The 2022 Geng meta-analysis pooled 15 randomized controlled trials and found sea buckthorn significantly improved blood lipids: triglycerides, total cholesterol, and LDL down, HDL up, with the subgroup analysis finding the reduction only in participants who started with abnormal lipid metabolism. The same analysis found no significant effect on blood glucose, blood pressure, or BMI. This page gives both halves equal weight, because the nulls are what make the positives believable.
What pooling 15 trials actually means
This is the best-supported entry in the evidence library, so it earns a slow read. A meta-analysis averages many trials into one estimate, trading each study’s noise for a shared signal. Fifteen trials is a decent pool for a supplement, and pooling matters here because any one trial could be dismissed as a fluke. Together, they either converge on a real effect or on a shared bias, which is why the corroboration below matters.
The positives, and who drove them
Four lipid markers moved significantly: triglycerides down, total cholesterol down, LDL down, HDL up.
The subgroup finding is the heart of the paper. The subgroup analysis found the reduction only in participants who entered with abnormal blood lipids, and no reduction in healthy volunteers with normal levels. Sea buckthorn behaved like a corrector of abnormal lipids, not an improver of normal ones, and that pattern repeats so consistently across the pool that it shapes our advice on the cholesterol page: normal labs, no reason to bother.
The subgroup logic, and why it matters for you
The corrector-not-improver pattern is the single most useful sentence in the paper, because it tells you which reader you are. It also has an independent echo. In a 2013 crossover (Larmo 2013) that fed 80 overweight women dried sea buckthorn berries, sea buckthorn oil, a phenolic extract and bilberries in four one-month periods, each diet’s effect on the serum metabolite profile differed by the woman’s baseline metabolic profile, though most single-metabolite changes were not significant after correction for multiple testing. That is one small study, not a confirmation, and it points the same direction as the pooled subgroups. Baseline biology shapes the response, which is why the heart hub routes people by their labs rather than by their hopes.
The nulls, printed in the same size
The same 15-trial pool found no significant effect on:
- Blood glucose
- Blood pressure
- BMI
These are not footnotes. They are the same analysis and the same methods, answering three other questions with “no.” Any page or label quoting the lipid half of Geng 2022 without the null half is clipping the paper to fit a sales pitch. Our metabolic health page maps all four outcomes marker by marker.
The numbers in everyday terms
A pooled lipid shift is a lab report, not a life outcome, and the translation matters. Picture two readers. The first has normal cholesterol and normal triglycerides: the subgroup analysis found no reduction in healthy people, and no amount of capsule loyalty changes that. The second has abnormal lipids their doctor is already watching: the pooled trials say sea buckthorn moved the panel in the right direction, on average. A shift in a marker is not a change in risk, and no sea buckthorn trial has ever counted a heart attack prevented. The same logic governs the triglyceride result: it is a number on a lab draw, valuable as a trend line, silent as a guarantee. For that second reader, and only that reader, an oil at an amount their doctor agrees could be worth the shelf space, so see how the oils compare before you pick a bottle.
The quality caveats, because they travel with the numbers
- Unknown size. The abstract gives no sample sizes, so we cannot say how many people the pooled trials enrolled.
- Unknown follow-up. The abstract gives no follow-up length, so nothing here speaks to years of use.
- Markers, not events. Cholesterol numbers predict risk, but they are not heart attacks or strokes. No trial has followed sea buckthorn users to actual cardiovascular events.
- Heterogeneity. The review itself reports high heterogeneity, even within its subgroups, so the pooled estimate averages over real differences between the 15 trials.
Heterogeneity deserves one more paragraph, because it is where the pooled average can mislead. Across the 15 trials, the review reports high heterogeneity, and the participants ranged from healthy volunteers to people with abnormal lipids. The meta-analysis handles that statistically, and the subgroup finding survived it, but a pooled estimate is still an average over genuinely different experiments.
