Passionflower: Independent Evidence on Anxiety, Sleep & Safety

Table of contents

• Evidence summary
• What passionflower is
• How it works
• What works and what does not
• Benefits by claim
• Combination products: why their results are not passionflower's
• Risks and all side effects
• All interactions
• Who should avoid passionflower
• Doses studied and product quality
• Follow the money: who makes it and who funded the evidence
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

Independence tiers: 1 = regulator or government body; 2 = academic or Cochrane team with no declared commercial funding for the work; 3 = independent authors analysing mostly manufacturer-run trials, or minor declared industry ties; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D. In this article, tier 3 also covers trials that used a commercial product but whose funding statement we could not access.

Sources cluster by country. The clinical trials come mainly from Iran (five of the nine trials in Janda 2020) and Brazil. The leading extracts are made in Germany and Switzerland, and the regulators we relied on are in the EU and North America. No large, independent, multi-country trial exists.

What passionflower is

Medicinal passionflower is the dried above-ground part of Passiflora incarnata L., a climbing vine native to the south-eastern United States and Central and South America. It is also called maypop. It has traditionally been used as a sedative; Spanish explorers encountered it in the 16th century, and it later entered European folk medicine (NCCIH).

• Plant part: the EU monograph covers only the aerial parts of P. incarnata, as tea, powder, or liquid and dry extracts (EU monograph). The European Pharmacopoeia requires at least 1.5% total flavonoids, expressed as vitexin (HMPC assessment report).
• Not the same as passion fruit: passion fruit is mainly P. edulis. Brazil's public health system has studied a third species, P. alata (Barros 2026). Results from one species cannot be transferred to another.
• Common forms: teas, tinctures and drops, tablets of dry extract (for example 425 mg tablets), and many combination products with valerian, hops, lemon balm or St John's wort.

Legal status by country

How it works

No one knows for certain. The EU assessors concluded that the mechanism of action is not yet clarified, although recent animal work suggests effects on the GABA system, the same calming signalling system that benzodiazepines act on (HMPC assessment report).

• Constituents: mainly flavonoids such as isovitexin, isoorientin, schaftoside and vitexin, plus traces of essential oil, a cyanogenic glycoside (gynocardin), and maltol, which may be a processing artefact. β-carboline (harmala-type) alkaloids such as harman may be present in traces but are undetectable in most commercial material (HMPC).
• A disputed active compound: a benzoflavone derivative once proposed as the key ingredient could not be found by a later team in Indian or French plant material, and only in traces in Italian material (HMPC).
• Other proposals: Memorial Sloan Kettering's herb database describes possible MAO inhibition by the alkaloids and GABA activation by maltol-type compounds and chrysin (MSK About Herbs). These are laboratory findings.

What is proven in humans: very little. The HMPC found no human pharmacokinetic or pharmacodynamic data. In one small volunteer study, passionflower's effect on brain-wave (EEG) recordings could not be told apart from placebo (HMPC, citing Schulz 1998). A later EEG study that reported calming-type brain-wave patterns tested a three-herb tablet, not passionflower alone, and a co-author works for its maker, Pascoe (Dimpfel 2011). The animal doses that showed effects appear high compared with human use (HMPC).

What works and what does not

Benefits by claim

Passionflower for anxiety disorders (GAD)

The Cochrane review searched for randomised trials in any anxiety disorder and found two, with 198 participants in total. One suggested no difference in effectiveness between passionflower and a benzodiazepine, with similar dropout rates. Differences in job performance and drowsiness were reported but did not reach statistical significance. The authors concluded that trials were too few for any conclusions and called for larger placebo-controlled studies (Miyasaka et al., Cochrane 2007; plain-language summary).

The two trials were:

• Akhondzadeh 2001 (Iran): 36 outpatients with GAD received 45 drops a day of passionflower extract or oxazepam 30 mg a day for four weeks. Both groups improved, with no significant difference at the end. Oxazepam worked faster, and the authors reported more problems with job performance on oxazepam (Akhondzadeh 2001). The design had no placebo-only arm: each drug was matched with a placebo for the other.
• Mori 1993 (Japan): a 28-centre double-blind comparison of an extract called Passiflamin with the benzodiazepine mexazolam (63 vs 71 patients). Passiflamin improved four of eight neurotic-symptom items; mexazolam improved all eight (HMPC summary).

