Abstract
Background: The concurrent use of nutraceuticals, botanical supplements, and standardized herbal extracts alongside prescribed pharmacotherapy has reached a prevalence that renders their pharmacological effects clinically inseparable from those of conventional drugs. Despite this, structured elicitation of supplement use during clinical encounters remains inconsistently practised, contributing to unexplained treatment failures, iatrogenic toxicities, and medico-legal risk.
Objective: This review systematically characterises six mechanistically distinct classes of herb-drug interaction (HDI) responsible for the highest burden of clinical harm, with specific attention to their molecular basis, pharmacokinetic and pharmacodynamic consequences, and practical implications for the prescribing clinician.
Methods: A narrative review of peer-reviewed literature was conducted, drawing on established pharmacokinetic interaction studies, controlled clinical trials, and case-report databases. Interactions were selected on the basis of mechanistic clarity, frequency of patient exposure, and clinical severity.
Conclusions: Each of the six interaction classes operates through a distinct molecular mechanism — nuclear receptor-mediated enzyme induction, irreversible mechanism-based inhibition, pharmacodynamic receptor antagonism, enzymatic blockade of cortisol inactivation, pharmacological duplication, and physicochemical chelation — and each carries the potential for life-threatening clinical consequences. Recognition requires targeted, mechanism-aware history-taking rather than generic supplement inquiry.
1. Introduction
The regulatory boundary separating dietary supplements from medicinal products is largely jurisdictional, not pharmacological. An increasing proportion of commercially available nutraceutical preparations — particularly those manufactured to standardised extract ratios (Drug-to-Extract Ratio, DER) — contain phytochemical constituents whose plasma concentrations, receptor affinities, and enzyme-modulating potencies are comparable to, and in some cases exceed, those of registered pharmaceutical agents. [^1]
Epidemiology consistently documents high rates of concurrent supplement and prescription drug use, with survey data from multiple healthcare systems indicating that between 30% and 70% of patients taking chronic medications also consume one or more botanical supplements, the majority of whom do not disclose this to prescribers. [^2] The clinical consequences are disproportionate when co-administered drugs have a narrow therapeutic index (NTI) — a category encompassing anticoagulants, immunosuppressants, antiepileptics, antineoplastic agents, and cardiac glycosides.
Herb-drug interactions are pharmacologically classifiable as pharmacokinetic (PK) — altering drug absorption, distribution, metabolism, or excretion — or pharmacodynamic (PD) — modifying the drug's effect at the receptor or effector organ level. Both categories can be clinically devastating. A comprehensive review of clinically documented HDIs confirmed that the majority of serious interactions involve modulation of cytochrome P450 (CYP) enzymes and/or the ATP-binding cassette transporter P-glycoprotein (P-gp), encoded by the ABCB1 gene. [^3]
The following six interaction types represent the most clinically significant and mechanistically well-characterised HDIs encountered in general medical practice. They are presented not as a comprehensive catalogue, but as exemplars of molecular mechanisms that a modern evidence-based prescriber must understand operationally.
2. Mechanism I — Hypericum perforatum (St. John's Wort): Pregnane X Receptor-Mediated Transcriptional Induction of Elimination Pathways
2.1 Epidemiological and Botanical Context
Hypericum perforatum (St. John's Wort, SJW) is one of the most widely consumed herbal products globally, used predominantly for self-management of mild to moderate depressive symptoms. Its over-the-counter availability ensures broad, often undisclosed use. Despite a favourable monotherapy tolerability profile, SJW represents, from a pharmacokinetic standpoint, one of the most consequential perpetrators of HDI currently available without prescription.
2.2 Molecular Mechanism
The primary bioactive constituent responsible for SJW's interaction profile is hyperforin, a phloroglucinol derivative present in varying concentrations across commercial preparations. Moore et al. demonstrated in 2000 that hyperforin is a potent ligand (Ki ≈ 27 nM) for the Pregnane X Receptor (PXR), an orphan nuclear receptor that functions as the master xenobiotic sensor in hepatocytes and enterocytes. [^4] Upon binding, hyperforin activates the PXR/Retinoid X Receptor (RXR) heterodimer, which translocates to the nucleus and drives transcriptional up-regulation (de novo protein synthesis) of CYP3A4, CYP2C9, and the efflux transporter P-glycoprotein. Crystallographic analysis has confirmed that hyperforin induces a substantial conformational expansion of the PXR ligand-binding pocket, facilitating high-affinity engagement. [^5]
Crucially, the magnitude of CYP3A4 induction correlates significantly with the hyperforin content of the specific preparation consumed, a relationship established across both commercial preparations and dry extracts (R = 0.87 for commercial preparations). [^6] Preparations with hyperforin content below 1% have not demonstrated clinically relevant enzyme induction, a finding with implications for the potential development of low-interaction SJW formulations.
