channels, and relieves severe pain, used for wind-damp painful obstruction, cold-pattern joint pain, traumatic injury, and severe abdominal cold pain. Strict dose limits and prolonged decoction are required.
Active Constituents
Aconitine
Diester diterpenoid (C19-norditerpenoid) alkaloidConcentration: 4.5 +/- 2.1 micrograms per gram in processed Caowu, against 83.7 +/- 6.9 micrograms per gram in the raw root (Sun et al. 2012) - roughly a 19-fold reduction
Sodium-channel site 2 agonist: it holds voltage-gated sodium channels open, producing sustained depolarisation, calcium overload and ventricular arrhythmia, with an oral toxic threshold near 0.2 mg and reported lethal doses of 2-4 mg. Residual aconitine is the reason correctly processed Zhi Cao Wu is still a Pharmacopoeia-graded toxic drug rather than a safe herb.
Mesaconitine
Diester diterpenoid alkaloidConcentration: 4.7 +/- 0.8 micrograms per gram in processed Caowu, against 527.1 +/- 66.7 micrograms per gram raw (Sun et al. 2012) - roughly a 112-fold reduction
Overwhelmingly the dominant diester of raw Cao Wu and the compound whose destruction accounts for most of the drop in toxicity on processing. It hydrolyses readily at the C8 acetyl group to benzoylmesaconine. Its cardiotoxic mechanism is the same as aconitine's.
Hypaconitine
Diester diterpenoid alkaloidConcentration: 24.0 +/- 1.4 micrograms per gram in processed Caowu, against 240.5 +/- 6.4 micrograms per gram raw (Sun et al. 2012) - roughly a 10-fold reduction
The heat-resistant diester. Because it survives boiling far better than aconitine and mesaconitine, it becomes the largest single residual diester in a correctly processed drug - about five times the residual aconitine content - and is the alkaloid most likely to be detected after poisoning by an insufficiently decocted preparation.
Benzoylaconine
Monoester diterpenoid alkaloidConcentration: One of the three monoesters used as the Pharmacopoeia potency marker for processed Caowu
First hydrolysis product of aconitine, one to two orders of magnitude less acutely toxic while retaining analgesic and anti-inflammatory activity. Its accumulation, not merely the disappearance of the diesters, is what defines a correctly processed drug.
Benzoylmesaconine
Monoester diterpenoid alkaloidConcentration: Usually the most abundant monoester in processed Caowu
Hydrolysis product of mesaconitine and the principal intended active constituent of the processed drug. Its plasma exposure after Hezi-processed Caowu differs measurably from that after the raw drug.
Benzoylhypaconine
Monoester diterpenoid alkaloidConcentration: Third component of the Pharmacopoeia monoester assay
Hydrolysis product of hypaconitine. Because hypaconitine resists heat, this monoester accumulates more slowly than the other two during processing.
Aconine
Amine (non-esterified, aminoalcohol) diterpenoid alkaloidConcentration: Endpoint of the two-step hydrolysis; accumulates with prolonged heating
Loss of the C14 benzoyl group yields an alkaloid that is essentially non-toxic and correspondingly weak. Processing is therefore a titration rather than a maximisation: driving hydrolysis to completion destroys the drug's activity along with its danger.
⚠ Drug Interactions
Raw Cao Wu (unprocessed Aconitum kusnezoffii root)
Dispensing the raw root where the processed root is intended is the most consequential error possible with this genus. Raw Cao Wu carries roughly 19-fold more aconitine, 112-fold more mesaconitine and 10-fold more hypaconitine than the processed drug (Sun et al. 2012); its total diester burden is on the order of 850 micrograms per gram against about 33 in processed material. At the 1.5-3 g daily dose set for the processed drug, that difference alone reaches the milligram range of alkaloid that is documented as lethal. Chen et al. attributed 13.5 percent of 52 confirmed Hong Kong aconite poisonings to inadequately processed herbs and 3.9 percent to dispensary error. The corresponding data hazard is a Latin name: a raw aconite record labelled Praeparata or Cocta is indistinguishable from the processed drug on the label.
