Ashwagandha at the supplement aisle: a practical buyer's overview

For the cognitive/stress framing of ashwagandha and the trial summary, see ashwagandha-research. This page focuses on the practical supplement-aisle considerations: standardisation, quality, the hepatotoxicity signal, and how to buy something that resembles what the trials used.

The standardisation problem#

Ashwagandha supplements vary dramatically:

• KSM-66: patented root extract standardised to 5% withanolides. The most-studied form in modern RCTs.
• Sensoril: patented root + leaf extract standardised to 10% withanolides. Used in some trials.
• Generic root powders: withanolide content can be near zero in cheap products; quality varies dramatically
• Whole-plant extracts: different active profile from root-only extracts
• Bare-bones ashwagandha tablets — often very low active content

For products with no standardisation listed, you have essentially no information about what's inside.

What the trials use#

The bulk of modern positive-result ashwagandha RCTs use:

• KSM-66 at 300–600 mg/day, root only
• 8–12 week treatment durations
• Standardised batches manufactured by Ixoreal Biomed

Trial doses vary by extract, so "the trial dose" isn't a single number: the placebo-controlled trial cited on this page (Lopresti et al. 2019) used 240 mg/day of a different standardised extract (Shoden), not KSM-66 or Sensoril at 300–600 mg/day. If you're trying ashwagandha based on trial evidence, matching the specific product and dose used in a trial you trust is more useful than assuming any standardised extract is interchangeable. Cheaper unstandardised products may or may not work; you can't predict.

The price gap#

Standardised KSM-66 or Sensoril products are typically £20–£35/month at standard doses. Cheaper unstandardised ashwagandha can be under £10/month.

The price gap is largely the standardisation and quality control. For an intervention where dose matters and where you're hoping to replicate trial outcomes, paying for standardisation is reasonable.

The hepatotoxicity signal#

A small but real safety signal has emerged in case reports:

• Several published case reports of acute hepatitis associated with ashwagandha use
• Cases from US, Iceland, Japan, and elsewhere
• One documented 2023 case (acute hepatitis with jaundice) improved after intensive pharmacological treatment and plasmapheresis, without needing a transplant
• The base rate is uncertain — likely uncommon but not negligible

Risk factors aren't well-characterised. The case reports don't show a clear pattern by product or population. Some occurred with standardised products.

What this means for users:

• Awareness of jaundice, dark urine, abdominal pain, unexplained fatigue
• Don't combine with other potentially hepatotoxic substances (alcohol, kava, high-dose paracetamol)
• People with existing liver disease should not use ashwagandha
• Periodic liver function tests are reasonable in long-term users (every 6 months)

Other side effects in practice#

• GI symptoms at higher doses (loose stools, nausea)
• Drowsiness: some users find this useful (evening dose); some find it interferes with daytime functioning
• Thyroid effects — modest TSH/T4 changes; problematic in hyperthyroidism, potentially helpful in subclinical hypothyroidism (don't self-manage)
• Sedating medication interactions — additive effect
• Pregnancy — traditionally avoided; theoretical abortifacient at high doses

Quality markers to look for#

If buying ashwagandha:

• Standardisation specified (e.g. "300 mg KSM-66, 5% withanolides")
• Branded extract (KSM-66 by Ixoreal Biomed, Sensoril by Natreon, etc.)
• Third-party testing mentioned (NSF, Informed Sport for athletes, or independent batch testing)
• Clear root vs leaf vs whole-plant labelling

Avoid:

• "Ashwagandha 1000 mg" with no standardisation info
• "Proprietary blends" hiding the actual dose
• Products with multiple poorly-evidenced ingredients combined ("stress complex" type formulations)

Cycling vs continuous use#

Some users cycle ashwagandha (8 weeks on, 2–4 weeks off) on the theory that adaptogens lose efficacy with continuous use. The trial evidence is mostly 8–12 weeks continuous; longer-term data is sparse.

There's no specific evidence ashwagandha requires cycling. Some users cycle because they don't want chronic exposure to an unfamiliar compound; that's reasonable conservative practice.

What the marketing overstates#

• "Reduces stress / fixes burnout." Real anxiolytic effects but burnout has structural causes that supplements don't address
• "Boosts testosterone significantly." Modest effects in specific populations; not equivalent to clinical TRT
• "Treats depression." Adjunct at most; NICE first-lines exist for depression
• "Improves sleep." Modest effects via stress/cortisol pathway

What we'd actually say#

If you're trying ashwagandha for stress-related symptoms, use a standardised branded extract (KSM-66 most-studied) at 300–600 mg/day for 8–12 weeks. If you don't notice anything by week 12, stop.

For someone with diagnosed anxiety, depression, or insomnia, ashwagandha is not first-line — discuss NICE-recommended treatments with a GP first.

The hepatotoxicity signal is uncommon but real. Awareness of liver-related symptoms matters.

For most users, the effects are modest and worthwhile but not transformative. Manage expectations to match the trial effect sizes.

When to talk to a clinician#

• Existing liver disease or abnormal liver function tests
• Symptoms suggestive of liver injury after starting (jaundice, dark urine, persistent fatigue)
• Diagnosed anxiety, depression, or insomnia
• Thyroid disorders, particularly hyperthyroidism
• Autoimmune disease
• On sedating medications, thyroid medications, or immunosuppressants
• Pregnancy or trying to conceive
• Yellow Card reporting for any unexpected reaction