Ashwagandha: how it works, safety and what your blood shows

Ashwagandha is a root extract sold for stress, and it is one of the few supplements whose effect shows up in your blood. In March 2024 the Dutch institute RIVM advised against using it, after reports of liver damage. The trials do show measurable shifts in cortisol, testosterone and thyroid values.

What strikes me as a training partner: almost every Dutch page about ashwagandha also sells it. RIVM holds position 1 in Google. The nine results below it are mostly web shops.

So here is the other angle. Not whether you should take it, but what the trials actually measured, and which values on your own lab report could move.

What exactly is ashwagandha?

Ashwagandha is the root of Withania somnifera, a plant used in Ayurveda. Its active compounds are called withanolides. Supplements are usually standardised to a percentage of withanolides, and that percentage varies a lot between brands. Most trials used 300 mg of root extract, twice a day.

The name roughly means smell of a horse. That refers to the sharp smell of the fresh root, not to what it does to you.

Sellers call it an adaptogen. That is not a medical term. It is a label for herbs said to help the body cope with stress.

That vagueness is the problem. Something that supposedly does everything a little is hard to test. What can be tested are the values in your blood.

What does RIVM say about ashwagandha?

On 5 March 2024, RIVM advised against using food supplements containing ashwagandha, and especially against use during pregnancy. The trigger was reports from doctors about poisonings. According to RIVM, those cases involved harmful effects in the liver, among other things.

RIVM adds that little scientific research has been done into its harmful effects. The advice rests on reports, not on a large study.

That is worth reading carefully. A warning based on reports is not proof that it goes wrong for everyone.

It is a good reason to know what you are measuring.

What did the trials do to cortisol?

In the best known study, 64 adults with chronic stress took 300 mg of extract twice a day for sixty days. Serum cortisol ended up lower than in the placebo group, at p=0.0006 (Chandrasekhar, 2012). Stress questionnaires scored lower too.

There is a detail here that almost everyone gets wrong. Countless pages quote a 27.9 percent drop. That number does not appear in the abstract of that study.

What it does say is substantially reduced, with that p-value. A significant difference, then, without a percentage.

When you read a precise number with no source attached, that is a reason to click through.

And there is a counterweight. In men aged 40 to 70, cortisol did not change significantly against placebo (Lopresti, 2019). Same compound, different group, different result. More on how cortisol moves in cortisol and stress.

What does ashwagandha do to testosterone and strength?

This is the study the fitness world keeps quoting, and it is genuinely striking. 57 young men did eight weeks of resistance training. Half took 300 mg of root extract twice a day (Wankhede, 2015).

The differences were large. On the bench press, the ashwagandha group gained 46.0 kg on their 1RM against 26.4 kg on placebo. On the leg extension it was 14.5 kg against 9.8 kg.

Testosterone rose by 96.2 ng/dL, against 18.0 ng/dL on placebo.

That last figure is where the translation to your own result breaks down. Dutch labs do not report testosterone in ng/dL but in nmol/L. 96.2 ng/dL works out at roughly 3.3 nmol/L.

Put that next to a real result and it becomes manageable. Say you sit at 15 nmol/L. A shift of 3.3 nmol/L puts you around 18 nmol/L, still inside the same reference range.

That is a real change, not a different person. More numbers in ashwagandha and testosterone.

What does ashwagandha do to your thyroid?

This effect gets discussed least and can distort your result most. Fifty people with subclinical hypothyroidism took 600 mg a day for eight weeks (Sharma, 2018). TSH, T3 and T4 all changed significantly against placebo.

For those participants that was the point. For you it means something else.

If you take ashwagandha and have your TSH drawn, you measure your thyroid plus the supplement. So mention it if your result comes back off.

How TSH and free T4 relate is covered in thyroid and your metabolism.

Ashwagandha and your liver: what has been reported?

This is the core of the RIVM warning. The medical literature holds several cases of liver injury after ashwagandha use. The pattern LiverTox describes is cholestatic or mixed, usually two to twelve weeks after starting, with jaundice and itching.

The largest series so far comes from India (Philips, 2023). It covered people using ashwagandha as a single-ingredient product, not as part of a herbal blend. That makes the injury harder to pin on something else.

It is rare. Millions take this without trouble, and the clinical trials reported no rises in liver enzymes.

But rare is not zero, and early liver injury has no symptoms. ALT and AST rise before you notice anything.

What training itself does to those values is covered in liver values and supplements in athletes. That distinction matters, because a heavy leg session lifts your AST too. The reported complaints and the full picture are in ashwagandha side effects.

Which blood values can change?

