Guide Medicinal Plants · Adaptogens and the Stress Axis

Ashwagandha: What the Winter Cherry Can Really Do and Who It Makes Sense For

Not a miracle cure and not a placebo. A plant with a clear address in the body, the stress axis. And with a clear condition: measure first, then decide.

Cortisol & HPA axis Evidence, honestly Contraindications Diagnostics first Integrative medicine
Shukri Jarmoukli · Physician, Integrative Medicine · ViveCura Berlin
My starting point

Ashwagandha is probably the most frequently bought plant against a feeling that nobody has ever measured. Exhaustion is not a diagnosis. It is a signal. And signals deserve to be understood before they are covered up.

I bet you know this scene. You are standing in front of a shelf or in front of a phone screen. It says: for stress, for calm, for sleep, for balance. You are tired. You are not ill enough for a diagnosis and not healthy enough for your life. So you reach for it.

And then often nothing happens. Or something very small. Or you feel oddly flat in the second week and do not know why.

In this text I do not want to explain to you that ashwagandha is bad. That would be dishonest, because the data are better than for many other plants on the shelf. I want to explain to you where this plant docks in the body, and why it can change something in some people and not in others. That is not a question of brand. It is a question of physiology.

What you can expect here

  • What Withania somnifera is and what withanolides are
  • How the HPA axis works, in images instead of jargon
  • Why a flat cortisol rhythm is something different from too much cortisol
  • What meta-analyses show for stress, anxiety and sleep
  • Where the studies are weak, named openly
  • When ashwagandha does not make sense, including liver and thyroid
  • Which values belong on the table before a supplement
  • Study dosages from the literature, clearly marked as such
Evidence markers: Clinical trial in humans Observation in humans Animal study Cell culture

What ashwagandha actually is

Ashwagandha is the root of Withania somnifera. The plant belongs to the nightshades, just like tomato, potato and bell pepper. Its English common name is winter cherry, in German Schlafbeere, the sleep berry. In the Ayurvedic tradition it belongs to the Rasayana plants, so to the remedies meant to support strength and regeneration.

The most interesting group of compounds in it are the withanolides. These are steroidal lactones, molecules with a backbone that resembles our own steroid hormones. Picture withanolides as molecules that know their way around the hormone system, because they speak a similar language. Withanosides, alkaloids, flavonoids and phenolic compounds come on top of that.

From the literature Review

A comprehensive review from 2021 compiled the ethnopharmacology, phytochemistry and toxicology of Withania somnifera. The authors describe a very broad range of investigated effects, from inflammation-modulating to neuroprotective, and draw a surprisingly sober conclusion from it.

Their bottom line: both deeper clinical studies in humans and a detailed toxicological analysis are needed before the plant is used widely as a health-promoting agent. For you that means: the abundance of promised effects online does not come from an abundance of good human trials.

Paul S et al. Biomed Pharmacother. 2021;143:112175. DOI: 10.1016/j.biopha.2021.112175

KSM-66, Sensoril, Shoden: why the name on the tin matters

Studies almost always use standardized extracts with brand names. The best known are called KSM-66, Sensoril and Shoden. They differ in the plant part used, in the solvent, in the extraction method and in the declared withanolide content.

That is not a marketing detail. It is the reason why studies can only be pooled to a limited degree. A systematic review from 2023 on cortisol names exactly this variety of formulations, dosages and study durations as a central cause of contradictory results.

In practical terms A product without a declared withanolide content, without information on the plant part and without documented heavy metal testing is not a basis for a medical decision. It is a powder with a story.
Why purity is a genuine issue Product analysis

An investigation published in JAMA in 2008 bought 193 Ayurvedic preparations over the internet and analyzed them with X-ray fluorescence. In roughly one in five products, detectable amounts of lead, mercury or arsenic were found. The proportion was just as high for products manufactured in the USA as for Indian ones.

Important for context: what was examined were Ayurvedic preparations in general, not ashwagandha specifically, and the data are older. Still, the lesson holds. With plant products bought online, the question of batch testing and heavy metal analysis is part of the deal.

Saper RB et al. JAMA. 2008;300(8):915-923. DOI: 10.1001/jama.300.8.915

And now you know why the question is not whether ashwagandha can do something, but which ashwagandha was studied at which dose in which person.

A story I hear often

From clinical practice, anonymized

Seven months of ashwagandha, and the tiredness stayed

A patient in her mid thirties, an office job with a lot of responsibility, came in with an exhaustion that she herself described as a permanent state. Not dramatic. Just uninterrupted.

“I have been taking this for seven months now. Everyone says it is the best thing for stress. With me, simply nothing happens.”

She had done everything right that you can learn from the internet. A standardized extract, regular intake, patiently kept up. What nobody had done with her: a diurnal cortisol profile.

What we measured was not a surplus. It was a flat curve. In the morning the rise that lifts a body out of sleep was missing. In the evening the drop that lets it back in was missing. Across the day the curve was almost a straight line. The rest of the picture fit with that: a ferritin in the lower range and a sleep that lasted eight hours but never really became deep.

