Spirulina, Audited: The Trials, the Protein and the B12

Spirulina is a cyanobacterium, Arthrospira platensis, grown in warm alkaline water and dried into a dark blue-green powder. The spirulina benefits that have actually been measured in people form a short list: blood lipid fractions, haemoglobin and red cell indices, oxidative stress markers around hard exercise, and nasal symptoms in allergic rhinitis. Everything else in the marketing is untested.

Three things a careful European buyer needs are missing from every first-page result. The protein figure is quoted per 100 g when you take half a gram. The vitamin B12 figure is mostly an analogue humans cannot use. And spirulina is grown in open ponds, which makes cyanobacterial contamination the real quality question. This page covers all three, then lists every trial with its dose, size and duration.

What spirulina is, and four things it is not

Spirulina is the dried biomass of Arthrospira platensis, harvested from open ponds or closed photobioreactors at a culture pH of around 10. That alkalinity is why it can be grown outdoors at all: it suppresses most competing organisms. Jars say Spirulina and papers say Arthrospira for the same material.

It is not an alga. The phrase spirulina algae benefits is a misnomer that has outlived its correction. Spirulina is a prokaryote, a photosynthetic bacterium with no nucleus. Chlorella, its usual shelf neighbour, is a true green alga and a eukaryote, with a tough cellulose wall that has to be cracked before digestion. They are not close relatives and their composition is not interchangeable.

It is not a seaweed. Spirulina is a freshwater organism, so the iodine question that drives European debate about kelp, wakame and bladderwrack does not apply to it. If you are worried about iodine, spirulina is the wrong product to worry about.

It is not an adaptogen. Spirulina is a food organism with a nutrient profile, not a stress-axis herb, and it does not sit inside that pharmacological definition. What an adaptogen is, and which ones have human trials behind them, is covered in our guide to adaptogens. We say this once and move on.

Blue spirulina is not spirulina. Products sold as blue spirulina are extracted C-phycocyanin, the blue pigment protein. ANSES reports phycocyanin levels of 5 to 15 per cent of spirulina dry matter, so a blue extract is a fraction of the organism, not the whole one, and none of the trials below used it.

How we read a spirulina trial before we quote it

Six things decide whether a spirulina study means anything. Every trial in the next section is described against all six.

Trial size. The exercise trials run from nine to seventeen participants. The largest anaemia trial here enrolled 179 preschool children. Nothing in this literature is large.

Population. Recreational cyclists, elite rugby players, Cretan outpatients with fatty liver, Californian over-fifties, Cambodian preschoolers. These groups barely overlap.

Dose and duration. Doses cluster at 2 to 6 g a day, far above what any capsule delivers. Durations run from 14 days to six months. If your daily amount is a fifth of the trial dose, you are not running the trial.

What was compared. Several of the most quoted studies are single-arm, before-and-after designs with no placebo group, which folds in everything else that changed over six months.

Who supplied the material. The 2011 anaemia and immune-marker study lists an author employed by a spirulina producer. That does not invalidate it, and a page quoting the result should tell you.

What was not measured. Energy, hair, skin and mood, the four things most people are searching for, appear in none of these trials.

What the human trials on spirulina measured

Blood lipids

Mazokopakis and colleagues (2014, Annals of Gastroenterology) gave 6 g of Greek-produced spirulina daily for six months to fifteen Cretan outpatients with non-alcoholic fatty liver disease, thirteen of them men, median age 48. They reported mean falls of 37.5 per cent in ALT, 38.5 per cent in AST, 9.1 per cent in total cholesterol and 24.8 per cent in triglycerides, and a mean fall of 8.1 per cent in body weight. Sonographic findings did not change. Fifteen people, no placebo arm, and a group whose body weight also fell: this is a pilot, and its authors called it one.

The pooled picture is stronger than any single trial. Rahnama and colleagues (2023, Pharmacological Research) ran a GRADE-assessed meta-analysis of 20 randomised trials, 23 arms and 1,076 participants, reporting reductions in LDL cholesterol, total cholesterol and triglycerides of about 0.6 standardised mean difference and a smaller rise in HDL. It cannot tell you which population that applies to, because most included trials recruited people with an existing metabolic diagnosis.