Below is what the trials measured, translated into the name and unit you actually see on a Dutch lab report. That step is missing from almost every other page, and without it you cannot lay a trial figure next to your own result.

Look at rows three and four. Wankhede measured testosterone in serum, Lopresti in saliva. Those are two different measurements.

So you cannot add them together, even though both describe the same hormone.

What should you not combine with ashwagandha?

This is one of the most asked questions in Google, and the honest answer is that the data are thin. The combinations flagged most often are thyroid medication, sedatives and sleep aids, and drugs that suppress the immune system.

The logic behind that is simple. The supplement appears to touch the same dials as those medicines.

If you take medication, raise it with your GP or pharmacist before combining anything. They can weigh that up. A blog cannot.

Who should be extra careful?

RIVM names pregnancy explicitly, and there is a reason for that. In some countries the root was historically used to induce abortion. That alone makes use during pregnancy unwise.

Beyond that, three groups have reason to pay closer attention.

People on thyroid medication, because the Sharma trial shows it touches the same values. People already known to have a liver condition, given the pattern LiverTox describes. And people taking drugs that suppress the immune system.

In all three cases your GP or pharmacist is the right person to ask, not a product page.

Why the trials disagree so much

Four trials, four different extracts, four different doses. Chandrasekhar used 600 mg a day split into two, Sharma 600 mg in one dose, and Lopresti an extract delivering 21 mg of withanolide glycosides a day.

That last number sits on a completely different scale from the first two. So you are not comparing like with like.

On top of that the participants differed sharply. Stressed adults, young lifters, and men aged 40 to 70 with fatigue. Something that does anything in one group need not do it in another.

What ashwagandha does not do

Ashwagandha is not a testosterone booster in the sense the marketing suggests. The rise in Wankhede came on top of eight weeks of resistance training, and the placebo group trained just as hard.

Training did most of the work there. The supplement sat on top of it.

It is not a sleeping pill either, even though the packaging often carries a little moon. The species name somnifera means sleep-inducing, but that is eighteenth-century botanical naming, not a research finding.

And it does not replace treatment. If you sleep badly week after week, or feel low, a conversation with your GP beats a jar of capsules.

What does ashwagandha do to your emotions?

In the trials, stress and anxiety scores on questionnaires improved, mostly in participants who already reported stress (Salve, 2019). In men with mild fatigue there was no significant difference against placebo on mood and vitality.

The picture is mixed, then, and it depends on who you measure.

Some users report a flat or blunted feeling instead. That is not in the trials, but it recurs in user reports. What people report themselves is covered in ashwagandha reviews.

How long before you notice anything?

The trials ran eight to sixteen weeks, and measurements were taken at the end of that period. No solid study shows what to expect after three days. Anyone who feels something in week one is probably feeling expectation.

Trial-by-trial timelines are in how long ashwagandha takes to work.

Which extract did you actually buy?

The doses from the trials mean little if your product is standardised differently. KSM-66 and Sensoril are two different patented extracts, with a different withanolide content and a different research base.

A jar holding 500 mg of powder is not the same as 500 mg of extract. That difference can be a factor of ten in active compound.

Which version was studied for what is covered in KSM-66, Sensoril or powder. Other supplements judged the same way are in 9 supplements for athletes.

What you can do with this

My own line is simple. RIVM advises against this, and that carries enough weight not to wave away. At the same time, pretending nothing has been measured is nonsense, because plenty has.

If you use it anyway, it helps to know where you started. A cortisol test gives insight into your cortisol, and total testosterone shows where you stand.

If your result comes back off, discuss it with your GP and mention which supplements you take. That one sentence saves a lot of digging.

Every blood test result includes a professional assessment from a BIG-registered doctor. For treatment decisions, discuss your results with your GP.

References

• RIVM. Advice on products containing Huperzia serrata, Tabernanthe iboga or ashwagandha. 5 March 2024.
• Chandrasekhar K, Kapoor J, Anishetty S. Indian J Psychol Med. 2012 (PMID 23439798).
• Wankhede S, Langade D, Joshi K, Sinha SR, Bhattacharyya S. J Int Soc Sports Nutr. 2015 (PMID 26609282).
• Sharma AK, Basu I, Singh S. J Altern Complement Med. 2018 (PMID 28829155).
• Lopresti AL, Drummond PD, Smith SJ. Am J Mens Health. 2019 (PMID 30854916).
• Salve J, Pate S, Debnath K, Langade D. Cureus. 2019 (PMID 32021735).
• Philips CA, Valsan A, Theruvath AH, et al. Hepatol Commun. 2023 (PMID 37756041).
• LiverTox. Ashwagandha. NCBI Bookshelf, NBK548536.