We worked on that straight line first. Light in the morning, meals at fixed times, strength training at a dose that does not add to the exhaustion, and the measurable gaps filled in a targeted way. Ashwagandha was not the question at that point.

Her experience of herself changed noticeably over the following months. I cannot claim causality, but I do document the temporal connection.

The lesson in one sentence: a remedy against too much stress hormone can achieve nothing if the problem is a lost rhythm.

The HPA axis: why cortisol is not an enemy

Cortisol has a bad reputation, and it is undeserved. Cortisol is not the hormone that wrecks you. It is the hormone that lifts you out of bed in the morning, keeps your blood sugar stable, puts a brake on inflammation and keeps you able to act in danger.

It arises at the end of a chain that specialists call the HPA axis. That stands for hypothalamus, pituitary gland and adrenal cortex, so for three stations from your brain down to your kidneys.

1

Hypothalamus

The weather station in the brain. It registers strain, light, blood sugar, pain, inflammation. Then it releases CRH, a start signal.

2

Pituitary gland

The switchboard. It hears the start signal and sends ACTH off through the bloodstream, like a circular email to the adrenal glands.

3

Adrenal cortex

The production site. This is where cortisol is made and, from a related branch, DHEA. Both are built from cholesterol.

4

Cortisol in the blood

It reaches almost every cell, binds to receptors in the cell nucleus and changes which genes are read there. That is why the effect is so broad.

5

Feedback upward

Cortisol reports back to the hypothalamus and the pituitary and throttles its own production. A thermostat. And this very thermostat can become imprecise under lasting strain.

What matters is not the level at a single point in time. What matters is the daily course. A healthy rhythm rises steeply in the first 30 to 45 minutes after waking, falls across the day and is almost down at night.

Two patterns that look completely different in the lab

Waking +30 min
Late morning
Afternoon
Before bed

Above: a rhythmic course with a clear morning rise and quiet at night.

Waking +30 min
Late morning
Afternoon
Before bed

Below: a flattened course. The morning does not get going, the evening does not come to rest. The total daily amount can be completely unremarkable in this case. Schematic illustration, not measured values of any particular person.

Why the flatness is medically relevant Meta-analysis

A group led by Emma Adam at Northwestern University systematically summarized the literature on the diurnal cortisol slope in 2017. They evaluated 179 associations from 80 studies.

What they observed: a flatter diurnal slope was overall associated with poorer physical and mental health. This pattern appeared in 10 of 12 domains examined, most clearly for immune and inflammatory markers.

What that means for you: a flat rhythm is a finding to take seriously, even when the single value looks normal. The effect sizes are moderate, though, and the authors explicitly leave open whether the flatness is cause, consequence or both.

Adam EK et al. Psychoneuroendocrinology. 2017;83:25-41. DOI: 10.1016/j.psyneuen.2017.05.018
This changes the perspective

“Too much cortisol” and “no rhythm left” are two different states. Both feel like exhaustion. But they need different answers.

A remedy that in studies mainly lowered the measured cortisol level is plausible for the first pattern. For the second pattern it misses the point, and in the unfavorable case it pushes a morning rise even further down that is already too weak.

Staying fair: what the term adrenal fatigue is not

Online, the term adrenal fatigue circulates for this picture. A systematic review from 2016 examined 58 studies and came to a clear result: as a distinct disease entity it is not supported. Endocrinology does not recognize it, and the test procedures used were often methodologically weak.

I consider this criticism justified and I adopt it. What is also true alongside it: the diurnal cortisol course is a measurable parameter that is well established in research, and a flattened slope correlates with poorer health data. One of the two is a label without a foundation. The other is a measurement with literature behind it. I work with the measurement and leave the label aside.

Cadegiani FA, Kater CE. BMC Endocr Disord. 2016;16:48. DOI: 10.1186/s12902-016-0128-4

And now you know why the sentence “I have too much stress, so I take something against cortisol” falls biologically short.

What the studies show, and what they do not

Now to the numbers. Among medicinal plants, ashwagandha is relatively well studied. There are several meta-analyses. They point in the same direction, and they all show the same weaknesses.

Stress and anxiety Meta-analysis, k=12, n=1,002

A group at the University of Tehran pooled twelve randomized controlled trials with a total of 1,002 participants in 2022. They found a clear reduction in anxiety and stress scores compared with placebo, with the most favorable dose range for stress at 300 to 600 milligrams per day.

And then comes the sentence you have to read along with it: the authors rated the certainty of the evidence for both endpoints as low. The scatter between the studies was extremely high. For you that means: the direction is recognizable, the size of the effect is not.

Akhgarjand C et al. Phytother Res. 2022;36(11):4115-4124. DOI: 10.1002/ptr.7598
Cortisol and the experience of stress diverge Meta-analysis, k=7, n=488

This 2025 paper from Saudi Arabia deliberately looked at two things separately: the laboratory value cortisol and the subjective experience of stress on the Perceived Stress Scale.

