Rhodiola rosea

What roseroot is

Rhodiola rosea grows in cold, high-altitude regions of the Northern Hemisphere. The root and rhizome are used; when freshly cut, they smell of roses. Two groups are considered the active constituents: the phenylethanoid salidroside, also called rhodioloside, and the cinnamyl alcohol glycosides rosavin, rosin and rosarin.

In the scene, roseroot is considered the stimulating counterpart to ashwagandha. This is not an invention: the reported side effects include nervousness and insomnia, not sedation.

What the authority made of it

In Europe, there are two routes for herbal medicinal products. One is called traditional use: demonstrate at least 30 years of medicinal use and then register, without proving efficacy. The other is called “well-established use” and requires exactly that proof.

Roseroot took the first route. The EU monograph of March 20, 2024 lists dry extract as a traditional herbal medicinal product, daily dose 144 to 400 milligrams, from age 18. In Germany, two such products have been registered since 2014 and 2016.

The authority closed the second route. The assessment report states that the published studies had considerable quality shortcomings, and that “well-established use” could therefore not be accepted. To this day, there is no Cochrane review on roseroot.

One root, two extract families

Anyone going through the studies on exhaustion comes across a pattern. The placebo-controlled studies almost all use the same extract, SHR-5: Darbinyan 2000 in 56 physicians on night duty, Spasov 2000 in 40 students, Shevtsov 2003 in 161 cadets, Olsson 2009 in 60 people with exhaustion, Darbinyan 2007 in 89 people with depression. In three of them, the same scientist from the institute behind the extract is a co-author.

The second product family, WS 1375, has a different problem: it was almost never tested against placebo. Edwards 2012 with 101 participants, Kasper and Dienel 2017 with 118 burnout patients, Lekomtseva 2017 with 100 people with exhaustion, Koop 2020 with 50 participants — all open-label, single-arm studies. Cropley 2015 in 80 mildly anxious people had a comparison group, but an untreated one.

The independent review from Exeter summed up the result in 2011 in one sentence: independent replications of the individual studies are lacking. The second review, Ishaque 2012, screened 206 papers, included 11 with 446 participants and found two of six studies positive for physical fatigue and three of five for mental fatigue. None met the CONSORT standard. An assessment of 39 studies on all Rhodiola species came to the same result.

Endurance: better supported than its reputation

In one area, the picture is better than the usual summary suggests. A 2025 meta-analysis evaluated 26 randomized trials with 668 healthy participants, mean duration 33 days: VO2max ES 0.32, time to exhaustion ES 0.38, time trial performance ES -0.40 — all significant, all small. The antioxidant markers were clearer, superoxide dismutase ES 1.16 and malondialdehyde ES -1.21. There was no effect on interleukin-6 and C-reactive protein.

Two limitations belong with this. Only one of the ten individual analyses is based on more than ten studies, and heterogeneity remains high. And the title refers to Rhodiola rosea, while the reference list also includes studies with Rhodiola crenulata.

Rosavins, salidroside and what is really in the jar

The rule of thumb that a good extract contains 3 percent rosavins and 1 percent salidroside is on almost every package. It has a real origin: this is how the extract in the exercise study De Bock 2004 was specified, and Mao 2015 worked with 3.07 percent rosavin and 1.95 percent rhodioloside. But the effect on exhaustion is based on other products: in Darbinyan 2000, 170 milligrams of SHR-5 contained about 4.5 milligrams of salidroside. The figures are a market convention, not an efficacy optimum.

The rosavins matter for a different reason. Salidroside occurs in many Rhodiola species, rosavin practically only in Rhodiola rosea. It is therefore not a measure of efficacy but proof of authenticity. If it is missing, another species was probably in the pot, usually Rhodiola crenulata.

That is exactly what is found regularly. Of around 40 European commercial products, about one fifth contained no rosavin; of the rest, around 80 percent were below the content of the registered medicinal products. Among 13 food supplements tested, almost 60 percent lacked the declared amount or the markers, and rosavin was detectable in only 9, 4 of them in traces. An analysis of the US market measured rosavins between 0.01 and 3.08 percent — plus one product with apparently synthetically added, undeclared salidroside.

Since February 2023, the genus Rhodiola has been listed in Appendix II of the CITES endangered species convention. This covers raw material and extracts, not finished packaged end products. The supply chain has become tighter and more expensive — the incentive to adulterate no smaller.

