Adhatoda (Justicia adhatoda)

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

Adverse effects in trials. None attributed to Adhatoda-containing preparations: no adverse effects in the COVID-19 trial [47]; no adverse events in the open probiotic study [45]; no serious events and only mild treatment-emergent events, none judged related, in the COPD trial [44]; fewer side effects with Linkus than with the drug comparators [46]. The trials are small and short (5 days to 12 weeks, a few hundred participants in all), so uncommon effects could have been missed. In the COVID-19 trial, eosinophils rose and PT/INR changed within the normal range in the Adhatoda arm; the paper describes the PT/INR change in opposite directions in two places [47].

Case reports and pharmacovigilance. No case report of harm from Adhatoda leaf was found in the searches above. Braun and Cohen record "insufficient reliable information" on adverse reactions [14]. The TGA Database of Adverse Event Notifications (DAEN, medicines) was searched in November 2026 for adhatoda, justicia, vasaka, malabar, PulmaCo, Lung Care and Kan Jang, covering reports from 01/01/1971 to October 2026 [60]. One product containing Justicia adhatoda appears: a topical head-lice product that also contains echinacea, tea tree oil and Stemona, with two reports (2003, "hair texture abnormal"; 2009, "application site reaction"). No report names an oral Adhatoda product. A DAEN entry does not show that a medicine caused the event, and these two cannot be assigned to Adhatoda among four ingredients. A search of the TGA website for adhatoda and justicia in November 2026 returned four ARTG product entries and no safety alert, advisory or recall [61].

Toxicology. No full-text repeat-dose toxicity study of the leaf or of vasicine was read for this version. A chronic toxicity study of vasicine in rats and monkeys (1987), a 2000 critical review of the toxicology and a 1997 rat study whose title reports no abortive effect of an oral leaf extract exist and have been requested in full text for the next revision; nothing is concluded from them here.

Pregnancy: contraindicated.

• Practice texts and labels (grade: expert practice opinion): MediHerb: contraindicated in pregnancy [13]. Bone and Mills: contraindicated, "except at birth" [19]. Braun and Cohen: contraindicated, as the herb may have abortifacient activity [14]. The ARTG entries for PulmaCo carry the warning "Do not use if pregnant or likely to become pregnant" [20,59].
• Animal evidence (grade: animal, known only through a secondary source), as reported by Braun and Cohen: leaf extract was 100 per cent abortive in rats at a dose equivalent to 175 mg/kg of starting dry material, and another extract showed anti-implantation activity in 60-70 per cent of animals; the authors note that in vivo results have been inconsistent [14].
• Pharmacology and tradition (grade: preclinical description and traditional use): vasicine is described as uterotonic and abortifacient [3]; a 1980 monograph from the Jammu laboratory is titled A New Oxytocic and Abortifacient [19]; the folk and Ayurvedic record includes abortifacient use and use to assist delivery [14,31].
• Regulatory (grade: regulatory assessment of the same animal and traditional data): the EMA’s committee found information on possible abortive, anti-implantation and uterotonic effects, asked for further reproductive-toxicity data, and gave reproductive-safety concerns as one reason no EU monograph could be established [2]. A 2018 call for new data has not, on the EMA record, led to a revised outcome [53].
• No human pregnancy data were found. Each strand is weak alone; together they point the same way, and the contraindication stands.

Lactation: avoid. No human or animal data on vasicine in breast milk or on exposure of the breastfed infant were found. MediHerb advises professional supervision [13]; the ARTG entries for PulmaCo say "Do not use while breastfeeding" [20,59]. AIP position (Medicine): avoid while breastfeeding, as a precaution in the absence of data and in line with the Australian label (grade: no data; precautionary).

Children and older people. The WHO South-East Asia manual cited by Bone and Mills recommends long-term use in children [19], and one open trial treated children with a polyherbal syrup containing Adhatoda [46]. The PulmaCo ARTG entry says use in children under 12 years is not recommended [20]; the Lung Care entry says not for children under 5 [21]. No paediatric safety study of the leaf was found (grade: no data beyond one open trial of a polyherbal syrup). No data for older people.

Surgery. MediHerb says to stop 7 days before general anaesthesia [13]. The same sentence appears on other herbs on the same page (Agrimony, Albizia), so it reads as a range-wide precaution; no evidence specific to Adhatoda was found.

Australian labelling. The Permissible Ingredients Determination sets no specific requirement or warning for Justicia adhatoda [7]. Two listed products carry no pregnancy warning on their ARTG records: a tablet with 500 mg of dried-leaf equivalent [21] and a capsule with 1.8 g, whose record states "No Warnings included on Record" [58]. The practitioner tablet carries one [20,59]. AIP comment: given the EMA’s concern, the absence of a mandatory pregnancy warning for this ingredient is a gap in the labelling rules.

Interactions (each graded; none is supported by a human interaction study).

Overdose. No reports found.

Referral wording. Anyone considering Adhatoda while pregnant or trying to conceive, while breastfeeding, for a child, or alongside prescribed medicines should consult a degree-qualified herbalist with competencies and adequate training in pharmacognosy.