A 2022 network meta-analysis of 29 trials across 12 herbs estimated passionflower's effect on the Hamilton Anxiety scale (HAM-A) at −4.20 points, but the 95% credible interval ran from −8.82 to +0.16, which includes no effect. The authors judged it worth further study (Zhang 2022). In plain terms: the best guess is a modest benefit, but the data are consistent with none.

Passionflower before surgery or dental treatment

This is where most of the trials are. They test a single dose 30 to 90 minutes before a procedure.

• Movafegh 2008 (Iran): 60 day-surgery patients; 500 mg (commercial extract) or placebo 90 minutes before. Anxiety scores were lower with passionflower (P < 0.001) (Movafegh 2008).
• Aslanargun 2012 (Turkey): 60 patients before spinal anaesthesia; passionflower syrup or placebo. Passionflower suppressed the rise in state anxiety, with no change in psychomotor tests, sedation or blood pressure (Aslanargun 2012). The EU assessors described the difference as statistically significant but small, and noted the placebo (mineral water) did not match the syrup's taste and colour (HMPC).
• Azimaraghi 2017 (Iran): 128 day-surgery patients; passionflower vs oxazepam 10 mg. Anxiety at 90 minutes was lower with passionflower, with similar psychomotor effects. Despite its title, the published abstract describes no placebo arm (Azimaraghi 2017).
• Rokhtabnak 2017 (Iran): 52 surgical patients; 1,000 mg passionflower vs melatonin 6 mg. Both reduced anxiety, but passionflower caused more cognitive impairment on a digit-symbol test (Rokhtabnak 2017).
• Kaviani 2013 (Iran): 63 periodontal patients; passionflower drops vs placebo vs nothing. Dental anxiety scores fell from 12.09 to 8.47 with passionflower, against 12.00 to 10.52 on placebo. The trial was only single-blind (Kaviani 2013).
• Dantas 2017 (Brazil): 40 volunteers, crossover, passionflower compared with midazolam for dental extraction; 20% of the midazolam group reported amnesia. Competing interests and funder were not stated in the record we could read.
• da Cunha 2021 (Brazil): 200 people having wisdom teeth removed; 500 mg passionflower, 500 mg mulungu, midazolam 15 mg or placebo. Passionflower was similar to midazolam and better than placebo (da Cunha 2021).
• Velasquez 2024 and Lima Neto 2025 (Brazil): one 54-person trial found anxiety fell with passionflower or valerian but not placebo (Velasquez 2024). A 30-person, publicly funded trial found no difference between passionflower, midazolam and placebo on anxiety questionnaires, though salivary cortisol was lower after surgery with passionflower or midazolam (Lima Neto 2025).

Two recent systematic reviews summarised dental trials in words rather than pooling them (Araújo 2021; Stonkutė 2026). In PubMed and Europe PMC searches in October 2026 we did not find a meta-analysis that pools passionflower's preoperative trials. The direction is fairly consistent, but every trial is small, the products and doses differ, and a finding of "similar to midazolam" in a small trial does not show equivalence.

Passionflower for sleep and insomnia

In the Australian tea study, 41 healthy adults aged 18 to 35 drank one cup of passionflower tea (2 g) or a placebo tea nightly for a week each. Of six sleep-diary measures, only sleep quality was better with passionflower. Overnight recordings in 10 participants showed no differences (Ngan & Conduit 2011; HMPC).

In the Korean trial, 110 adults with diagnosed insomnia disorder took a passionflower extract or placebo for two weeks. Total sleep time on polysomnography was higher with passionflower than placebo at P = 0.049; we could not verify the effect size from the abstract. We could not confirm the other sleep measures from the abstract (Lee 2020). A P-value this close to 0.05 from one trial needs replication. NCCIH summarises the sleep evidence as showing passionflower might improve total sleep time, with mixed effects on falling and staying asleep (NCCIH).

An Indian trial of a branded extract (SIVI, 600 mg at bedtime for 30 days, 65 people) reported less perceived stress and longer self-reported sleep than placebo. It was funded by the extract maker, JK Botanicals, and two authors were its employees (Harit 2024).