What was observed was a statistically significant reduction in cortisol of 1.16 micrograms per deciliter. On the stress scale there was no significant difference from placebo. The author calls this a discrepancy and asks for longer studies.

For you this is the most important finding in this article. A biomarker can move without your life moving with it. Anyone who looks only at cortisol confuses a number with a result.

Albalawi AA. Nutr Health. 2025;31(4):1395-1408. DOI: 10.1177/02601060251363647

A lowered cortisol value is not proof of a better life. It is an indication that a substance changes something in a system. What it changes in you is decided by your baseline findings.

Sleep Meta-analysis, k=5, n=400

A Malaysian group at Universiti Sains Malaysia analyzed five randomized trials with 400 participants in 2021. They found a small but significant effect on overall sleep.

The subgroups are the interesting part: the effect was clearer in people with diagnosed insomnia, at doses of 600 milligrams daily and above, and with use for at least 8 weeks. For general quality of life, no significant effect was found.

For you that means: with genuine problems falling asleep, the data are at their best. With diffuse tiredness and no sleep problem, they are at their thinnest.

Cheah KL et al. PLoS One. 2021;16(9):e0257843. DOI: 10.1371/journal.pone.0257843
The possible mechanism for sleep In vivo, mouse

A Japanese group at the University of Tsukuba tested in mice in 2017 which component of ashwagandha leaves induces sleep. Surprisingly, it was not the alcoholic extract with its many withanolides, but the water extract with triethylene glycol that increased the proportion of deep sleep, and it did so in a dose-dependent way.

This is an animal study on leaves, not on the root, and in humans this pathway is not established. But it does show something important: the sleep-promoting fraction is possibly not the same as the fraction that manufacturers standardize.

Kaushik MK et al. PLoS One. 2017;12(2):e0172508. DOI: 10.1371/journal.pone.0172508

Where the studies are weak, named openly

Four limitations that appear in every meta-analysis
  • Small samples. The individual trials often include 50 to 80 people. A network meta-analysis of medicinal plants for anxiety names exactly this as a limitation for Withania somnifera.
  • Very large scatter. In several analyses the heterogeneity is above 80 percent. That means the studies are barely comparable with each other.
  • Short durations. Mostly 30 to 112 days. What happens after a year is practically unresearched. A systematic review from 2023 stresses that none of the included studies examined the long-term consequences of lowering cortisol on adrenal function.
  • Closeness to manufacturers. Many studies were carried out with the preparation of a particular manufacturer or supported by them. That does not make the results wrong, but it shifts expectations.
The mechanism, as far as it holds up RCT, n=60

An Australian group around Adrian Lopresti investigated 240 milligrams of a standardized extract daily over 60 days in 60 stressed adults in 2019.

What was observed was a reduction in anxiety scores, a clear drop in morning cortisol and, less noticed, a drop in DHEA-S as well. The authors conclude from this that the stress-easing effects could run via a dampening of the HPA axis.

For you the DHEA-S finding is the intriguing part. The plant does not appear to lower cortisol selectively, but rather to pull back the adrenal axis as a whole a little. In someone whose axis is already running flat, that is not a desirable goal.

Lopresti AL et al. Medicine (Baltimore). 2019;98(37):e17186. DOI: 10.1097/MD.0000000000017186

And now you know why, with ashwagandha, I speak neither of hype nor of humbug, but of a substance with a plausible address and a thin long-term file.

When ashwagandha does not make sense

This section is the most important one to me. Because with plants, contraindications are rarely discussed, as if herbal meant harmless. It does not.

Pregnancy and breastfeeding

Here the rule is: do not use. The data really are unclear, and that is exactly the reason. A systematic review from 2025 examined the historical reports of an abortifacient effect and concludes that they rest on chains of citations without a reliable primary source. Animal studies showed no reproductive toxicity at relevant doses. Reliable human data on pregnancy are likewise missing, though. In pregnancy, uncertainty is not an argument for an experiment.

Do not use

Thyroid: Hashimoto, hyperthyroidism, levothyroxine

Ashwagandha can interfere with the thyroid axis. In subclinical hypothyroidism, a small study shifted TSH, T3 and T4 toward the normal range. The same mechanism can pull in the wrong direction in hyperthyroidism or under levothyroxine. There is a case report of a painless thyroiditis with hyperthyroidism after two months of use, which receded after the preparation was stopped.

Only with medical guidance

Autoimmune disease and immunosuppressants

The plant is credited with immunomodulating properties. With an autoimmune disease, stimulating the immune system is not a neutral event, and under immunosuppressive therapy it could work against the treatment. Reliable studies on this combination are missing. Where data are missing, restraint is the more honest answer.

Caution

Sedatives, benzodiazepines, alcohol

In studies, ashwagandha showed sleep-promoting and dampening effects. In combination with substances that do the same, these effects can add up. That is relevant for driving, night shifts and working with machinery. With alcohol the rule also applies: do not combine.

