Is Ashwagandha an Abortifacient?

Ashwagandha (Withania somnifera (L.) Dunal, Solanaceae, AS) is a traditional adaptogenic herb used in Ayurvedic medicine. The use of adaptogenic ingredients in dietary supplements and herbal medicines has enjoyed continuously increasing popularity. In Europe (European Union and United Kingdom) AS is regulated as a food supplement (EU Directorate-General for Health and Food Safety 2024). Despite being considered safe based on its long-standing traditional use, regulators started to question aspects of AS safety of as early as 2013 (Klenow et al. 2013), leading to a recent recommendation for an Article 8 procedure under Regulation (EC) No 1925/2006 (Heads of Food Safety Agencies 2024). If executed, this may result in a ban on its use as food supplement, which may lead to removal of the ingredient from the EU marketplace entirely. One of the primary concerns raised by regulators is its purported abortifacient effect (Brendler et al. 2025). Overall, abortifacient claims in the “canon” are largely present as unreferenced assertions (~60%) and indiscriminate citation of secondary, tertiary, quaternary etc. literature (~35%), leaving ~5% of publications fit for further consideration. For example, a well-cited and otherwise thorough review of botanical products of concern during pregnancy (Bruno et al. 2018) references an equally well-cited paper on general effects of AS (Mir et al. 2012). In the section “folklare” [sic], Mir et al. claim abortifacient effects for AS roots, citing three references (Dutt 1877; Kumar et al. 1980; Singh and Kumar 1998), the first two of which do not appear in the list of references. Dutt (1877) does not mention abortifacient effects, Kumar et al. (1980) discuss aphrodisiacs, and Singh and Kumar (1998) cite Garg and Parasar (1965) (discussed below) and Duke's Handbook of Medicinal Herbs, which yields another five references (Duke 1985). These, in turn, cite additional references, some of which circle back to previous citations. The German risk assessment of 2013 (Klenow et al. 2013) first points to the WHO monograph for AS (WHO 2009), which in the contraindications section states: “Due to the lack of safety data and the fact that the crude drug has been used in traditional medicine to induce abortion, its use during pregnancy or breastfeeding is contraindicated,” citing the American Herbal Pharmacopoeia's AS monograph (Gattu 2000). There, the abortifacient effect is backed up by citing as a popular book by Svoboda (1992), which claims uterotonic properties and abortifacient in large doses without reference (Svoboda 1992), and the AS monograph in The Wealth of India (Chadha 1976), which attributes abortifacient properties to four sources: Bector et al. (1968), Garg and Parasar (1965), Kirtikar and Basu (1933) and Chopra's Indigenous Drugs of India (1958). Chopra's quotes Kirtikar & Basu verbatim without actually citing them (Chopra et al. 1958). Bector et al. discuss arthropathies (Bector et al. 1968), Garg & Parashar are discussed below, and Kirtikar & Basu claim use as an abortifacient in Sindh (Pakistan). They cite Pammel (1911), A Manual of Poisonous Plants, who bases his claim on Greshoff (1907) The Distribution of Prussic Acid in the Vegetable Kingdom, which evidently is a mis-citation1 (Kirtikar and Basu 1933; Pammel 1911). Thus, the only promising trace to follow is the ethnobotany of the province of Sindh (Pakistan). An authoritative and plausible source for this is The Plants and Drugs of Sind (Murray 1881). Another, Stocks' Notes upon the Materia Medica of Sind (1857), makes no mention of Physalis or Withania species (Stocks and Haines 1857). And indeed, it is Murray's entry for Physalis somnifera L. (a synonym of AS) that states: “It is often used in Sind to cause abortion.” Here, Murray cites William Roxburgh, John Lindsey Stewart and Whitelaw Ainslie as sources. Ainslie's Materia Indica (Ainslee 1826) reports deobstruent and diuretic qualities as well as alexipharmic with reference to an earlier edition of Flora Indica. Roxburgh, for Physalis flexuosa, claims it to be alexipharmic (Roxburgh 1832). The primary source is, in fact, Punjab Plants (Stewart 1869) which states, now under Withania somnifera: “At least one case has occurred in the Punjab in which the root was used with a view to effect criminal abortion, and I learned that in Sind this practice is not uncommon.” No such claims are made in the preceding literature, but this assertion appears randomly throughout the literature to this day, often verbatim and often without reference. Of particular note is that Stewart is a singular source, and his claim is not backed by any other contemporary, including Vicary's Some notes on the botany of Sinde, one of the most likely field sources at the time (Vicary 1848). The irony must not be lost, however, that contemporaries also classified Physalis somnifera as a narcotic poison (Beck and Dunlop 1825) and reported on multiple cases of purported lethal poisonings with Physalis flexuosa L., another synonym of AS (Burton Brown 1888). These associations—surprisingly but thankfully—were not carried over into modern literature. Also noteworthy is the lack of contemporary sources citing Stewart as the primary source—the most recent that I could identify was published almost 