What the studies show

The largest controlled study: 161 cadets

Randomized, double-blind, placebo-controlled, with an additional untreated group. 161 cadets aged 19 to 21 received a single dose of SHR-5, 41 people 370 milligrams and 20 people 555 milligrams. Mental performance under fatigue was measured as an antifatigue index: 1.0385 and 1.0195 versus 0.9046 on placebo, both significant, no difference between the doses. A later statistical review found misprints and mix-ups in it.

Exhaustion syndrome over four weeks

60 men and women aged 20 to 55 with diagnosed exhaustion syndrome, 30 each on 576 milligrams of SHR-5 daily or placebo, 28 days. The primary endpoint was the Pines burnout scale: 4.27 to 4.01 on active treatment, 4.33 to 4.26 on placebo, p equal to 0.047 — narrowly reached. Two of five attention indices also improved. The data of at least five participants were excluded; the analysis was per protocol.

Rhodiola versus sertraline versus placebo

Phase II study in 57 patients with mild to moderate depression over 12 weeks, three arms. The Hamilton score fell by 8.2 points on sertraline, by 5.1 on Rhodiola and by 4.6 on placebo; no group difference was significant. Tolerability was clear: 63.2 percent adverse events on sertraline, 30.0 on Rhodiola, 16.7 on placebo. Rhodiola is less effective but has the more favorable benefit-risk ratio, according to the authors.

The burnout study cited as evidence

118 outpatients with burnout symptoms took 400 milligrams of WS 1375 daily for 12 weeks. Most measures improved, several after just one week. The catch lies in the design: no control group, no blinding, no placebo arm. The authors themselves call the study exploratory. It is not suitable as proof of efficacy, even though it is cited as such.

Where the data stop

The weakness of this evidence base is less that the studies are small — though they are. It is more structural: independent replications are lacking. The placebo-controlled studies on exhaustion almost all use the same extract, some with a co-author from the manufacturer’s institute; the second product family left out the control group. On top of that comes criticism of details: in the 2007 depression study, the placebo score remained almost unchanged at 24.17 to 23.41, which the authority considers atypical and takes as grounds for regarding the significance results as questionable. The longest controlled study lasts 12 weeks. Data on genotoxicity are missing entirely. And the adaptogen concept itself comes from animal experiments; in humans, only an altered cortisol awakening response has been measured so far.

Status, approval and legal

In Germany, roseroot dry extract has been registered in two products as a traditional herbal medicinal product since 2014 and 2016 — registration, not approval, single dose 200 milligrams, daily dose 400 milligrams, from age 18. The EU monograph refers to a physician if complaints persist for more than two weeks. As a food, there is no authorized health claim: 1,548 botanical claims are on hold in the EU pending a decision by the Commission and the member states.

Safety

The monograph lists headache, nervousness, insomnia, dizziness, nausea, vomiting, skin rash and itching, frequency unknown in each case. As a contraindication, it names only hypersensitivity. Use under 18 and during pregnancy and breastfeeding is not recommended because data are lacking. The monograph states that no clinically relevant interactions have been observed — but the assessment report names signals: a 21 percent change in the ratio of losartan metabolite to parent compound after 14 days in 13 volunteers, individual reports with sertraline, paroxetine, duloxetine and amitriptyline, and one suspected case of serotonin syndrome with paroxetine. So if you take antidepressants, discuss it. The widespread advice to avoid it in case of manic tendencies, on the other hand, is not supported: in the systematic review of 35 case reports of herb-associated mania, Rhodiola does not appear.

Sources

• EMA, European Union herbal monograph on Rhodiola rosea L., rhizoma et radix, Revision 1, March 20, 2024
• EMA/HMPC/24186/2023, Assessment report on Rhodiola rosea L., rhizoma et radix, Revision 1
• Ishaque et al., BMC Complementary and Alternative Medicine 2012 (PMID 22643043)
• Hung, Perry & Ernst, Phytomedicine 2011 (PMID 21036578)
• Wang et al., Frontiers in Nutrition 2025 (PMID 41080184)
• Shevtsov et al., Phytomedicine 2003 (PMID 12725561)
• Olsson, von Schéele & Panossian, Planta Medica 2009 (PMID 19016404)
• Mao et al., Phytomedicine 2015 (PMID 25837277)
• Booker et al., Phytomedicine 2016 (PMID 26626192)
• Porwollik & Jafari, PLoS ONE 2026 (PMID 41544018)