Physiologically based pharmacokinetic (PBPK) modelling has further established that hyperforin's inductive effect is substantially greater in intestinal enterocytes than in hepatocytes — simulations suggesting up to 15.5-fold induction in intestinal CYP3A4 versus only approximately 1.1-fold in hepatic CYP3A4 at clinically encountered hyperforin doses — placing the primary site of interaction at the gut wall rather than the liver. [^7]
2.3 Clinical Pharmacokinetics and Consequences
The inductive mechanism operates through de novo enzymatic protein synthesis, which introduces two critical temporal characteristics. Maximal induction develops over 5–14 days of continuous SJW use, and — crucially — enzymatic activity remains elevated for up to 14 days following cessation, meaning drug toxicity from withdrawal of the inducer is as clinically important as the initial loss of efficacy. [^8]
Documented clinical consequences include: acute allograft rejection in cardiac and renal transplant recipients due to sub-therapeutic ciclosporin and tacrolimus plasma levels; loss of virological control in HIV-positive patients on protease inhibitor regimens; breakthrough seizures with reduced valproate or phenytoin exposure; unexpected thromboembolic events in patients anticoagulated with direct oral anticoagulants (DOACs) — rivaroxaban, apixaban, dabigatran — or vitamin K antagonists due to critically reduced area under the plasma concentration-time curve (AUC). Hormonal contraceptive failure has also been reported. [^9]
Of particular diagnostic relevance: a patient presenting with an apparently compliant medication profile and documented therapeutic drug monitoring results may have commenced SJW use in the interval between their last monitoring visit and the adverse event. The interaction should be considered in any unexplained loss of therapeutic effect for NTI drugs.
3. Mechanism II — Furanocoumarins (Grapefruit and Related Citrus): Irreversible Mechanism-Based Inhibition of Intestinal CYP3A4
3.1 Botanical Source and Clinical Underappreciation
Grapefruit (Citrus paradisi), pomelo (Citrus maxima), and bitter orange (Citrus aurantium) — a common ingredient in commercial thermogenic and weight-loss supplements — contain furanocoumarin derivatives, principally bergamottin and 6',7'-dihydroxybergamottin (DHB). This interaction class is widely cited yet persistently misunderstood by prescribers and patients, resulting in dangerous clinical guidance: specifically, the recommendation to temporally separate grapefruit consumption from drug administration.
3.2 Molecular Mechanism
Furanocoumarins are mechanism-based ("suicide") inhibitors of CYP3A4. They are activated by the very enzyme they subsequently inactivate: CYP3A4-mediated metabolism generates a reactive epoxide intermediate that forms a covalent adduct with the apoprotein at the active site of the enzyme, rendering it permanently non-functional. [^10] The furanocoumarin dimer components of grapefruit juice (e.g., paradisin A and GF-I-4) demonstrate particularly high inhibitory potency for CYP3A4 in vitro. The primary site of action is the brush-border enterocytes of the small intestine, where CYP3A4 normally provides a substantial first-pass metabolic barrier for lipophilic substrates.
3.3 Clinical Pharmacokinetics and Consequences
Because the inhibition is irreversible rather than competitive, the critical pharmacokinetic parameter is not the concentration of grapefruit components at any given moment but rather their cumulative access to intestinal enterocytes. The clinical implication is unambiguous: temporal separation of grapefruit consumption from drug administration offers no protection. Recovery of intestinal CYP3A4 activity requires the generation of new enzymatic protein, which follows the natural turnover of enterocytes — a process requiring approximately 72 hours. This is not a pharmacokinetic interaction that dissipates within hours.