80 years ago (Badhwar et al. 1946). Apart from the claim being not much more than hearsay, it is likely that it is based on a taxonomic and possibly a plant part confusion, as well as a mistranslation into English of the native concept of “uterotonic”. The taxonomic confusion may arise from overlapping common names for Physalis indica Lam. (nom. illeg.), Withania coagulans (Stocks) Dunal, Alkekengi officinarum Moench and other Solanum spp. (Birdwood 1862; Stocks 1850), some of which have documented abortifacient effects. For instance, antifertility effects are reported for Physalis angulata L. (Mohana et al. 1979; Sudhakaran et al. 1999). This assertion, however, is not corroborated in the contemporary literature (Joseph and Ravi 2022; Wu et al. 2021). Contemporary investigators regularly cite menstrual disorders as within the traditional use spectrum of AS (Sharma et al. 2023). Given the prominence of the abortifacient assertion in the literature, we may very well be looking at investigator bias (Jain et al. 2004; Sharma et al. 2014; Singh and Mall 2008) or a mislabeling of established effects on reproduction and hormonal regulation (Dodke et al. 2020), if an uterotonic or even antifertility effect is reported as abortifacient. More aptly, authors of a recent review of Indian tribal ethnogynecology recommend AS for female infertility (Surendran et al. 2023). Another recent review of contraceptives in Ayurveda makes reference to earlier research finding spermicidal and infertility inducing effects for the fruit of AS in male rats (Bhatt and Deshpande 2021; Mali 2013), which is clearly distinct from abortifacient. The same authors associate AS with teratogenicity at high doses, a claim that is made but not demonstrated (or referenced) by the publication they cite (Priyadarsini et al. 2018). AS root powder (25 mg for 10 days prior to mating and continued until day of littering) was fed to four male and female mice; another eight mice were kept as control. Delayed mating, decreased litter and infertility in one case were observed (Garg and Parasar 1965). This study is widely cited, but it is neither sufficiently powered nor indicative, for example, from its setting one cannot derive whether it reports an effect on the female or male reproductive system. Mali (2013) is also routinely cited as (tertiary) evidence even though he studied male infertility, not female. A highly referenced and comparatively contemporary investigation of commonly used Indian abortifacient plants interestingly does not mention or investigate the by then already very popular AS (Nath et al. 1992). As early as 1970, multiple investigations cited in Mahajan and Gaitonde (1991) showed that AS root preparations do not possess antifertility activity, which the authors confirm in their own investigation in female albino rats (Mahajan and Gaitonde 1991). More recently, orally administered methanolic extract (500 mg/kg or 3000 mg/kg body weight per day) or water extract (500 mg/kg body weight per day) of AS root consecutively from Days 1–7 of pregnancy did not show abortifacient or antifertility effects in female rats (Ilayperuma et al. 2002). This outcome was confirmed in an investigation of pregnant rats (Days 5–19 of gestation, 25 in each group) fed with either 500 mg/kg, 1000 and 2000 mg/kg body weight per day of a hydroalcoholic AS root extract, which showed no evidence for maternal or fetal toxicity (Prabu and Panchapakesan 2015). This investigation was performed in full compliance with OECD guidelines 414 for the assessment of prenatal developmental toxicity. As such, its dismissal by the DTU report (DTU Fødevareinstituttet 2020) that has served as the basis for all regulatory interventions in the EU hitherto is unwarranted. A further study found ameliorating effects of AS root extract (100 mg/kg body weight) on reproductive pathology and teratogenic effects of methimazole induced hypothyroidism in female Sprague Dawley rats (Ramya et al. 2024). The adage that one only needs to repeat a falsehood often enough to create an “illusion of truth” is well-known. What started off as an indiscriminate citation of sources without verification created a canonized “modern myth” that will be almost impossible to eradicate, a phenomenon that sadly is of increasing frequency in the scientific literature (Nissen et al. 2016). Nonetheless, the record must be set straight: There is effectively no authoritative evidence that AS has an abortifacient effect. On the contrary, the few modern studies available all appear to dispel this myth, namely by identifying beneficial effects of AS on teratogenicity and reproductive pathologies (Tallon et al. 2025; Tharangani et al. 2025). AS doubtlessly has endocrinological effects on female hormone levels (Wiciński et al. 2023), which are further documented in clinical trials (Gopal et al. 2021; Smith et al. 2023), and have been assessed for associated risks (Vollmer and Brendler 2025; Williamson and Brendler 2025). Those may warrant caution under specific circumstances but cannot be characterized as abortifacient. T. Brendler: conceptualization, data curation, investigation, methodology, project administration, writing – original draft, writing – review and editing. The author declares no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.