Addition possible

Liver disease and planned surgery

With pre-existing liver disease or elevated liver values, ashwagandha does not belong on the shelf, more on that below. Before planned procedures it should be stopped, usually two weeks in advance, because dampening effects can meet anesthetic agents. Discuss this with the team that operates on you.

Stop beforehand

Nightshade allergy, children, adolescents

Withania somnifera belongs to the Solanaceae. With a known allergy or intolerance to nightshades, caution is warranted. For children and adolescents, controlled data are almost entirely missing, as they are for people with prostate cancer or hormone-dependent tumors.

No data

Liver injury: rare, but real

Case series from Iceland and the USA Case series, n=5

A group around Helgi Björnsson at Landspitali University Hospital in Reykjavik described five cases of liver injury after ashwagandha preparations in 2020, three from Iceland and two from the US Drug-Induced Liver Injury Network.

All five developed jaundice, along with nausea, fatigue and itching, after a latency of 2 to 12 weeks. The pattern of injury was cholestatic or mixed. In four of five, liver values recovered within 1 to 5 months after the preparation was stopped. A chemical analysis of the products found ashwagandha, but no other toxic admixtures.

Björnsson HK et al. Liver Int. 2020;40(4):825-829. DOI: 10.1111/liv.14393
Case series from India, the more serious side Case series, n=8

In 2023 a group around Cyriac Philips published the largest series to date: eight patients with liver injury after single-substance ashwagandha, out of 23 reported cases in total. The chemical analysis of the preparations showed only natural plant compounds, no adulteration.

The most frequent presentation was a cholestatic hepatitis. Five patients had a pre-existing chronic liver disease. Three of them developed acute-on-chronic liver failure and died in the course of it.

For you that means, very concretely: a known liver disease is not a side note, it is a reason to keep your hands off.

Philips CA et al. Hepatol Commun. 2023;7(10):e0270. DOI: 10.1097/HC9.0000000000000270
Stop immediately and have it clarified medically in case of
  • yellowing of the skin or of the white of the eyes
  • dark urine or strikingly pale stool
  • persistent itching without a rash
  • newly appearing nausea, upper abdominal complaints, unusual fatigue
Context, so the fear does not grow larger than the risk

Measured against use worldwide, these are rare events. A randomized safety study in 80 healthy adults found no notable changes in blood count, liver values or thyroid profile and no adverse events over 8 weeks at 300 milligrams twice daily.

Both are true. In most people nothing notable happens. In individual people something serious happens. Because we do not know in advance who belongs to that group, the decision belongs in a conversation and not in a purchase recommendation.

Verma N et al. Complement Ther Med. 2021;57:102642. DOI: 10.1016/j.ctim.2020.102642
On the regulatory side A regulatory review article from 2025 describes that ashwagandha is increasingly under observation in the EU. Most recently, a procedure under Article 8 of Regulation (EC) No 1925/2006 was recommended. At the end of that road there could be a restriction or a ban on its use in food supplements. Individual EU countries have already limited its use. Professionally, this assessment is discussed controversially. What follows from it for you: the legal status can change, and a product on the shelf is no proof of tested safety.

What belongs on the table before you swallow anything

If you are exhausted, ashwagandha is not the first step. The first step is the question: what is actually behind this? Exhaustion is a catch basin. At least a dozen different causes swim in it, and most of them are measurable.

The stress axis itself

  • Diurnal cortisol profile across several time points, not a single value
  • Morning rise after waking, methodologically tricky and only meaningful with documented times
  • DHEA-S as the second branch of the adrenal cortex
  • With abnormal findings, further endocrinological workup

The look-alike candidates

  • TSH, free T3, free T4, TPO antibodies
  • Ferritin and transferrin saturation, not just hemoglobin
  • Vitamin D, vitamin B12 or holotranscobalamin, folate
  • Fasting glucose and HbA1c
  • Inflammatory markers, kidney values, liver values, blood count
  • Sleep apnea screening, polygraphy if suspected
Why the measurement method matters so much with cortisol The morning rise in cortisol is one of the most sensitive measurements in stress research. The international society for psychoneuroendocrinology has published consensus guidelines for it and updated them in 2022. The core point: without objectively documented waking and sampling times, the value is hard to interpret. A saliva sample taken at some point in the morning is therefore almost worthless. If someone derives a stress diagnosis from a single cortisol value, healthy skepticism is warranted.

The differential diagnosis of sleep apnea deserves a sentence of its own. If your sleep is repeatedly interrupted at night without you noticing, you will wake up exhausted in the morning, no matter which plant you take. Snoring, observed pauses in breathing, morning headaches and a tendency to fall asleep during the day are reasons to have this clarified before any supplement is considered.

“Exhaustion is not a diagnosis. It is a question. And questions are answered with measurements, not with capsules.”

Shukri Jarmoukli, ViveCura Berlin

And now you know why, with me, the lab comes before the shelf.

When ashwagandha can be one building block

Let us assume the workup has been done. The look-alike candidates have been ruled out. What remains is a picture that fits an overactive stress axis, or a stubborn problem falling asleep. Then ashwagandha can be a building block. Not the solution. A building block.

