Licorice (Glycyrrhiza glabra)

Free to every reader. The safety section of every monograph is open to everyone, signed in or not. In an emergency call 000. For poisoning, call the Poisons Information Centre on 13 11 26.

15. Safety

15.1 The dose-response for harm

15.2 Adverse effects

From trials. The meta-analytic findings at 13.3 are themselves the adverse-effect data: systolic +5.45 mm Hg, diastolic +3.19, potassium −0.33 mmol/L, renin and aldosterone suppressed [38,39,40].

From case reports and series (each with the causality judgement stated):

TGA safety alerts, recalls and DAEN. Medicine attempted all three Australian sources in September 2026 and reports honestly what each returned.

• TGA safety alerts. No safety alert, product alert or hazard communication for Glycyrrhiza glabra, licorice or glycyrrhizin was located. The TGA website’s own search sits behind a bot-challenge interstitial that refuses automated requests, so the finding rests on an external index of tga.gov.au rather than on the site search, and it is a negative finding of the weakest kind: no alert was found; that is not the same as confirming none exists.
• Database of Recalls, Product Alerts and Product Corrections (DRAC). Searched for licorice, liquorice and glycyrrhiza in September 2026. The database’s search only accepts a product or ingredient chosen from its own suggestion widget, which does not respond to an automated request, so no result set was obtained. DRAC covers actions from 01/07/2012 onward.
• Database of Adverse Event Notifications (DAEN). The medicines search is a Microsoft Power Pages application that will not return a result set outside an interactive browser session, and no downloadable DAEN dataset exists on data.gov.au. DAEN was therefore not searched successfully, and the gap stays open at 20.5. A manual browser search is a ten-minute job for whoever next has the draft in front of them, and it must be done before release.

Nothing in this section turns on the Australian post-market data, because the harm is established from controlled trials rather than from spontaneous reports. For comparison, the EMA searched EudraVigilance and found 282 cases under "liquorice root extract", "Glycyrrhiza glabra L. radix" and "Glycyrrhiza uralensis root" out of 1,733 reports retrieved [8].

15.3 Contraindications

15.4 Herb-drug interactions, graded

15.5 Pregnancy and lactation

Pregnancy: the strongest non-cardiovascular safety signal this herb has.

• The foetus is shielded from the 5- to 10-fold rise in maternal cortisol by highly expressed placental 11β-HSD2. In dually perfused human placentas, inhibiting 11β-HSD with glycyrrhetinic acid allowed cortisol direct access to the foetal circulation [8].
• In 1,049 Finnish women, heavy glycyrrhizin exposure (≥ 500 mg/week) did not significantly affect birth weight or maternal blood pressure but was significantly associated with lower gestational age: odds ratio 2.5 (95% CI 1.1-5.5) for birth before 38 weeks, with a mean shortening of 2.52 days [46].
• In a follow-up case-control study, heavy consumption was associated with a more than two-fold increased risk of preterm delivery before 37 weeks, and an odds ratio of 3.07 (95% CI 1.17-8.05) in the fully adjusted model for early preterm delivery before 34 weeks [47].
• A follow-up of 321 of those children at about eight years of age found that the high-exposure group showed deficiencies in verbal and visuospatial ability and narrative memory, and more externalising, attention, rule-breaking and aggressive behaviour, dose-relatedly; the same children had higher salivary cortisol [10]. Räikkönen and colleagues re-examined 378 of the cohort at a mean age of 12.5 years and found the cognitive and behavioural findings still there — 7 IQ points lower (95% CI 3.1 to 11.2), poorer memory, 3.3-fold odds of ADHD-type problems — together with taller stature, higher BMI-for-age and more advanced puberty in the exposed girls, but no difference in cortisol at that age [48]. The cortisol signal at eight years and its absence at twelve are both reported here as the two follow-ups found them; they are not reconciled, and the endocrine mechanism in the child is unsettled.
• The EMA’s position, adopted 04/03/2026: "The use is not recommended during pregnancy and in women of childbearing potential not using contraception." Animal studies show reproductive toxicity: 18β-glycyrrhetinic acid crosses the rat placenta, and glycyrrhizin at 239-1,000 mg/kg from day 7 of pregnancy was minorly embryotoxic to rat foetus kidney and lung [7].
• Mills and Bone assign Category C (has caused, or may be suspected of causing, harmful effects on the foetus or neonate without causing malformations) and add that doses up to 3 g/day of root are likely to be safe for short periods [10]. Medicine’s ruling, 23/09/2026 (Dr Eleanor Whitcombe, Head of Medicine): the EMA position stands and the 3 g/day allowance is not carried into this monograph. The reason is arithmetic rather than preference. The Finnish signal attaches to a heavy exposure of 500 mg glycyrrhizin a week, which is about 71 mg a day [46,47]. Root that meets the Ph. Eur. minimum of 4.0 per cent glycyrrhizin delivers, at 3 g a day, about 120 mg a day [3]; at the 6 per cent upper end of the range Mills and Bone give, about 180 mg [10]. A 3 g/day allowance therefore permits an intake well above the exposure associated with preterm delivery, and above the 100 mg ceiling at which a randomised crossover trial has already measured a pressor effect in healthy young adults [40]. Mills and Bone’s own sentence is hedged ("likely to be safe for short periods") and carries no supporting study. Braun and Cohen, writing for the same Australian readership, say licorice is contraindicated in pregnancy without qualification [11]. Licorice is not prescribed in pregnancy at AIP, in any preparation containing glycyrrhizin, at any dose, and the incidental exposure from a flavouring quantity in a formula is avoided as well. Deglycyrrhizinated licorice is a separate question on which there is no pregnancy data either way, and it is not recommended in the absence of any.
• Women with or prone to pre-eclampsia should avoid licorice: 11β-HSD activity is significantly lower in pre-eclamptic placentas and umbilical cord cortisol higher, with decreased foetal weight [10].