Haemoglobin, red cell indices and anaemia

Selmi and colleagues (2011, Cellular and Molecular Immunology) supplemented 40 volunteers aged 50 or over for 12 weeks; 30 completed. They reported a steady rise in mean corpuscular haemoglobin in both sexes, rises in mean corpuscular volume in men, and increases in white blood cell count and indoleamine 2,3-dioxygenase activity at weeks 6 and 12. There was no placebo group and the authors explicitly asked for randomised trials to follow.

Barennes and colleagues (2022, BMC Pediatrics) ran a single-blinded placebo-controlled crossover trial in 179 Cambodian preschool children at 2 g a day. Anaemia counts fell in both arms, further on spirulina than on placebo, and weight gain differed only as a trend. That trial addresses undernourished children in a low-income setting, not an adult in Zurich or Manchester.

Exercise, oxidative stress, and the trial that found nothing

Kalafati and colleagues (2010, Medicine and Science in Sports and Exercise) gave nine moderately trained men 6 g a day for four weeks in a double-blind placebo-controlled crossover. After a two-hour treadmill run, time to fatigue at 95 per cent of maximal oxygen uptake was longer on spirulina, carbohydrate oxidation fell by 10.3 per cent, fat oxidation rose by 10.9 per cent, and reduced glutathione was higher at rest and at 24 hours. Nine men.

Chaouachi and colleagues (2022, Journal of Human Nutrition and Dietetics) supplemented seventeen elite male rugby players in Rennes with 5.7 g a day for seven weeks against an isoproteic placebo. F2-isoprostanes, C-reactive protein and creatine kinase rose after exhaustive exercise in the placebo group and did not rise in the spirulina group.

Then the trial that most pages skip. Ali, Aubeeluck and Gurney (2024, Journal of Dietary Supplements, University College London) gave 6 g a day for 14 days to seventeen recreationally active cyclists in a double-blind randomised crossover with a 14-day washout. Haemoglobin rose 3.4 per cent against placebo. Maximal oxygen uptake, peak work rate, time to fatigue, heart rate, respiratory exchange ratio and blood lactate all showed no difference whatsoever. The blood marker moved and the performance did not.

Trial	Population	Dose	Duration	Design	Measured
Mazokopakis 2014	15 adults with fatty liver	6 g/day	6 months	Single arm	Liver enzymes, lipids, body weight
Selmi 2011	40 adults aged 50 plus	Supplement	12 weeks	Single arm	Red cell indices, white cell count
Barennes 2022	179 preschool children	2 g/day	Crossover	Placebo controlled	Anaemia, growth
Kalafati 2010	9 trained men	6 g/day	4 weeks	Crossover, blinded	Fatigue time, glutathione
Chaouachi 2022	17 elite rugby players	5.7 g/day	7 weeks	Placebo controlled	Creatine kinase, isoprostanes
Ali 2024	17 recreational cyclists	6 g/day	14 days	Crossover, blinded	Haemoglobin, no performance change

Allergic rhinitis is the fourth measured area. Cingi and colleagues (2008, European Archives of Oto-Rhino-Laryngology) ran a double-blind placebo-controlled trial and reported better symptom and physical-finding scores for nasal discharge, sneezing, congestion and itching. Mao, Van de Water and Gershwin (2005, Journal of Medicinal Food) fed 2,000 mg a day for 12 weeks in a randomised crossover and reported interleukin-4 production from stimulated cells 32 per cent lower, with no change in the interferon-gamma or interleukin-2 arms. Two small trials, one symptom score, one laboratory measure.

What to take from this section: the doses that produced measurable changes were 2 to 6 g a day, sustained for weeks. Anything that promises the same from a single small capsule is describing a different experiment.

The spirulina protein arithmetic, per gram and per capsule

ANSES puts the protein content of spirulina at 60 to 70 per cent of dry matter, with essential amino acids at about 47 per cent of that protein. That is a genuinely high density and it is the number every page quotes.