Consequences for affected drug classes are severe: the peak plasma concentration (Cmax) of lipophilic statins (simvastatin, lovastatin, and to a lesser extent atorvastatin) may increase by several-fold, dramatically raising the risk of myopathy and rhabdomyolysis. Dihydropyridine calcium channel blockers (felodipine, amlodypine) exhibit similarly augmented bioavailability, producing hypotensive crises. Immunosuppressants, benzodiazepines, and certain antiretroviral agents are similarly affected. [^3]
For clinical practice, the only operationally sound guidance for patients on affected NTI drugs is complete avoidance of grapefruit, pomelo, and bitter orange, as well as supplements containing Citrus aurantium extract.
4. Mechanism III — Ginkgo biloba: PAF Receptor Antagonism, Pharmacodynamic Synergy, and the Diagnostic Trap of the Normal Coagulogram
4.1 Population Exposure
Ginkgo biloba extract (GBE) is one of the most frequently purchased dietary supplements among geriatric patients, used primarily for cognitive support and peripheral circulatory complaints. This population overlaps substantially with patients receiving antiplatelet therapy for cardiovascular or cerebrovascular disease.
4.2 Molecular Mechanism
The terpene lactone fraction of GBE — specifically ginkgolide B — functions as a specific, competitive antagonist of the membrane receptor for Platelet-Activating Factor (PAF). [^11] PAF receptor activation is one of several pathways contributing to platelet degranulation and primary aggregation. Ginkgolide B's receptor occupancy attenuates this activation signal. Separately, the bilobalide component of GBE has been shown in murine models to induce hepatic CYP enzymes, which may reduce the efficacy of simultaneously administered anticoagulants such as warfarin through a pharmacokinetic mechanism. [^11]
The clinical picture is further complicated by the evidence base regarding GBE's net haemostatic effect. Koch et al. reported that the concentrations of ginkgolide B required to inhibit human platelet PAF-mediated aggregation at standard therapeutic GBE doses are more than 100 times higher than measured peak plasma levels, raising questions about the clinical relevance of this mechanism at recommended doses. [^12] A systematic review and meta-analysis by Kellermann and Kloft similarly found no significant effect of standardised EGb 761 extract on validated haemostatic parameters. Conversely, a 2025 retrospective observational study of 2,647 hospital prescriptions found that GBE drug interactions, particularly with clopidogrel and aspirin, were associated with a statistically significant increase in both abnormal coagulation results and clinical bleeding events (OR 1.49 and 1.08, respectively). [^13]
4.3 The Diagnostic Trap
Regardless of the precise quantitative contribution of ginkgolide B to haemostatic impairment, the pharmacodynamic character of the interaction defines its diagnostic challenge. Both PAF antagonism and the incremental effects of GBE components act on primary haemostasis — platelet function — rather than on the coagulation cascade. Consequently, routine coagulation screening (prothrombin time, activated partial thromboplastin time, INR) will be entirely normal in a patient who is actively haemorrhaging due to GBE-potentiated antiplatelet therapy. This can induce false reassurance during peri-operative or emergency assessment. Platelet function analysis (e.g., PFA-100, aggregometry) is required if GBE-mediated haemostatic compromise is suspected.
In clinical practice, GBE should be discontinued at least one week prior to any surgical procedure, and its use in patients receiving dual antiplatelet therapy (DAPT) should be specifically discussed and documented.
5. Mechanism IV — Glycyrrhiza glabra (Liquorice): 11β-HSD2 Inhibition and Iatrogenic Apparent Mineralocorticoid Excess
5.1 Pharmacological Identity in the Supplement Landscape
Liquorice root extracts appear across multiple supplement categories — gastrointestinal protective formulations, antitussive agents, and adaptogenic blends — often without explicit disclosure of their mineralocorticoid activity potential. Unlike most HDI scenarios, the principal mechanism here bypasses drug metabolism entirely; it is an enzymatic blockade of a key glucocorticoid-inactivating step in the distal nephron.
5.2 Molecular Mechanism
Glycyrrhizic acid, the primary bioactive triterpenoid saponin in liquorice root, undergoes intestinal bacterial hydrolysis to its active metabolite 18β-glycyrrhetinic acid. This metabolite is a potent inhibitor of 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2), a renal tubular enzyme whose physiological function is to rapidly convert active cortisol to inactive cortisone at the level of the mineralocorticoid receptor (MR). [^14] This enzymatic barrier is the principal mechanism preventing cortisol — which is present in plasma at concentrations several orders of magnitude greater than aldosterone — from occupying and constitutively activating renal MRs.