Passionflower for everyday stress

This is the use regulators accept, but on tradition alone. The HMPC said the clinical studies support the plausibility of traditional use, not proof of efficacy (HMPC). A 12-week German study of a Pascoe passionflower dry extract in 154 adults with nervous restlessness reported improvements in stress resistance and quality of life, but it had no control group and was funded by Pascoe, with two Pascoe employees as authors (Gibbert 2017). Uncontrolled before-and-after studies cannot separate a treatment effect from natural recovery and expectation. NCCIH lists stress among conditions without enough evidence to judge (NCCIH).

Combination products: why their results are not passionflower's

A large share of passionflower is sold in mixtures. The EU assessors noted that many products on the market are combinations (HMPC), and Janda's review excluded 21 multi-herb studies for that reason (Janda 2020).

• Ze 185 (Relaxane), Max Zeller Söhne / Zeller Medical AG, Switzerland: each tablet contains butterbur root 90 mg, valerian root 90 mg, passionflower herb 90 mg and lemon balm leaf 60 mg (dry extracts) (Keck 2020). In 182 people with somatoform disorders, the four-herb mix beat both a three-herb mix without butterbur and placebo over two weeks (Melzer 2009). That design isolated butterbur's contribution, not passionflower's. In 72 healthy men, Ze 185 reduced self-reported anxiety under a lab stress test but not cortisol; three co-authors worked for Zeller (Meier 2018). A retrospective study of 3,252 psychiatric inpatients linked Ze 185 to fewer benzodiazepine prescriptions; four of six authors were Zeller employees, although the paper states there was no manufacturer funding (Keck 2020). Butterbur itself needs careful processing to remove liver-toxic pyrrolizidine alkaloids (LiverTox: butterbur).
• Euphytose, Bayer HealthCare SAS, France: the 1997 trial in 182 people with adjustment disorder tested six extracts. At day 28, 42.9% had a HAM-A score below 10, against 25.3% on placebo (Bourin 1997). We could not verify the composition of the Euphytose product sold today, so the trial product should not be assumed to match it.
• Pascoe's St John's wort mixtures: Neurapas balance combines St John's wort, valerian and passionflower, and Pascoe holds a UK registration for Botanova with the same three extracts at the same strengths (Dimpfel 2011; MHRA THR list). These products therefore carry St John's wort's interaction risks.

To credit passionflower, a trial would need a passionflower-only arm, or a factorial design that adds and removes it. None of the combination trials we found did this. A positive combination trial tells you about that exact mixture at that exact dose, and nothing more.

Risks and all side effects

Other sources are less reassuring:

• Drowsiness, dizziness and confusion are possible (NCCIH). Excessive doses may cause sedation (HMPC).
• Thinking speed: one trial found more short-term cognitive impairment after passionflower than after melatonin (Rokhtabnak 2017).
• Label-listed effects: the German maker of Pascoflair lists stomach and gut upset, allergic skin reactions and a faster heart rate, frequency unknown (Pascoe product information).
• Liver: LiverTox says passionflower "has not been implicated" in liver injury (LiverTox).

All interactions

We found no formal interaction studies in the sources we could read, which is not proof of no interactions.

Who should avoid passionflower

• Pregnancy: NCCIH says it should not be used, as it may induce uterine contractions (NCCIH). Swedish product information advises women who are pregnant or breastfeeding to avoid it.
• Breastfeeding: not recommended in the EU; NCCIH says little is known.
• Children under 12: not recommended (Swedish product information).
• Before surgery: tell your surgeon or anaesthetist if you have taken it in the previous two weeks (NCCIH).
• Drivers and machine operators: it may impair ability to drive (EU monograph 4.7).
• People taking sedatives or drinking alcohol, as above.
• People with a heart-rhythm disorder or on QT-prolonging drugs: the evidence is one case report, but it is worth raising with a pharmacist.
• Allergy: hypersensitivity is the only formal EU contraindication.
• People avoiding alcohol: many liquid extracts are ethanol-based (EU monograph).
• Liver disease: no signal for passionflower itself (LiverTox), but check combination products for butterbur.
• Anyone with persistent symptoms: the EU advises seeing a doctor if symptoms last more than two weeks or worsen.

Doses studied and product quality

The figures below are doses studied in trials or stated in official monographs and registrations. They are not personal dosing advice.