It is also the wrong unit. Per 100 g is not how anyone eats this. Do the arithmetic in the unit you actually swallow:

• One gram of powdered spirulina carries 0.6 to 0.7 g of protein.
• A 500 mg capsule or a 500 mg spirulina tablet carries roughly 0.30 to 0.35 g.
• Six grams a day, the highest dose in any trial above, carries 3.6 to 4.2 g. That is less protein than one egg.

So spirulina protein is real and it is not a protein source. Treat it as a micronutrient-dense food ingredient with a pigment profile, and buy your protein somewhere cheaper by the gram.

The B12 question, answered plainly

Spirulina contains 0.1 to 2 mg of B12-like compounds per kilogram, and a lot of labels print that. Watanabe and colleagues (1999, Journal of Agricultural and Food Chemistry) purified those compounds from spirulina tablets and characterised them by thin-layer chromatography, reversed-phase HPLC and nuclear magnetic resonance. The major analogue, 83 per cent of the total, was pseudovitamin B12: a corrinoid that does not bind human intrinsic factor and therefore is not absorbed as B12. Only the minor 17 per cent fraction was the real thing.

ANSES states the consequence directly in its 2017 opinion: the B12 in spirulina is mostly an inactive analogue, and spirulina is not a reliable B12 source for people avoiding animal foods. The opinion also notes that supplementing children who were B12 deficient with spirulina did not correct their macrocytic anaemia.

If you eat no animal foods, take a B12 supplement containing cyanocobalamin or methylcobalamin, and read any B12 figure on a spirulina label as an artefact of the microbiological assay rather than a nutrient you will absorb.

Iron in spirulina: real, but not the iron you think

Iron in spirulina is genuine and it is not haem iron. Masten Rutar and colleagues (2022, Foods) analysed 46 spirulina supplements on the Slovenian market and found iron ranging from 0.28 to 3.48 g per kilogram, a more than tenfold spread between products with similar labels. More usefully, they found that 82 to 92 per cent of that iron was present as Fe(III), the ferric form, whose absorption is markedly lower than ferrous iron.

The same study found that about 86.7 per cent of products carried declarations of elemental content that did not match the analysis, some deviating by more than 45 per cent. A label figure for iron on a spirulina jar is therefore a weak guide to what is in it, and a much weaker guide to what you will absorb. Iron in the specific context of our own products is covered in shilajit for women, and the mineral-carrier chemistry sits in fulvic acid benefits.

Microcystins, trace metals, and how to read a certificate

This is the section the first page of results does not write, and it is the one that should decide your purchase.

Spirulina is grown in open ponds. Open ponds can be colonised by other cyanobacteria, including microcystin producers such as Microcystis. ANSES notes that Arthrospira itself has essentially not been shown to produce cyanotoxins, so any microcystin in a spirulina product is contamination rather than the organism. That is a manufacturing problem, and manufacturing problems are fixable and auditable.

The measured picture is mixed, which is the honest way to put it. Vichi and colleagues (2012, Food and Chemical Toxicology, Istituto Superiore di Sanità) tested 17 blue-green algae supplement brands on the Italian market and found the spirulina-only samples free of contamination, with microcystins confined to Aphanizomenon flos-aquae products at up to 5.2 micrograms of MC-LR equivalents per gram. A decade later, Rhoades and colleagues (2023, Microorganisms) examined five retail spirulina products and foods and detected microcystins in all five at levels that could take a consumer past recommended daily limits, alongside Bacillus cereus and Klebsiella pneumoniae. The authors attribute both findings to open-pond production.

The reference number to hold on to: the World Health Organization tolerable daily intake for microcystins in chronic exposure is 0.04 micrograms per kilogram of body weight per day. ANSES concluded in 2017 that a specific threshold for microcystins in spirulina supplements still needs to be set, taking other dietary intake into account. There is, in other words, no harmonised European maximum for this contaminant in supplements yet.

So what do you check before buying spirulina powder or tablets?