When 11β-HSD2 is blocked, unmetabolised cortisol engages renal MRs with high affinity, producing a clinical syndrome biochemically indistinguishable from primary hyperaldosteronism but characterised — critically — by suppressed plasma renin and suppressed plasma aldosterone. This is the syndrome of Apparent Mineralocorticoid Excess (AME) or pseudohyperaldosteronism.
5.3 Clinical Consequences
The resulting phenotype is resistant hypertension, sodium and water retention, and hypokalaemia. The dose-dependency and reversibility of this effect following liquorice discontinuation have been confirmed in multiple case reports, and the clinical presentation is well characterised: a patient with previously controlled hypertension presenting with treatment-resistant blood pressure elevation and unexplained hypokalaemia. [^14]
Of particular concern is the compound pharmacological risk in patients receiving loop diuretics to manage the refractory hypertension: these agents, by promoting urinary potassium losses, will substantially deepen the pre-existing kaliuresis induced by mineralocorticoid receptor activation. The clinical sequelae of progressive hypokalaemia are especially dangerous in patients co-prescribed digoxin (where hypokalaemia potentiates glycoside-receptor binding and arrhythmogenicity) or class III antiarrhythmic agents such as amiodarone or sotalol (where hypokalaemia prolongs cardiac action potential duration and increases the risk of Torsade de Pointes). Assessment of plasma renin activity and aldosterone when investigating resistant hypertension should always be accompanied by a targeted supplement history, as the suppressed-renin/suppressed-aldosterone biochemical pattern is pathognomonic for this interaction and should prompt active liquorice enquiry.
6. Mechanism V — Red Yeast Rice (Monascus purpureus): Pharmacological Duplication and the GMP Deficit
6.1 Regulatory and Biochemical Identity
Red yeast rice (RYR) is marketed as a dietary supplement for lipid management, positioned in patient perception as a "natural" alternative to pharmaceutical statin therapy. This positioning constitutes a pharmacological misrepresentation with measurable clinical consequences.
6.2 Molecular Mechanism and Regulatory Problem
The primary active constituent of RYR is monacolin K, produced during fermentation of rice by Monascus purpureus. Monacolin K is not merely structurally similar to lovastatin — it is stereochemically identical to lovastatin, the registered pharmaceutical HMG-CoA reductase inhibitor. The mechanism of action — competitive inhibition of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), the rate-limiting enzyme in the cholesterol biosynthetic pathway — and its consequent systemic effects, including skeletal muscle toxicity mediated by mitochondrial dysfunction and coenzyme Q10 depletion, are therefore identical to those of the pharmaceutical product.
The critical distinction is regulatory rather than pharmacological: RYR supplements, unlike pharmaceutical statins, are not subject to mandatory GMP analytical quality control. Monacolin K content in commercially available RYR products varies enormously between batches and manufacturers, from trace quantities to amounts substantially exceeding pharmaceutical statin doses. This batch-to-batch variability makes dosing unpredictable and precludes safe use in patients who are simultaneously being titrated to a therapeutic statin dose.
6.3 Clinical Consequences
The most consequential clinical scenario is unrecognised pharmacological duplication: a patient prescribed atorvastatin for primary or secondary cardiovascular prevention who independently purchases RYR as a "complementary" measure is, in effect, receiving dual HMG-CoA reductase inhibition. The resultant myopathy — ranging from asymptomatic creatine kinase elevation to fulminant rhabdomyolysis with myoglobinuria and acute kidney injury — is clinically indistinguishable from pharmaceutical statin-induced myopathy and is frequently not attributed to the supplement.
An additional safety concern specific to RYR is contamination with citrinin, a nephrotoxic mycotoxin produced during suboptimal fermentation conditions by certain Monascus strains. Citrinin has demonstrated dose-dependent nephrotoxicity and genotoxicity in preclinical models, and its presence in commercial RYR supplements has been documented across multiple quality surveillance studies. Patients purchasing RYR from unregulated online sources are particularly exposed to this risk. Unexplained myalgia or elevated creatine kinase in a patient on statin therapy should prompt direct enquiry about RYR use.