Milligram doses are not comparable across products. A "500 mg" tablet may be raw powder or a concentrated extract, and several trials did not say which. The HMPC noted that the type of extract in the Movafegh tablets was not reported (HMPC), and an Italian academic review called poor description of extract ratios a crucial weakness across trials (Miroddi 2013).

Product quality issues

• Chemotypes: Australian researchers found two chemically distinct types of P. incarnata. They advised using the established isovitexin-rich type until the effects of the difference are understood (Wohlmuth 2010).
• Species and plant part: supplements may contain other Passiflora species or fruit rather than leaf and stem. An FDA-funded lab developed a method to tell P. incarnata apart from other species and from Peganum harmala, a harmala-alkaloid plant, in supplements (Avula 2012). The Norwegian product linked to hospital admissions was made mainly from the fruit (Solbakken 1997).
• Regulated versus unregulated: EU and UK registered products must meet pharmacopoeia and good manufacturing standards. The Swedish assessment of Pascoflair, for example, checked plant source, heavy metals, pesticides and an Ames mutagenicity test (Swedish MPA). US dietary supplements are not approved before sale (NCCIH).

Follow the money: who makes it and who funded the evidence

Who makes and sells it

We found no citable market-size figure for passionflower specifically. In the US, products are sold by many supplement brands without pre-market approval.

Who funded the evidence

• Independent and public: the Cochrane review (funding not visible to us), Janda 2020 (Polish government programme), Zhang 2022 (funder not verified), Ngan & Conduit 2011 (Monash University), Lima Neto 2025 (Brazil's CNPq), Rokhtabnak 2017 (an Iranian hospital research centre) and Kaviani 2013 (Isfahan university research funds). Several of these trials still used commercial or pharmacy-made products.
• Unknown: we could not access funding statements for Akhondzadeh 2001, Movafegh 2008, Aslanargun 2012, Lee 2020, da Cunha 2021, Melzer 2009 or Bourin 1997. Dantas 2017: competing interests and funder not stated in the record we could read.
• Maker-linked: Pascoe funded the Gibbert 2017 study and a Pascoe employee is a co-author of the Dimpfel 2011 EEG study (Dimpfel 2011). Zeller staff co-authored the 2018 and 2020 Ze 185 papers. JK Botanicals funded Harit 2024.
• The watchdogs' own money: EMA is mostly funded by fees from companies (EMA), but HMPC monographs are not company applications. NCCIH is funded by Congress (NCCIH) and cites the commercial NatMed Pro database. ESCOP, a European phytotherapy federation, sells its 2023 passionflower monograph (ESCOP).

Documented integrity events

• In 2017 the FDA warned Duncan's Botanical Products (US) that claims for its Maypop-Passion Flower product, including anxiety and high blood pressure, made it an unapproved new drug (FDA warning letter).
• In 2024 an FDA advisory letter to Galpin Naturals (US) listed a passion flower tincture among products sold with disease-treatment claims (FDA advisory letter).
• We found no recalls, warning letters or enforcement actions involving the main European makers' passionflower products in our searches.

What this means: passionflower's regulatory standing rests on tradition, and its clinical evidence rests mostly on small academic trials from Iran and Brazil that used local commercial products. The most commercially important European extracts have very little independent placebo-controlled testing of their own; the main published work on them is maker-funded or maker-co-authored. Neither fact shows passionflower does not work. Together they show that nobody has yet paid for a large, independent trial that could settle the question.

Related research

For wider, evidence-graded context, see these Pure City Research guides:

• Valerian: sleep evidence, safety and funding limits
• Chamomile for sleep and anxiety
• Oral lavender and Silexan
• St John's wort: depression evidence and interaction risks
• L-theanine
• Stress, anxiety and depression supplements
• Sleep supplements: evidence limits, product quality and safety
• Generalised anxiety disorder
• Insomnia
• CBT-I: cognitive behavioural therapy for insomnia
• Lorazepam
• Sleep and pregnancy