Clinical note: Verify that the label reads Zhi Cao Wu (Radix Aconiti Kusnezoffii Praeparata) and not Cao Wu (Radix Aconiti Kusnezoffii). Decoction time is a second, independent safety variable after processing: boiling for 30-60 minutes leaves residual toxicity while two hours removed measurable toxicity in the reviewed material, so decoct alone for at least 30-60 minutes before adding the rest of the formula, keep to 1.5-3 g per day, and never dispense as an undecocted powder, pill or tincture.
Antiarrhythmic drugs (amiodarone, flecainide, quinidine, lidocaine, sotalol)
The residual diester alkaloids are sodium-channel agonists that produce precisely the arrhythmias these drugs are given to suppress, and aconitine-induced arrhythmia is notoriously refractory to conventional antiarrhythmic therapy - published case management has needed combined blood purification alongside drug treatment. Adding a channel blocker or repolarisation-prolonging agent to a persistent depolarising stimulus gives unpredictable, potentially additive electrophysiological effects. European TCM safety guidance names antiarrhythmic co-medication as a specific hazard for aconite drugs.
Clinical note: Do not prescribe to a patient on antiarrhythmic therapy, or with a known arrhythmia, conduction defect or structural heart disease. Suspected poisoning in such a patient is a cardiology emergency.
Digoxin and other cardiac glycosides
Digoxin inhibits the sodium-potassium ATPase, aconite alkaloids drive sodium influx that reverses the sodium-calcium exchanger; both routes end in myocyte calcium overload, triggered activity and ectopy. The toxidromes overlap closely enough that either can mask the other. Co-medication with digitalis is an explicit contraindication in European TCM safety guidance for aconite drugs.
Clinical note: Absolute contraindication. In a patient on digoxin presenting with arrhythmia, obtain a digoxin level and send blood and urine for aconitine alkaloid screening rather than assuming glycoside toxicity.
He Zi (Terminalia chebula fruit)
Hezi-processing of Cao Wu is the standard Mongolian-medicine method and is pharmacologically active rather than merely traditional. Zhi et al. showed that Hezi-processed Caowu gave lower Cmax and AUC for aconitine (7.23 to 2.31 ng/mL; 10.31 to 4.96 ng.h/mL) and mesaconitine (37.03 to 19.83 ng/mL; 77.12 to 46.90 ng.h/mL) than the raw drug, with hypaconitine barely changed. Zhu et al. then found that Zhicaowu alone downregulated rat CYP1a2, CYP2d2, CYP3a1 and CYP2c11, while the Zhicaowu-Hezi pair upregulated them dose-dependently - so the combination accelerates clearance of alkaloids that the herb alone would slow. Wang et al. showed toxicity mitigation depends jointly on the Hezi processing, the powder dosage form and a small dose.
Clinical note: Do not treat Hezi-processed and water-processed Cao Wu as interchangeable at the same dose, and do not assume the classical Mongolian small-dose powder regimen transfers to a decoction. Because Zhicaowu alone inhibits several CYP isoforms, expect co-prescribed conventional drugs cleared by CYP1A2, CYP2D6 or CYP3A4 to be affected differently depending on whether Hezi is present.
CYP3A4 inhibitors (ketoconazole, itraconazole, clarithromycin, ritonavir, grapefruit juice)
Aconitine alkaloids are predominantly metabolised by CYP3A4, and rodent work on Zhicaowu herb pairs shows that manipulating CYP3A activity moves systemic alkaloid exposure in the expected direction. Inhibiting the route should raise and prolong exposure to residual diester. The magnitude has not been established in humans, but a drug with a 0.2 mg oral toxic threshold has no room to absorb an unquantified increase.