Important for the following section What comes now are study dosages from the published literature. This is not a recommendation for intake and not a prescription for you. Whether, how much, for how long and in which form something makes sense in your case belongs in medical guidance that looks at your findings, your diagnoses and your medications.
StudyDosage in the study protocolDurationEndpoint studied
Chandrasekhar 2012, n=64300 mg high-concentration full-spectrum root extract, twice daily60 daysStress scales, serum cortisol
Lopresti 2019, n=60240 mg standardized extract, once daily60 daysAnxiety scores, cortisol, DHEA-S, testosterone
Langade 2019, n=60300 mg root extract, twice daily10 weeksSleep onset latency, sleep efficiency, anxiety
Langade 2020, n=80300 mg root extract, twice daily8 weeksSleep parameters in healthy people and in insomnia
Sharma 2018, n=50600 mg root extract daily8 weeksTSH, T3, T4 in subclinical hypothyroidism
Meta-analysis 2022, k=12most favorable range for stress: 300 to 600 mg dailypredominantly 8 weeksStress and anxiety scales
Meta-analysis 2021, k=5clearer sleep effects from 600 mg daily upwardfrom 8 weeksSleep quality and sleep quantity

Four practical points that are missing from package inserts

1

Quality before dose

A standardized extract with a declared withanolide content, a clear statement of the plant part and documented heavy metal testing. Without these three pieces of information, nobody knows what is in the capsule. The analysis of Ayurvedic internet products in JAMA showed that this question is not a formality.

2

Timing according to the goal

The studies are not consistent here. The stress trials mostly split into two doses or gave it in the morning. The sleep trials gave it twice daily. If it is about falling asleep, the evening is biologically plausible. If it is about resilience during the day, more speaks for the morning.

3

A time window instead of a permanent solution

The study durations end between 30 and 112 days. Over years there are no controlled data, and a systematic review stresses explicitly that nobody has examined the long-term consequences of a lasting reduction in cortisol on adrenal function. A limited time window with a clear goal therefore makes more sense than a subscription.

4

A withdrawal trial instead of belief

After a defined period, deliberately pause and observe what changes. An effect that is missed when you leave it out was one. An effect that is not missed was habit. That is the simplest and most honest test you can carry out yourself, ideally agreed with your doctor.

The ViveCura approach: rhythm first, then the plant

I look at exhaustion through four lenses. Nervous system, immune system, metabolism, hormonal system. They are connected, and a plant alone does not reach them.

The nervous system needs predictability. It learns from repetition when activity and when recovery are due. If waking times, light stimuli and meals are in a different place every day, the thermostat has no reference. The hormonal system builds cortisol and DHEA from cholesterol, along the same axis. Whoever dampens this axis permanently dampens both branches. The metabolism reacts sensitively to blood sugar swings, and every sharp dip is a stress signal for the hypothalamus. The immune system is the area in which a flattened cortisol rhythm was mirrored most clearly in the Adam meta-analysis, namely in inflammatory markers.

The basics that come before any supplement

  • Light: bright daylight early in the morning, dimmed light in the evening. That is the strongest timekeeper you have for free.
  • Sleep window: the same waking time on seven days, including the weekend. The rhythm is created in the morning, not in the evening.
  • Meal rhythm: fixed times, enough protein, no breakfast made of sugar. Blood sugar swings are stress that you do not experience as stress.
  • Movement dose: strength training two to three times a week, at a dose you come out of recovered. With a flat rhythm, more is not better.
  • Recovery as an appointment: not as leftover time. Twenty minutes without a screen and without a task are a signal to the nervous system, not laziness.
  • Closing measurable gaps: iron, vitamin D, B12, thyroid. Targeted and based on findings, not on a hunch.
The reframe this is really about

An adaptogen is no substitute for adaptability. At best it can open a window in which your system finds its way back more easily. If nothing happens inside that window, it closes again.

That is why the most honest question is not “which ashwagandha is the best”, but “what exactly is supposed to change over the next eight weeks, and how will I notice it”.

Three levers for this week

First: For seven days, note your waking time and how you experience the first hour, in one sentence. You often recognize more patterns in that than in a single lab panel.

Second: In the first 30 minutes after waking, go outside for 10 minutes, even when it is cloudy. That is the stimulus that trains the morning rise.

Third: If you are already taking ashwagandha and notice nothing, plan a deliberate withdrawal trial and raise the topic of a workup instead. Diurnal cortisol profile, thyroid, ferritin, vitamin D, B12, blood sugar, sleep apnea. This list costs you one conversation and can save you months.

Energy is not a luxury. Energy is the precondition for leading your life instead of merely administering it. That is why it deserves a diagnosis and not a guess.

Frequently asked questions about ashwagandha

What exactly is ashwagandha?