Frequently asked questions

Sources and funding notes

What we could not verify: PubMed pages were blocked to our checks, so we used Europe PMC records, and publisher or PMC full texts where they opened. We could not open full texts or funding statements for Akhondzadeh 2001, Movafegh 2008, Aslanargun 2012, da Cunha 2021, Melzer 2009 or Bourin 1997. For Lee 2020 we could read the sample size and P value only, so the effect size is not given. The Cochrane review's funding and affiliations were not visible on the page we could open. Zhang 2022 funders and author affiliations were not verified, and its full text is paywalled. We could not read the HMPC assessment report or the EU monograph text, because the PDFs gave our tools no readable text. For that reason we removed the EU claims about 'pilot' wording, 'no known undesirable effects', interaction statements and dosing tables. The driving warning (EU monograph section 4.7) was checked against a copy of the monograph text and kept. We could not verify the Health Canada dosing range, so we removed it; the Health Canada indication is limited to restlessness or nervousness. The Swedish registration (2011, traditional use) was checked in search-indexed registration documents. We could not read the current UK list entries for passionflower, so UK product names were removed. The Ze 185 maker (Max Zeller Söhne, Switzerland) and the six-plant Euphytose composition were checked. We did not open the original 1985 German Commission E text or the paywalled 2023 ESCOP monograph. The outcome of the HMPC's 2025 periodic review was not published when we checked. We found no citable market-size data.