Lactation. Mills and Bone state that licorice is compatible with breastfeeding [10]. The EMA does not agree: glycyrrhizin is detectable in the breastmilk of some women taking liquorice, no study has measured glycyrrhetinic acid in milk, the effect on the infant is unknown, and use during breastfeeding is not recommended [7,8]. Medicine’s ruling, 23/09/2026: the EMA position is adopted and licorice is not recommended in lactation. The conflict is real and stays printed. Mills and Bone’s "compatible with breastfeeding" is a bare assertion with no study behind it in their text [10]; the EMA’s position rests on the finding that glycyrrhizin is detectable in the breastmilk of some women taking liquorice, on the absence of any measurement of glycyrrhetinic acid in milk, and on the absence of any infant data [7,8]. Where one source asserts safety without evidence and the other declines to assert it because the measurements have not been made, Medicine follows the one that matches the state of the evidence. MediHerb’s own guide says only that "professional supervision is suggested for pregnancy and lactation" [80], which is a supervision instruction and not a safety finding, and it does not change the ruling.

15.6 Children, older people, overdose

Children and adolescents. The EMA: use under 18 years is not recommended, because it has not been established for want of adequate data [7]. WHO: safety and effectiveness in children have not been established [9]. Mills and Bone warn that children like licorice confectionery and that its use as a flavouring in tinctures should be moderate [10]. A child with Addison’s disease consuming 300-400 g/day of confectionery developed reversible growth retardation [10].

Older people. Special precautions in the elderly, and in anyone with hypertension or cardiac, renal or hepatic disease; do not give high-glycyrrhizin preparations for prolonged periods, and monitor closely [10]. MediHerb’s own guide says "exercise care with elderly patients" [80].

Overdose. The EMA records overdose with prolonged use beyond 4 weeks or high intake, presenting as water retention, hypokalaemia, hypertension, cardiac rhythm disorders and hypertensive encephalopathy [7]. Five reported cases link high-dose ingestion to transient visual loss [10,11].

Toxicology. Acute intraperitoneal LD₅₀ for glycyrrhetinic acid in mice 308 mg/kg; oral LD₅₀ greater than 610 mg/kg. Little short-term, subchronic or chronic toxicity in rats given glycyrrhizin salts; genotoxic and carcinogenic effects have not been observed for glycyrrhizin derivatives [10]. The EMA notes that adequate tests on reproductive toxicity, genotoxicity and carcinogenicity have not been performed for the herbal preparations themselves [7]. A 2017 review gathers the acute, sub-acute, sub-chronic and chronic toxicity work on G. glabra and glycyrrhizin salts in one place and reaches the same practical conclusion as this monograph: the effects that matter in people are hypertension and the consequences of hypokalaemia, and the risk rises with prolonged gastrointestinal transit, reduced 11β-HSD2 activity, existing hypertension, anorexia nervosa, older age and female sex [64].

15.7 Practical risk reduction

1. Prescribe by glycyrrhizin, not by millilitres. Ask for the assayed content of the batch and calculate the daily glycyrrhizin.
2. Take a blood pressure before prescribing, and again during, as MediHerb’s own guide directs [80].
3. High potassium, low sodium diet for anyone on high-glycyrrhizin preparations for more than a short period; monitor weight and blood pressure [10,11,80].
4. Use DGL where the flavonoids are what is wanted — ulcer, dyspepsia, mucosal healing — and keep whole root for where the glycyrrhizin is the point [10,11].
5. Take a licorice-food history. Confectionery, liquorice tea and Chinese and Kampo formulas all contribute. A Danish survey found pure liquorice confectionery at a mean 17,922 mg/kg glycyrrhizic acid, and one Dutch herbal-tea brand at 450 mg/L of the prepared beverage [8].
6. Stop 7 days before general anaesthesia [80].
7. Anyone considering licorice should consult a degree-qualified herbalist with competencies and adequate training in pharmacognosy.