Ashwagandha is the root of Withania somnifera, a nightshade plant from South Asia. Its English common name is winter cherry, in German Schlafbeere, the sleep berry. In the Ayurvedic tradition it belongs to the Rasayana plants, so to the tonics. The best studied group of compounds are the withanolides, a family of steroidal lactones. Withanosides, alkaloids and flavonoids come on top of that. What ends up in the jar depends strongly on the plant part and on the extraction method. That is exactly what makes comparing studies and products difficult.

Does ashwagandha lower cortisol?

Across the studies taken together, a reduction in measured cortisol was observed. A meta-analysis from 2025 pooled seven randomized trials with 488 participants and found a statistically significant reduction of 1.16 micrograms per deciliter. The second finding of the same paper is the interesting one: on the Perceived Stress Scale there was no significant effect. A laboratory value can therefore move without your experience moving with it. For you that means: cortisol is a measured value, not a lived experience.

How solid is the evidence on ashwagandha?

There are several meta-analyses with favorable results for stress, anxiety and sleep. The authors themselves, however, limit clearly how much can be concluded from them. The individual trials are small, the scatter between them is very large, the durations are mostly 8 to 12 weeks, and many investigations were funded by the manufacturers of the tested extracts or carried out with their preparations. One large meta-analysis explicitly rated the certainty of the evidence for stress and anxiety as low. That is no reason to write ashwagandha off. It is a reason not to expect miracles.

Who is ashwagandha not suitable for?

Do not use it during pregnancy or while breastfeeding. Caution is warranted with thyroid conditions such as Hashimoto or an overactive thyroid, when taking levothyroxine, with autoimmune disease and immunosuppressive therapy, with sedatives, benzodiazepines and alcohol, with known liver disease or abnormal liver values, before planned surgery and with an allergy to nightshades. Reliable data are also missing for children and adolescents. In all of these cases the decision belongs in a medical conversation, not in a shopping cart.

Does ashwagandha make sense with Hashimoto or thyroid problems?

Particular restraint is appropriate here. A small randomized trial in 50 people with subclinical hypothyroidism observed a shift of TSH, T3 and T4 toward the normal range under 600 milligrams of root extract daily over 8 weeks. At first that sounds favorable. But this very effect can go in the wrong direction in an overactive thyroid or under levothyroxine. There is also a case report of a painless thyroiditis with hyperthyroidism after two months of ashwagandha, which regressed after the preparation was stopped. With Hashimoto the immunomodulating component is added, and that has not been studied sufficiently so far.

Can ashwagandha harm the liver?

There are documented cases. In 2020 a case series from Iceland and the US Drug-Induced Liver Injury Network described five people with liver injury after taking ashwagandha preparations, typically with jaundice and itching after 2 to 12 weeks. Liver values recovered within 1 to 5 months after the preparation was stopped. In 2023 a series from India with eight cases under single-substance preparations followed. There, three cases with pre-existing liver disease were fatal. A single case report from Heidelberg comes on top of that. Measured against how widely the plant is used, these are rare events. But they are real. Jaundice, dark urine, pale stool, persistent itching or severe fatigue while taking it are reasons to stop immediately and to have it clarified medically.

Which dosages have been studied?

These are study dosages from the literature, not a recommendation for you. The classic stress trial from 2012 used 300 milligrams of a high-concentration full-spectrum root extract twice daily over 60 days. An Australian investigation from 2019 worked with 240 milligrams of a standardized extract once daily over 60 days. A dose-response analysis found the most favorable range for stress at 300 to 600 milligrams per day. For sleep, a meta-analysis showed clearer effects from 600 milligrams daily upward and from 8 weeks of use. Which dose makes sense in your case, or does not make sense, belongs under medical guidance.

Take it in the morning or in the evening?

The studies give no clear answer on this, because they proceeded differently. The stress trials mostly worked with two doses spread across the day or with one dose in the morning. The sleep trials gave the extract twice daily over 8 to 10 weeks. If the goal is falling asleep, an evening dose is biologically plausible. If the goal is resilience during the day, more speaks for the morning. More important than the time of day is the question of whether there is any stress dysregulation at all.

What is the difference between KSM-66 and Sensoril?

Both are commercial, standardized extracts, as is Shoden. They differ in the plant part used, in the extraction method and in the declared withanolide content. That is why they are not simply interchangeable pharmacologically, and that is why studies using different extracts can only be pooled to a limited degree. A systematic review on cortisol named exactly this variety of formulations and dosages as a central reason for the contradictory results. For practice that means: a standardized extract with a declared withanolide content and documented heavy metal testing is better than cheap powder without any information.

Why did ashwagandha change nothing for me?

The most common reason is that the exhaustion did not come from an overactive stress axis. A flat diurnal cortisol curve, an iron deficiency, an untreated sleep apnea, a vitamin B12 deficiency, a thyroid disorder or a fragmented sleep-wake rhythm behave completely differently from acute ongoing stress. A plant that in studies mainly lowered measured cortisol can simply miss the point when the rhythm is already flat. So measure first, then decide.

Read on in the ViveCura guide

Ashwagandha never stands alone. These topics are physiologically attached to it.