Batch-level testing, not brand-level. Vichi’s Italian survey found up to 50-fold variation between batches of the same brand. A certificate that is not tied to your batch number is decoration.

The analytes, named. Microcystins, plus cadmium, lead, mercury and arsenic. The Slovenian survey found those four metals below regulatory maxima across 46 products, which is reassuring and is not a substitute for the certificate on your jar.

The method. LC-MS/MS or ELISA for microcystins. If a seller cannot name the method, they have not seen the report either.

A traceable supply chain. ANSES’s headline consumer advice is to buy through channels controlled by public authorities, with traceability and an identifiable manufacturer. That is the whole recommendation, and it is a purchasing rule rather than a health claim.

Organic is not a contamination test. An organic certificate covers inputs and cultivation practice. It does not certify microcystin content. Buy organic if you want the cultivation standard; ask for the toxin panel separately.

Powder, tablets or capsules, and how people take it

Bags sold as spirulina algae powder and jars sold as spirulina powder are the same material. Tablets, sometimes sold as spirulina pills, are that powder compressed with a binder. Capsules are that powder in a shell. The format changes the dose you will realistically take and how much of the taste you meet, and nothing else.

The taste is the practical variable. Spirulina in powder form is strongly savoury and marine, and most people who quit quit because of that, not the pill count. Powdered spirulina disappears into a strongly flavoured base and does not disappear into water. Tablets and capsules bypass the taste and make the trial doses almost impossible to reach: 6 g a day is twelve 500 mg capsules.

Our own spirulina capsules are a maintenance-sized daily amount, not a way of reproducing a 6 g research protocol, and we would rather say so.

Honest limits: who should not take spirulina

This is the section that matters most, and every point in it comes from the ANSES opinion of 4 August 2017, which reviewed 49 adverse-effect reports collected through the French nutrivigilance scheme up to February 2017.

Phenylketonuria. Spirulina averages about 2.75 g of phenylalanine per 100 g of dry powder. ANSES advises against spirulina supplements in anyone with phenylketonuria. This is unambiguous and it is not negotiable.

Allergic predisposition. Two cases of anaphylaxis in teenagers with an allergic background, plus reported atopic dermatitis, led ANSES to advise against spirulina for anyone with an allergic predisposition.

Muscle and liver events. ANSES describes reported cases of muscle and liver damage after spirulina supplement use, without being able to characterise a dose or duration relationship: an acute rhabdomyolysis case in a 28-year-old man taking 3 g a day for one month, and a myositis case with creatine kinase at 68,000 IU/L. These are individual reports among very many users, and they are the reason nobody honest calls this an inert powder.

Anticoagulants and other medication. One report from Switzerland describes a decrease in anti-vitamin K activity in someone taking spirulina. If you take an anticoagulant, or any prescribed medicine, ask your doctor or pharmacist before adding this.

Beta-carotene ceiling. ANSES calculates that 5 g of spirulina a day supplies 7 to 8.5 mg of beta-carotene, against a maximum recommended daily intake from food supplements estimated at 7 mg. High spirulina doses can put you over that on their own.

Purines. Spirulina’s nucleic acid content is around 4.29 g per 100 g of dry matter. That is a non-trivial purine load if you have gout or have been told to limit purines.

Pregnancy, breastfeeding and children. No trial supports use in these groups in Europe, and we will not say it is safe. Speak to a doctor or pharmacist first.

What actually decides this

Three factors, and none of them is a benefit list. Dose realism: the measured effects came from 2 to 6 g a day for weeks, so decide whether you will genuinely take that much of a strongly flavoured powder before you buy any format. Contamination control: batch-level microcystin and trace-metal testing from a traceable manufacturer is the whole quality question for an organism grown in open water. And the correct expectation, which is a nutrient-dense food ingredient with four narrow measured outcomes.

If that is the deal you want, take spirulina as a food and let it be a food. If you came here for the more heavily trialled end of our range, the shilajit evidence is audited the same way, and the mushrooms are covered in lion’s mane and reishi.