Clinical note: Treat strong CYP3A4 inhibition as a reason to withhold rather than to reduce the dose. Ask specifically about azole antifungals, macrolides and antiretrovirals.
Ban Xia (Pinellia ternata), Gua Lou (Trichosanthes), Bei Mu (Fritillaria), Bai Lian (Ampelopsis japonica), Bai Ji (Bletilla striata)
These five drugs form the aconite arm of the classical eighteen incompatibilities, a contraindication applied to Chuan Wu, Cao Wu and Fu Zi alike. The pairing has modern support for at least one member: in rats, Pinellia ternata enhanced absorption of aconitine and benzoylaconine from Aconitum carmichaelii, altered the metabolism of mesaconitine, benzoylmesaconine, songorine and fuziline, and changed CYP isoenzyme activity across a five-probe cocktail. That work used A. carmichaelii rather than A. kusnezoffii, so the extension to Cao Wu is inference from a shared alkaloid profile, not direct evidence.
Clinical note: Avoid these combinations. Where a source formula appears to pair them, verify the text and the intended processed forms before dispensing, and never raise the aconite dose to compensate.
Dosage
| Form | Amount | Frequency | Duration | Population | Notes |
|---|---|---|---|---|---|
| decoction | 1.5–3 g | — | — | — | ChP 2020. Must be pre-decocted (先煎) 30–60 min before adding other herbs. Corrected from a generic 9–15 g filler value. |
Dui Yao — Herb Pairs
The classical two-herb combinations this herb appears in, each with an action neither herb has alone.
The two processed aconites together powerfully dispel wind-cold-damp and relieve pain in stubborn painful obstruction.
Chronic wind-cold-damp obstruction with phlegm and blood stasis in the collaterals. Highly toxic; contraindicated in pregnancy; incompatible with Ban Xia, Gua Lou, Bei Mu, Bai Ji and Bai Lian.
Core pair of a classical formula — Xiao Huo Luo Dan, Tai Ping Hui Min He Ji Ju Fang
Evidence Tier
Moderate evidence · 6 studiesRecorded studies by study design, strongest design at the top. This is a study-design tier only, not a GRADE rating: it does not weigh risk of bias, consistency or precision.
Systematic review / meta-analysis
0
Randomized controlled trial
0
Other clinical trial
0
Observational / case report
1
0 verified · 1 unverified
In vitro / animal
5
1 verified · 4 unverified
Show 5 studies
- Quantitative and Qualitative Analysis of Aconitum Alkaloids in Raw and Processed Chuanwu and Caowu by HPLC in Combination with Automated Analytical System and ESI/MS/MS
- Influence of Different Dosage Forms on Pharmacokinetics of 6 Alkaloids in Raw Aconiti Kusnezoffii Radix (Caowu) and Chebulae Fructus- (Hezi-) Processed Caowu by UPLC-MS/MS
- Potential induction of the relative mRNA expression levels of CYP450 by Zhicaowu-Hezi (Aconiti kusnezoffii radix preparata and Terminalia chebula Retz.)
- A proteomics-based study on the mechanisms of Terminalia chebula Retz processed Aconitum kusnezoffii Reichb against rheumatoid arthritis and its cardiotoxicity reduction
- Mongolian medicine theory-based multidimensional evaluation of toxicity mitigation in Hezi-processed Caowu jointly mediated by powder dosage form and small dose
Other / unclassified
0
Verified: design read from PubMed for a DOI that resolves to the cited paper Unverified: taken from the study's recorded description
Clinical Studies
Quantitative and Qualitative Analysis of Aconitum Alkaloids in Raw and Processed Chuanwu and Caowu by HPLC in Combination with Automated Analytical System and ESI/MS/MS
Head-to-head assay of raw and processed Caowu. Processing cut aconitine from 83.7 to 4.5 micrograms per gram, mesaconitine from 527.1 to 4.7, and hypaconitine from 240.5 to 24.0 - about 19-fold, 112-fold and 10-fold respectively. Timed boiling at 15, 30, 60, 90 and 150 minutes showed aconitine and mesaconitine gone by 150 minutes while hypaconitine changed little, and heating in dioxane-water at 120 degrees Celsius for 50 minutes converted aconitine and mesaconitine largely to their benzoyl analogues. This is the quantitative basis for treating decoction time as a safety variable independent of processing.