SJ

Shukri Jarmoukli

Physician, Integrative Medicine · ViveCura Berlin

I work in my private practice in Berlin-Kreuzberg with people for whom standard diagnostics has not found an answer. My approach combines conventional medicine with clinical psychoneuroimmunology, functional diagnostics and lifestyle medicine. Not as an opposite to conventional medicine, but as an extension by levels that often have no place in the standard routine.

With topics like ashwagandha, I am less interested in what a plant promises and more in where in the system it acts and in whom that place is the problem at all.

Privatpraxis Shukri Jarmoukli · Skalitzer Straße 137, 10999 Berlin · vivecura.com

Sources

  1. Akhgarjand C, Asoudeh F, Bagheri A et al. Does Ashwagandha supplementation have a beneficial effect on the management of anxiety and stress? A systematic review and meta-analysis of randomized controlled trials. Phytother Res. 2022;36(11):4115-4124. DOI: 10.1002/ptr.7598 [Meta-analysis, k=12, n=1,002]
  2. Albalawi AA. Dual impact of Ashwagandha: Significant cortisol reduction but no effects on perceived stress. A systematic review and meta-analysis. Nutr Health. 2025;31(4):1395-1408. DOI: 10.1177/02601060251363647 [Meta-analysis, k=7, n=488]
  3. Arumugam V, Vijayakumar V, Balakrishnan A et al. Effects of Ashwagandha (Withania somnifera) on stress and anxiety: A systematic review and meta-analysis. Explore (NY). 2024;20(6):103062. DOI: 10.1016/j.explore.2024.103062 [Meta-analysis, k=9, n=558]
  4. Alsanie SA, Alhodieb FS, Askarpour M. Effects of ashwagandha (Withania somnifera) on mental health in adults: A systematic review and dose-response meta-analysis of randomized controlled trials. Complement Ther Med. 2026;97:103325. DOI: 10.1016/j.ctim.2026.103325 [Meta-analysis, k=22]
  5. Cheah KL, Norhayati MN, Husniati Yaacob L, Abdul Rahman R. Effect of Ashwagandha (Withania somnifera) extract on sleep: A systematic review and meta-analysis. PLoS One. 2021;16(9):e0257843. DOI: 10.1371/journal.pone.0257843 [Meta-analysis, k=5, n=400]
  6. Fatima K, Malik J, Muskan F et al. Safety and efficacy of Withania somnifera for anxiety and insomnia: Systematic review and meta-analysis. Hum Psychopharmacol. 2024;39(6):e2911. DOI: 10.1002/hup.2911 [Meta-analysis, k=5, n=254]
  7. Della Porta M, Maier JA, Cazzola R. Effects of Withania somnifera on cortisol levels in stressed human subjects: A systematic review. Nutrients. 2023;15(24):5015. DOI: 10.3390/nu15245015 [Systematic review, k=9]
  8. Zhang W, Yan Y, Wu Y et al. Medicinal herbs for the treatment of anxiety: A systematic review and network meta-analysis. Pharmacol Res. 2022;179:106204. DOI: 10.1016/j.phrs.2022.106204 [Network meta-analysis, k=29]
  9. Chandrasekhar K, Kapoor J, Anishetty S. A prospective, randomized double-blind, placebo-controlled study of safety and efficacy of a high-concentration full-spectrum extract of ashwagandha root in reducing stress and anxiety in adults. Indian J Psychol Med. 2012;34(3):255-262. DOI: 10.4103/0253-7176.106022 [RCT, n=64]
  10. Lopresti AL, Smith SJ, Malvi H, Kodgule R. An investigation into the stress-relieving and pharmacological actions of an ashwagandha (Withania somnifera) extract: A randomized, double-blind, placebo-controlled study. Medicine (Baltimore). 2019;98(37):e17186. DOI: 10.1097/MD.0000000000017186 [RCT, n=60]
  11. Langade D, Kanchi S, Salve J, Debnath K, Ambegaokar D. Efficacy and safety of ashwagandha (Withania somnifera) root extract in insomnia and anxiety: A double-blind, randomized, placebo-controlled study. Cureus. 2019;11(9):e5797. DOI: 10.7759/cureus.5797 [RCT, n=60]
  12. Langade D, Thakare V, Kanchi S, Kelgane S. Clinical evaluation of the pharmacological impact of ashwagandha root extract on sleep in healthy volunteers and insomnia patients: A double-blind, randomized, parallel-group, placebo-controlled study. J Ethnopharmacol. 2020;264:113276. DOI: 10.1016/j.jep.2020.113276 [RCT, n=80]
  13. Adam EK, Quinn ME, Tavernier R et al. Diurnal cortisol slopes and mental and physical health outcomes: A systematic review and meta-analysis. Psychoneuroendocrinology. 2017;83:25-41. DOI: 10.1016/j.psyneuen.2017.05.018 [Meta-analysis, 179 associations from 80 studies]
  14. Stalder T, Kirschbaum C, Kudielka BM et al. Assessment of the cortisol awakening response: Expert consensus guidelines. Psychoneuroendocrinology. 2016;63:414-432. DOI: 10.1016/j.psyneuen.2015.10.010 [Consensus guideline]