Why the 2017 meta-analysis matters
Five years earlier, Guo 2017 pooled 11 randomized trials on blood lipids and reached the same lipid conclusions. Two research groups, one direction, until a 2026 meta-analysis pulled the other way, as the next section shows. Independent convergence is what moves a finding from “interesting” toward “strongest evidence this plant has,” and it is why this page carries a moderate grade and not a weaker one, even though a 2026 meta-analysis found no significant pooled effect on any lipid. What holds it back from strong is the high heterogeneity, the unknown size and length of the pooled trials, and that it remains, in absolute terms, a supplement evidence base rather than a drug trial program. The doses used in the single trials are tabulated on the study dosages page, and the other human trials in our registry sit in the clinical trials catalog.
When a third meta-analysis and a trial disagree
Two later results cut against the lipid story, and they belong beside it. A 2026 meta-analysis pooled 12 randomized trials (901 people analyzed, across seven countries) of sea buckthorn whole berries, purees, powders, extracts and oils. It found no significant pooled effect on total cholesterol, LDL-C, HDL-C, triglycerides, systolic blood pressure or BMI. Its abstract rates every outcome very low certainty because of risk of bias, heterogeneity and imprecision, and says its trial sequential analysis did not reach the required information size. We could read only that abstract.
A single trial points the same way. Zhou 2022 gave 90 ml of sea buckthorn puree or placebo to 111 people with high cholesterol for 90 days and found that the puree did not affect total cholesterol, LDL-C or triglycerides. It was a puree and not an oil, from one research group, and we could read only its abstract.
Neither result cancels the 2022 and 2017 analyses. They mean the lipid evidence is split, which is why this page stays at a moderate grade for lipid markers and no higher, and why it stays there until the full text of the 2026 analysis can be read.
How this site spends that grade
A moderate grade for lipid markers means three things here and only three. The cholesterol page may say the evidence supports an improvement in lipid markers for people with abnormal levels. The blood pressure page may not borrow the grade, because the pool said no. And the weight loss page may not borrow it either, because the BMI null came from the same analysis as the lipid positive. One analysis, four answers, each honored at full strength.
What would settle this
The definitive study would enroll thousands of people with abnormal lipids, follow them for years, and count cardiovascular events rather than lab panels. No supplement maker has ever funded a trial like that, for any supplement, so the realistic hope is narrower: a 12-month trial with abnormal-baseline participants, pre-registered lipid endpoints, and a follow-up after stopping to see whether the shift persists. Until either exists, the claim ends at marker improvements, and the reading-studies page shows how to spot anyone claiming more.
Frequently asked questions
What did the 2022 sea buckthorn meta-analysis find?
Pooling 15 randomized controlled trials, it found sea buckthorn supplementation significantly reduced triglycerides, total cholesterol, and LDL cholesterol while raising HDL. Its subgroup analysis found those lipid improvements only in participants who started with abnormal blood lipids. It found no significant effect on blood glucose, blood pressure, or BMI.
How strong is the sea buckthorn cholesterol evidence?
The strongest this plant has. Two meta-analyses by different research groups (15 trials in 2022, 11 trials in 2017) reached the same lipid conclusions, though a 2026 meta-analysis of 12 randomized trials found no significant pooled effect on any lipid, with very low certainty (we could read only its abstract), and a 2022 placebo-controlled trial in 111 people with high cholesterol found no change in total cholesterol, LDL-C or triglycerides. The caveats are structural: the pooled trials differ widely, the abstracts give no trial size or follow-up length, and the outcomes are lab numbers rather than heart attacks or strokes. We grade it moderate for lipid markers, held back by high heterogeneity, and not for outcomes.
If it lowers cholesterol, why not blood pressure too?
Because they are different physiology, and markers move independently. Cholesterol synthesis and transport respond to dietary fats and plant compounds; blood pressure is governed by vessels, kidneys, and the nervous system. The same meta-analysis looked for a pressure effect and did not find one, which is why our blood-pressure page carries the null rather than the borrowed hope.
Illustration.