Aconite Poisoning over 5 Years: A Case Series in Hong Kong and Lessons Towards Herbal Safety
Fifty-two laboratory-confirmed aconite poisonings in Hong Kong, 2004 to 2009. Neurological toxicity in 94.2 percent, cardiovascular in 88.5 percent, gastrointestinal in 59.6 percent. Causes: overdose 32.7 percent, hidden or inadvertently dispensed aconite 32.7 percent, inadequately processed herbs 13.5 percent, dispensary error 3.9 percent. Inadequate boiling time, leaving diesters insufficiently hydrolysed, is identified as a distinct route to poisoning even where the crude drug was correctly processed.
Influence of Different Dosage Forms on Pharmacokinetics of 6 Alkaloids in Raw Aconiti Kusnezoffii Radix (Caowu) and Chebulae Fructus- (Hezi-) Processed Caowu by UPLC-MS/MS
Rat pharmacokinetics of six alkaloids after raw and Hezi-processed Caowu given as different dosage forms. Hezi processing (soaking Caowu in a Terminalia chebula decoction at a 2:1 ratio for three days) lowered aconitine Cmax from 7.23 to 2.31 ng/mL and AUC from 10.31 to 4.96 ng.h/mL, and mesaconitine Cmax from 37.03 to 19.83 ng/mL and AUC from 77.12 to 46.90 ng.h/mL; hypaconitine exposure barely moved, consistent with its resistance to hydrolysis. Dosage form materially changed exposure, so powder and decoction are not interchangeable.
Potential induction of the relative mRNA expression levels of CYP450 by Zhicaowu-Hezi (Aconiti kusnezoffii radix preparata and Terminalia chebula Retz.)
Male Sprague-Dawley rats dosed orally for 14 days. Processed Cao Wu given alone downregulated CYP1a2, CYP2d2, CYP3a1 and CYP2c11 mRNA and inhibited the corresponding enzyme activities, with CYP2b1, CYP2c13 and CYP2e1 unchanged. The Zhicaowu-Hezi pair reversed this, dose-dependently inducing the same four isoforms with the 1:3 ratio most pronounced. Since the aconite alkaloids are largely CYP3A substrates, the finding is directly relevant both to the classical Mongolian pairing and to co-prescribed conventional drugs.
A proteomics-based study on the mechanisms of Terminalia chebula Retz processed Aconitum kusnezoffii Reichb against rheumatoid arthritis and its cardiotoxicity reduction
Proteomic study of Terminalia chebula-processed Aconitum kusnezoffii in rheumatoid arthritis models, addressing both the anti-arthritic effect and the reduction in cardiotoxicity relative to less thoroughly processed material. Provides mechanistic support for the Mongolian Hezi-processing tradition as a genuine toxicity-attenuating step rather than a purely conventional one.
Mongolian medicine theory-based multidimensional evaluation of toxicity mitigation in Hezi-processed Caowu jointly mediated by powder dosage form and small dose
Evaluates toxicity mitigation of Hezi-processed Cao Wu as a joint product of the processing, the powder dosage form and the small dose used in Mongolian practice, rather than of processing alone. The practical implication is that the safety of the traditional preparation does not transfer automatically to the same processed material given in a different form or at a larger dose.
⚠ Safety & Contraindications
- Pregnancy
- Heart conditions
- Toxic — professional use only
Contraindications
Contraindicated in pregnancy and in cardiac disease.
Safety Warnings
- Processing lowers but does not remove the toxicity.
- Decoct first for at least 30 to 60 minutes.
- Incompatible with Ban Xia, Gua Lou, Bei Mu, Bai Lian and Bai Ji.