  15. Stalder T, Lupien SJ, Kudielka BM et al. Evaluation and update of the expert consensus guidelines for the assessment of the cortisol awakening response (CAR). Psychoneuroendocrinology. 2022;146:105946. DOI: 10.1016/j.psyneuen.2022.105946 [Review, guideline update]
  16. Cadegiani FA, Kater CE. Adrenal fatigue does not exist: A systematic review. BMC Endocr Disord. 2016;16(1):48. DOI: 10.1186/s12902-016-0128-4 [Systematic review, k=58]
  17. Björnsson HK, Björnsson ES, Avula B et al. Ashwagandha-induced liver injury: A case series from Iceland and the US Drug-Induced Liver Injury Network. Liver Int. 2020;40(4):825-829. DOI: 10.1111/liv.14393 [Case series, n=5]
  18. Philips CA, Valsan A, Theruvath AH et al. Ashwagandha-induced liver injury: A case series from India and literature review. Hepatol Commun. 2023;7(10):e0270. DOI: 10.1097/HC9.0000000000000270 [Case series, n=8]
  19. Tóth M, Benedek AE, Longerich T, Seitz HK. Withania somnifera-induced acute liver injury: A case report. Clin Case Rep. 2023;11(3):e7078. DOI: 10.1002/ccr3.7078 [Case report]
  20. Sharma AK, Basu I, Singh S. Efficacy and safety of ashwagandha root extract in subclinical hypothyroid patients: A double-blind, randomized placebo-controlled trial. J Altern Complement Med. 2018;24(3):243-248. DOI: 10.1089/acm.2017.0183 [RCT, n=50]
  21. Hayashi M, Hamada H, Azuma SI, Hayashi K. Painless thyroiditis by Withania somnifera (Ashwagandha). Cureus. 2024;16(3):e55352. DOI: 10.7759/cureus.55352 [Case report]
  22. Verma N, Gupta SK, Tiwari S, Mishra AK. Safety of ashwagandha root extract: A randomized, placebo-controlled study in healthy volunteers. Complement Ther Med. 2021;57:102642. DOI: 10.1016/j.ctim.2020.102642 [RCT, n=80]
  23. Wiciński M, Fajkiel-Madajczyk A, Kurant Z et al. Can ashwagandha benefit the endocrine system? A review. Int J Mol Sci. 2023;24(22):16513. DOI: 10.3390/ijms242216513 [Mechanism review]
  24. Paul S, Chakraborty S, Anand U et al. Withania somnifera (L.) Dunal (Ashwagandha): A comprehensive review on ethnopharmacology, pharmacotherapeutics, biomedicinal and toxicological aspects. Biomed Pharmacother. 2021;143:112175. DOI: 10.1016/j.biopha.2021.112175 [Review]
  25. Shinde S, Balasubramaniam AK, Mulay V et al. Recent advancements in extraction techniques of ashwagandha (Withania somnifera) with insights on phytochemicals, structural significance, pharmacology, and current trends in food applications. ACS Omega. 2023;8(44):40982-41003. DOI: 10.1021/acsomega.3c03491 [Review]
  26. Brendler T, Al-Mondhiry R, Lang L et al. Ashwagandha: Is it safe? Part 1: A regulatory review. Phytother Res. 2025 (online ahead of print). DOI: 10.1002/ptr.70151 [Review, regulatory]
  27. Tallon MJ, Koturbash I, Blum JL. A systematic and ethnobotanical review of ashwagandha's (Withania somnifera) teratogenic and abortifacient potentials. Phytother Res. 2025. DOI: 10.1002/ptr.70079 [Systematic review]
  28. Saper RB, Phillips RS, Sehgal A et al. Lead, mercury, and arsenic in US- and Indian-manufactured Ayurvedic medicines sold via the Internet. JAMA. 2008;300(8):915-923. DOI: 10.1001/jama.300.8.915 [Cross-sectional analysis, 193 products]
  29. Kaushik MK, Kaul SC, Wadhwa R, Yanagisawa M, Urade Y. Triethylene glycol, an active component of ashwagandha (Withania somnifera) leaves, is responsible for sleep induction. PLoS One. 2017;12(2):e0172508. DOI: 10.1371/journal.pone.0172508 [In vivo, mouse]
On how to read the evidence: The meta-analyses on ashwagandha cited here rest on small individual trials with high scatter, short durations and frequent manufacturer involvement. The authors themselves rate the certainty of the evidence predominantly as low. For long-term use over several years, for children and adolescents, and for the combination with autoimmune disease and immunosuppressants, no reliable controlled data are available. The statements in this text are perspectives and ways of putting things in context, not treatment instructions. This article does not replace a medical conversation or individual diagnostics. With existing conditions or ongoing medication, every decision about herbal preparations belongs under medical guidance.

Have questions or want to book an appointment?

We'd be happy to advise you personally at our practice.

Book appointment