1. EMA: Passiflorae herba, herbal medicinal product page — EU regulator summary and document list; EMA mostly fee-funded.
2. HMPC Community herbal monograph on Passiflora incarnata L., herba (EMA/HMPC/669740/2013, 2014) — EU regulator.
3. HMPC Assessment report on Passiflora incarnata L., herba (EMA/HMPC/669738/2013) — EU regulator; critique of trials and safety data.
4. EMA: Funding — 91.5% fees and charges in 2026 budget.
5. NCCIH: Passionflower (updated April 2025) — US government; cites NatMed Pro (commercial).
6. NCCIH: Funding appropriations history — US Congress.
7. LiverTox: Passionflower (2020) — NIDDK, US government.
8. LiverTox: Butterbur (2019) — NIDDK, US government.
9. Health Canada: Passionflower natural health product monograph (29 November 2024) — Government of Canada.
10. MHRA: Herbal medicines granted a traditional herbal registration (updated 28 October 2025) — UK regulator.
11. Swedish Medical Products Agency: Public assessment report, Pascoflair — Swedish regulator; applicant Pascoe.
12. Base de données publique des médicaments: Euphytose — French government database; holder Bayer HealthCare SAS.
13. 21 CFR 172.510: Natural flavoring substances — US federal regulation.
14. FDA warning letter: Duncan's Botanical Products (8 May 2017) — US regulator.
15. FDA advisory letter: Galpin Naturals (2024) — US regulator.
16. Miyasaka LS et al. Passiflora for anxiety disorder. Cochrane 2007 (summary) — Brazil-based authors; funding not visible on the page we could open.
17. Janda K et al. Passiflora incarnata in neuropsychiatric disorders: a systematic review. Nutrients 2020 — Poland; Ministry of Science and Higher Education; no conflicts.
18. Zhang W et al. Medicinal herbs for the treatment of anxiety: network meta-analysis. Pharmacol Res 2022 — funder not verified.
19. Araújo JO et al. Oral sedation in dental procedures: systematic review. BMJ Open 2021 — Brazil; no specific grant; no competing interests.
20. Stonkutė et al. Preoperative anxiolysis in surgical care: systematic review. Dent J 2026 — Lithuania; university funding; no conflicts.
21. Miroddi M et al. Passiflora incarnata L.: evaluation of clinical trials. J Ethnopharmacol 2013 — Italy (University of Messina); funding not stated in abstract.
22. Akhondzadeh S et al. Passionflower vs oxazepam in GAD. J Clin Pharm Ther 2001 — Iran; funding not accessible.
23. Akhondzadeh S et al. Passionflower in opiate withdrawal. J Clin Pharm Ther 2001 — Iran; funding not accessible.
24. Movafegh A et al. Preoperative oral Passiflora incarnata. Anesth Analg 2008 — Iran; funding not accessible.
25. Aslanargun P et al. Passiflora before spinal anesthesia. J Anesth 2012 — Turkey; funding not accessible.
26. Azimaraghi O et al. Passiflora and oxazepam before ambulatory surgery. Asian J Pharm Clin Res 2017 — Iran; no funding statement seen.
27. Rokhtabnak F et al. Melatonin vs Passiflora premedication. Anesth Pain Med 2017 — Iran; Firoozgar clinical research development centre.
28. Kaviani N et al. Passion flower for dental anxiety. J Dent (Shiraz) 2013 — Iran; Isfahan university research funds.
29. Dantas LP et al. Passiflora vs midazolam in dental extraction. Med Oral Patol Oral Cir Bucal 2016/2017 (sources differ on year) — Brazil; competing interests not stated in the record we could read; no funder named.
30. da Cunha RS et al. Herbal anxiolytics before third molar extraction. Clin Oral Investig 2021 — Brazil; funding not accessible.
31. Velasquez ACA et al. Passiflora and valerian in tooth extraction. Oral Maxillofac Surg 2024 — Brazil; funding not accessible.
32. Lima Neto TJ et al. Passiflora, salivary biomarkers and anxiety. Sci Rep 2025 — Brazil; CNPq and PIBIC; competing interests not stated in the record we could read.
33. Ngan A, Conduit R. Passionflower tea and sleep quality. Phytother Res 2011 — Australia; Monash University.
34. Lee J et al. Passiflora and polysomnography in insomnia disorder. Int Clin Psychopharmacol 2020 — South Korea; funding not verified.
35. Harit MK et al. Passiflora extract (SIVI) in stress and sleep. Cureus 2024 — India; funded by JK Botanicals.
36. Gibbert J et al. Passion flower dry extract and nervous restlessness. Complement Med Res 2017 — Germany; funded by Pascoe.
37. Dimpfel W et al. NEURAPAS balance and EEG. BMC Psychiatry 2011 — Germany; Pascoe co-author initiated the study.
38. Melzer J et al. Ze 185 in somatoform disorders. Phytother Res 2009 — Switzerland; funding not accessible.
39. Meier S et al. Ze 185 in acute stress. Phytomedicine 2018 — Switzerland; three Zeller co-authors.
40. Keck ME et al. Ze 185 and benzodiazepine prescribing. Phytother Res 2020 — Switzerland; four Zeller co-authors; states no manufacturer funding.
41. Bourin M et al. Plant-extract combination (Euphytose) in adjustment disorder. Fundam Clin Pharmacol 1997 — France; funding not accessible.
42. Fisher AA et al. Toxicity of Passiflora incarnata L. J Toxicol Clin Toxicol 2000 — case report; country not verified.
43. Smith GW et al. Vasculitis with a Passiflora preparation. Br J Rheumatol 1993 — UK; case report.
44. Solbakken AM et al. Nature medicine as intoxicant. Tidsskr Nor Laegeforen 1997 — Norway; case series.
45. Carrasco MC et al. Valerian and passionflower with lorazepam. Phytother Res 2009 — Spain; case report.
46. Sato AC et al. Passiflora edulis juice and blood coagulation. Protein Pept Lett 2012 — Brazil; laboratory study.
47. Avula B et al. Alkaloids and flavonoids in Passiflora species and supplements. Nat Prod Commun 2012 — US; FDA grant U01FD002071.
48. Wohlmuth H et al. Chemotypes of passionflower. Biol Pharm Bull 2010 — Australia (Southern Cross University).
49. Barros TB et al. Passiflora alata in anxiety. Daru 2026 — Brazil; no competing interests.
50. Memorial Sloan Kettering: About Herbs, Passionflower (2023) — US cancer centre database.
51. A.D.A.M. consumer health library: Passionflower interactions (reviewed 2007) — commercial content provider, hosted by a US hospital.
52. Altmeyer's Encyclopedia: Passiflorae herba — Springer-affiliated reference; Commission E indication.
53. ESCOP: Passiflorae herba monograph (2023) — European phytotherapy federation; paid monograph.
54. Pascoe: Pascoflair product information — manufacturer.
55. apotheke adhoc: Uriach buys Pascoe (May 2024) — German pharmacy trade news.
56. Schwabe group: About us — manufacturer.
57. Swiss SME portal: interview with Max Zeller Söhne — Swiss government portal.
58. Historical Dictionary of Switzerland: Max Zeller — Swiss academic reference.

Educational research reviewed 6 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.