⚠ Rule-Based Cautions
These entries come from the deterministic rule tables that gate Verscienta's formula tools — classical pair prohibitions, pregnancy and lactation contraindications, and dose ceilings.
Incompatibilities (十八反 / 十九畏)
- 十八反: Wu Tou × Ban Xia — avoid combining with Ban Xia / Pinellia (pair accepted when both members are processed)
- 十八反: Wu Tou × Gua Lou — avoid combining with Gua Lou / Trichosanthes / Tian Hua Fen
- 十八反: Wu Tou × Bei Mu — avoid combining with Bei Mu / Fritillaria / Chuan Bei / Zhe Bei
- 十八反: Wu Tou × Bai Lian — avoid combining with Bai Lian / Ampelopsis
- 十八反: Wu Tou × Bai Ji — avoid combining with Bai Ji / Bletilla
Pregnancy
Avoid toxic aconite
Breastfeeding
Avoid toxic to infant
Dose Ceiling
≤15g — prepared aconite 3–15g and must be pre-decocted; higher risks toxicity
≤6g — raw/less-processed aconite extremely toxic; keep low and pre-decoct
⚠ Toxicity Information
Oral and limb numbness, nausea, palpitations, arrhythmia — typically only at higher doses or after inadequate processing.
Historical Texts
Shen Nong Ben Cao Jing (Divine Husbandman's Classic of Materia Medica)
Han dynasty, compiled c. 200 CELei Gong Pao Zhi Lun (Master Lei's Treatise on Drug Processing), attributed to Lei Xiao
Southern and Northern Dynasties, c. 5th century CEBen Cao Gang Mu (Compendium of Materia Medica), Li Shizhen
Ming dynasty, 1596Mongolian medicine Hezi-processing tradition, recorded in the Mongolian materia medica literature
Traditional Mongolian medicine, documented from the pre-modern period and still standard practiceReferences
- Ruirui Zhang; Qing Yang; Zhuangzhuang Huang; Hongqian Liu; Jie Gao; Xizhou Nie; Yufei Liao; Ping Wang. A comprehensive review of the traditional usages, phytochemistry, pharmacology, toxicology, quality control and other applications of Aconitum kusnezoffii Reichb . Journal of Ethnopharmacology (2026) [DOI]
- Kenneth Happy; Sungyu Yang; Chang Ho Kang; Youngmin Kang. The pharmacology, toxicology, and detoxification of Aconitum kusnezoffii Reichb., traditional and modern views . Applied Biological Chemistry (2025) [DOI]
- Yabin Gao; Hang Fan; Anzheng Nie; Kang Yang; Haiyan Xing; Zhiqing Gao; Liujie Yang; Zheng Wang; Linqi Zhang. Aconitine: A review of its pharmacokinetics, pharmacology, toxicology and detoxification . Journal of Ethnopharmacology (2022) [DOI]
- Weiye Zhang; Zhubin Zhang; Ziwen Bian; Qi Zuo; Xue Quan; Jiahao Gong; Bingling Ju; Liang Feng; Xiaobin Jia; Bing Yang. Processing-induced detoxification of toxic Traditional Chinese Medicines: a systematic review of attenuation mechanisms at physical, chemical, and in vivo levels . Chinese Medicine (2026) [DOI]
- Ru Ya Tan; Pin Jun Ooi; Xiangyang Xu; Xu Wei; Xu Wen; Qiyue Qiu; Mun Fei Yam. Aconitum carmichaelii: a clinical-regulatory synthesis of traditional use, toxicological evidence, and global safety governance . Frontiers in Pharmacology (2026) [DOI]
This information is for educational purposes only and is not intended to replace professional medical advice. Always consult a qualified healthcare provider before using any herbal remedy, especially if you are pregnant, nursing, or taking medications.
📝 Notes
Public notes from the community and your own private notes on Zhi Cao Wu.
No notes yet.