7. Mechanism VI — Polyvalent Cation Chelation: Physicochemical Sequestration and Bioavailability Abrogation
7.1 Nature of the Interaction
Unlike the preceding mechanisms, polyvalent cation chelation operates entirely at the physicochemical rather than the biochemical level. It does not involve enzymatic, receptor, or transcriptional processes and is therefore unresponsive to any pharmacological intervention other than temporal separation sufficient to prevent simultaneous gastrointestinal occupancy.
7.2 Molecular Mechanism
In the acidic environment of the gastric and proximal intestinal lumen, drugs possessing specific oxygen or hydroxyl coordination sites — including fluoroquinolone antibiotics (ciprofloxacin, levofloxacin), tetracyclines, and levothyroxine — undergo coordinate bond formation with polyvalent metal cations commonly present in nutritional supplements: magnesium (Mg²⁺), calcium (Ca²⁺), iron (Fe²⁺/³⁺), and zinc (Zn²⁺). The resulting chelate complexes are bulky, electrically charged macromolecular structures that cannot traverse the lipophilic phospholipid bilayer of intestinal enterocytes. [^1]
The pharmacokinetic outcome is a dramatic reduction in oral bioavailability (F): studies have consistently documented reductions of 50–90% in fluoroquinolone and tetracycline absorption when co-administered with mineral-containing antacids or supplements. For levothyroxine, concomitant iron or calcium supplementation can reduce absorption sufficiently to produce clinically significant hypothyroidism, manifest as rising TSH despite apparent medication compliance.
7.3 Clinical Consequences and Temporal Requirements
For fluoroquinolone antibiotics, the practical consequence of chelation-mediated bioavailability reduction is the inability to achieve the minimum inhibitory concentration (MIC) necessary for bactericidal activity, effectively converting an antibiotic course into sub-therapeutic exposure and selecting for resistance. For levothyroxine, the interaction produces the clinical paradox of thyroid function tests that apparently indicate undertreated hypothyroidism despite nominal compliance with the prescribed dose.
The critical prescribing guidance is that temporal separation of 30 minutes before meals is insufficient. Literature consistently indicates that a separation interval of at least 4 hours is required to ensure that the drug and the chelating cation are not simultaneously present in the proximal gastrointestinal tract. This is particularly relevant for hospital inpatients receiving enteral mineral supplementation concurrently with fluoroquinolone or thyroid hormone regimens. The prescriber should explicitly document the required timing interval and should not assume that standard mealtime medication scheduling avoids this interaction.
8. Diagnostic Framework: Directed Supplement History-Taking
The six mechanisms described above share one avoidable point of failure: absence of an adequate supplement history at the prescribing encounter. Generic screening questions — "Do you take any supplements?" — are known to produce significant underreporting. Patients categorically differentiate "supplements" from "medicines" in their mental models, and non-disclosure is the norm rather than the exception.
An evidence-based clinical approach requires mechanism-directed enquiry:
- Before initiating or adjusting NTI drug therapy (anticoagulants, immunosuppressants, antiepileptics, antineoplastics): specifically ask about Hypericum perforatum (St. John's Wort), citrus-based supplements (Citrus aurantium/bitter orange), and any herbal products used for mood, sleep, or energy.
- Before any surgical, interventional, or haemostatic procedure: ask specifically about Ginkgo biloba, fish oil/omega-3 concentrates at supratherapeutic doses, and any supplements marketed for memory or circulation.
- In the workup of resistant hypertension with hypokalaemia: directly ask about liquorice-containing products, including digestive teas, confectionery-based preparations, and gastrointestinal supplements.
- In the workup of unexplained myalgia or creatine kinase elevation in a patient on statin therapy: ask directly about red yeast rice and any cholesterol-lowering supplements.
- When unexplained loss of antibiotic or hormone efficacy is suspected: ask about the timing of mineral supplement ingestion relative to medication administration, with specific inquiry about iron, calcium, magnesium, and zinc products.
9. Discussion
The clinical interactions described in this review share a common epistemic problem: their mechanisms operate silently, at the molecular level, without producing acute symptoms attributable to the supplement itself. The patient experiences either treatment failure or drug toxicity — both of which are attributed by default to the pharmaceutical regimen. This attribution error delays recognition, prolongs iatrogenic harm, and may prompt inappropriate dose escalation or regimen changes that further compound the problem.
Several conceptual shifts are required in clinical practice. First, the pharmacological distinction between "drug" and "supplement" must be abandoned at the level of prescriber cognition: both categories introduce biologically active molecules with defined receptor affinities, enzyme-modulating properties, and pharmacokinetic profiles. Second, the absence of a prescription for a potentially interacting agent does not constitute an absence of exposure. Third, the temporal logic of some interactions — particularly PXR-mediated induction and mechanism-based inhibition — extends significantly beyond the pharmacological half-life of either agent, requiring awareness that exposures occurring days to weeks prior to the adverse event remain causally relevant.
The evidence base for individual HDIs varies considerably in quality and clinical translatability. Interactions involving Hypericum perforatum/CYP3A4 and grapefruit furanocoumarins/CYP3A4 are supported by mechanistically convergent in vitro, animal, and human clinical data, including direct pharmacokinetic studies in relevant patient populations. [^8][^9] The haemostatic risk of Ginkgo biloba remains an area of genuine mechanistic debate: while PAF antagonism is biochemically established, its pharmacological relevance at recommended clinical doses is contested, with controlled studies and meta-analyses producing conflicting conclusions. [^12][^13] The literature on Glycyrrhiza glabra and 11β-HSD2 is mechanistically clear and supported by compelling clinical case series, but controlled dose-response data in the specific context of common supplement formulations remain limited. The red yeast rice/statin interaction is perhaps the most operationally straightforward: it is not a pharmacokinetic or pharmacodynamic interaction in the traditional sense, but a case of unrecognised pharmacological duplication whose risks are directly extrapolatable from the established statin literature. The chelation interaction is supported by robust clinical pharmacokinetic data for fluoroquinolones, tetracyclines, and levothyroxine.
10. Conclusion
The six interaction mechanisms characterised in this review — nuclear receptor-mediated enzyme induction (H. perforatum/PXR/CYP3A4), irreversible mechanism-based inhibition (furanocoumarins/CYP3A4), pharmacodynamic receptor antagonism and haemostatic potentiation (G. biloba/PAF), 11β-HSD2 blockade with apparent mineralocorticoid excess (G. glabra), pharmacological duplication and contamination risk (red yeast rice/monacolin K), and physicochemical chelation — each represent a distinct pharmacological pathway through which commercially available, non-prescription products can produce life-threatening modification of prescribed drug therapy.
Their integration into clinical decision-making does not require comprehensive knowledge of the entire herb-drug interaction literature. It requires, rather, a structured mechanistic framework that is applied at targeted decision points in pharmacotherapy: initiating NTI drug therapy, managing treatment-refractory presentations, preparing patients for procedures, and investigating unexplained laboratory or clinical anomalies.
The obligation to obtain this information rests with the prescriber. As nutraceutical standardisation continues to yield preparations of increasing pharmacological potency, and as polypharmacy and self-supplementation continue to increase in an ageing global population, the incorporation of phytopharmacological literacy into clinical practice is no longer discretionary.
1. Meng Q, Liu K. Pharmacokinetic interactions between herbal medicines and prescribed drugs: focus on drug metabolic enzymes and transporters. Current Drug Metabolism. 2015. [^1]
2. Izzo AA. Herb-drug interactions: an overview of the clinical evidence. Fundamental & Clinical Pharmacology. 2005. [^3]
3. Pal D, Mitra AK. MDR- and CYP3A4-mediated drug-herbal interactions. Life Sciences. 2006. [^4]
4. Moore LB, et al. St. John's wort induces hepatic drug metabolism through activation of the pregnane X receptor. PNAS. 2000. [^7]
5. Watkins RE, et al. Crystal structure of human PXR in complex with hyperforin. Biochemistry. 2003. [^5]
6. Gödtel-Armbrust U, et al. Variability in PXR-mediated induction of CYP3A4 by commercial preparations of St. John's Wort. Naunyn-Schmiedeberg's Archives of Pharmacology. 2007. [^6]
7. Nicolussi S, et al. Clinical relevance of St. John's Wort drug interactions revisited. British Journal of Pharmacology. 2019. [^8]
8. Adiwidjaja J, et al. Physiologically based pharmacokinetic modelling of hyperforin to predict drug interactions with St John's Wort. Clinical Pharmacokinetics. 2019. [^7]
9. Soleymani S, et al. Clinical risks of St John's Wort co-administration. Expert Opinion on Drug Metabolism & Toxicology. 2017. [^9]
10. He K, et al. Inactivation of cytochrome P450 3A4 by bergamottin, a component of grapefruit juice. Chemical Research in Toxicology. 1998. [^10]
11. Koch E. Inhibition of platelet activating factor-induced aggregation of human thrombocytes by ginkgolides. Phytomedicine. 2005. [^12]
12. Taki Y, et al. Ginkgo biloba extract attenuates warfarin-mediated anticoagulation through induction of hepatic CYP enzymes by bilobalide in mice. Phytomedicine. 2012. [^11]
13. Kellermann A, Kloft C. Is there a risk of bleeding associated with standardized Ginkgo biloba extract therapy? A systematic review and meta-analysis. Pharmacotherapy. 2011.
14. Ngo Thi Quynh Mai, et al. Impact of Ginkgo biloba drug interactions on bleeding risk and coagulation profiles. PLoS ONE. 2025. [^13]
15. Di Pierro F, et al. Liquorice-induced pseudohyperaldosteronism. Acta Bio-Medica. 2011. [^14]
16. Cho HJ, Yoon IS. Pharmacokinetic Interactions of Herbs with Cytochrome P450 and P-Glycoprotein. Evidence-Based Complementary and Alternative Medicine. 2015.
17. Cho HJ, Yoon IS. Pharmacokinetic Interactions of Herbs with Cytochrome P450 and P-Glycoprotein. Evid-Based Complement Alternat Med. 2015.
18. Kennedy DA, Seely D. Clinically based evidence of drug-herb interactions: a systematic review. Expert Opinion on Drug Safety. 2010. [^2]
This review synthesises the available evidence from a targeted literature search. The evidence base for individual interactions varies in study design and clinical translatability; deeper systematic review with extended analysis could surface additional controlled trial data for specific interaction pairs.
[^1]: Meng & Liu, 2015. Pharmacokinetic interactions between herbal medicines and prescribed drugs: focus on drug metabolic enzymes and transporters. Current drug metabolism.
[^2]: Kennedy & Seely, 2010. Clinically based evidence of drug–herb interactions: a systematic review. Expert Opinion on Drug Safety.
[^3]: Izzo, 2005. Herb–drug interactions: an overview of the clinical evidence. Fundamental & Clinical Pharmacology.
[^4]: Moore et al., 2000. St. John's wort induces hepatic drug metabolism through activation of the pregnane X receptor. Proceedings of the National Academy of Sciences of the United States of America.
[^5]: Watkins et al., 2003. 2.1 A crystal structure of human PXR in complex with the St. John's wort compound hyperforin. Biochemistry.
[^6]: Gödtel-Armbrust et al., 2007. Variability in PXR-mediated induction of CYP3A4 by commercial preparations and dry extracts of St. John’s wort. Naunyn-Schmiedeberg's Archives of Pharmacology.
[^7]: Adiwidjaja et al., 2019. Physiologically Based Pharmacokinetic Modelling of Hyperforin to Predict Drug Interactions with St John’s Wort. Clinical Pharmacokinetics.
[^8]: Nicolussi et al., 2019. Clinical relevance of St. John's wort drug interactions revisited. British Journal of Pharmacology.
[^9]: Soleymani et al., 2017. Clinical risks of St John’s Wort (Hypericum perforatum) co-administration. Expert Opinion on Drug Metabolism & Toxicology.
[^10]: He et al., 1998. Inactivation of cytochrome P450 3A4 by bergamottin, a component of grapefruit juice. Chemical Research in Toxicology.
[^11]: Koch, 2005. Inhibition of platelet activating factor (PAF)-induced aggregation of human thrombocytes by ginkgolides: considerations on possible bleeding complications after oral intake of Ginkgo biloba extracts. Phytomedicine.
[^12]: Kellermann & Kloft, 2011. Is There a Risk of Bleeding Associated with Standardized Ginkgo bilobaExtract Therapy? A Systematic Review and Meta‐analysis. Pharmacotherapy.
[^13]: Karthik et al., 2025. Liquorice-induced pseudohyperaldosteronism: a rare cause for severe hypertension. BMJ Case Reports.
[^14]: Cho & Yoon, 2015. Pharmacokinetic Interactions of Herbs with Cytochrome P450 and P-Glycoprotein. Evidence-Based Complementary and Alternative Medicine.
