Methylcobalamin vs Cyanocobalamin: Which B12 to Buy?

The most over-marketed corner of the supplement aisle

Two of the eight B vitamins are worth most people's attention: B12 and folate. Deficiency in the other six is rare in anyone eating enough food.

B12 is where the money is, and both forms of it work. Cyanocobalamin, the cheap one, is more stable and carries most of the research. Methylcobalamin, the expensive one, is sold as the natural and active version, and is manufactured by putting cyanocobalamin through one further chemical step.

MTHFR is the genetics attached to all of this. Depending on ancestry, 10 to 15% of people carry two copies of the variant, which is the version that slows the enzyme, and carrying one copy is more common still. That makes it ordinary human variation rather than a discovery. It is also a folate enzyme, with no role in converting B12 at all.

Homocysteine is the marker people chase next. B vitamins lower it reliably. Across 37,485 people in randomized trials, lowering it prevented no heart attacks, no strokes and no deaths.

Who needs B12, and who does not

Not everyone. Most guides open with a symptom list broad enough to include anybody, which is how every reader ends up suspecting they are deficient.

B12 is made by bacteria. Not by plants, and not by animals either. Grazing animals host those bacteria in their own guts, absorb what they produce, and store it in muscle, liver, milk and eggs, which is how it reaches you. Your own gut bacteria make B12 too, but B12 is absorbed high up, at the end of the small intestine ^[1], and those bacteria live further down in the colon, so what they make passes out unused. Shortage therefore tracks either what you eat or how well you absorb it, and four groups account for most of it.

• Vegans, where it is not optional. Long-term strict vegans without supplementation approach near-universal deficiency ^[2]. Vegetarians who eat dairy and eggs are in a much better position.
• Anyone past about sixty. Around 6% of adults over 60 are outright deficient, and roughly another 20% sit in a marginal band that ordinary screening waves through ^[2]. Stomach acid falls with age, and acid is what frees B12 from food.
• Long-term metformin or acid-reducing medication. Both interfere with absorption, and both are taken for years at a time ^[1].
• Anyone with pernicious anemia or gut surgery, where the absorption machinery itself is the problem.

If you are none of those and you eat meat, fish, eggs or dairy regularly, your B12 is probably fine and a supplement is unlikely to do anything for you.

Which form to buy: methylcobalamin or cyanocobalamin

Cyanocobalamin and methylcobalamin both work, and most of what is written about the difference, in both directions, is a bit dramatic.

Cyanocobalamin is the cheap default: more stable, and the form behind most of the research that established what B12 does and how much you need. The Vegan Society, which has more reason than most to get this right, recommends it. The only trial run in vegans, 42 of them, found it held holotranscobalamin, the fraction your cells can actually use, at nearly twice the level methylcobalamin managed ^[3].

Methylcobalamin is a good product, and on a different measure the evidence tilts its way: less of it is lost in urine, which implies more is held in tissue ^[4].

Those two findings sit oddly together, and the reason is that neither is strong. One measured what was circulating in blood, the other what was retained. The human studies behind the retention finding date to the 1960s and involved a handful of people. Nothing on either side measured whether anyone ended up better off.

Methylcobalamin is the better choice if you have reduced kidney function, where the small amount of thiocyanate released from cyanocobalamin can accumulate, and high-dose methylcobalamin has its own trial evidence in diabetic nerve pain.

Two things are worth knowing before you pay the premium, neither of which makes methylcobalamin a bad buy. Cyanocobalamin does not occur in nature: the cyanide group is an artifact of the charcoal purification step by which B12 was first isolated. But methylcobalamin is not harvested either. It is manufactured by chemically methylating cyanocobalamin or hydroxocobalamin in a reactor ^[5]. And whichever one you swallow, your cells strip the upper group off before use and rebuild it into the two forms they run on ^[6].

If the cyanide puts you off, that is a fair reason to buy methylcobalamin. The amount is small, about 20 mcg in a 1,000 mcg dose against a background intake of around 95 mcg a day from food and air ^[5], but you do not need a threshold argument to justify a preference.

What matters far more than the choice is taking it.

The same pattern runs through folate. Folic acid is the well-studied form that fortification programs are built on. Methylfolate is sold at a premium on the strength of the MTHFR story below.

Why do doctors prescribe cyanocobalamin instead of methylcobalamin?

Because it keeps, it costs very little, and it is the form nearly all the research used. Cyanocobalamin is unusually stable, which matters for something that sits in a pharmacy or on a shelf for a year, and the trials that established how much B12 people need were run with it.

MTHFR: common, not a diagnosis

MTHFR testing sells a lot of methylfolate. The C677T variant it looks for slows an enzyme in folate metabolism, and people carrying two copies do have measurably higher homocysteine ^[7]. That much is real. What gets left out is how ordinary it is. You inherit two copies of the gene, one from each parent, and it is the people with two copies of the variant who show the effect. Depending on ancestry, 10 to 15% of people are in that group, and carrying a single copy is more common still. It is one of the most common functional variants in human metabolism, which makes it a normal piece of human variation rather than a finding that explains your symptoms.

It is also, specifically, not a reason to buy methylcobalamin, which is where a lot of the money goes. MTHFR is a folate enzyme. It makes the methylfolate that a B12-dependent step downstream then uses. It has nothing to do with your ability to use cyanocobalamin. The two pathways meet at that step, which is where the confusion comes from, but an MTHFR variant does not impair B12 conversion.

The practical answer for almost everyone is that it changes nothing. Eat enough folate, and if you supplement, ordinary folic acid works. Where MTHFR matters clinically it is in specific contexts, such as pregnancy planning, and those are conversations with a clinician rather than a reason to buy a more expensive tablet.

So why is everything sold methylated?

For folate, the mechanism is sound. MTHFR is the enzyme that makes methylfolate, so if it runs slow, a pre-made version skips the step. That is a coherent argument, and it is where the whole idea started.

What it is not is settled. The enzyme is reduced, not absent, and ordinary folic acid still gets converted. A 13-week randomized trial that split its results by genotype found the reverse of the sales pitch: in people carrying two copies of the variant, folic acid lowered homocysteine by 30.7% against methylfolate's 7.8%.^[8] The authors note the methylfolate group started from a lower baseline, so the fair reading is that the trials go both ways, not that folic acid wins.

Then the logic traveled to nutrients the enzyme never touches. Methylated B12, activated B6, methylated multivitamins, all sold on a story about a folate enzyme. Three things made it stick. The variant is common, so millions of people who tested got a result to act on. Consumer genetic testing put that result directly in their hands. And it gives a commodity a reason to cost five times more, which is the part that keeps it alive: a two-dollar molecule with a personalized explanation attached is a far better business than a two-dollar molecule.

None of that makes methylated forms bad. It makes the reason you were given for buying them weaker than it sounded.

Oral B12 works as well as injections

Three randomized trials compared high-dose oral B12 against intramuscular injection. Oral at 1,000 to 2,000 mcg a day matched the injections at normalizing B12, and the 2,000 mcg dose produced higher levels than injecting. Both routes corrected the anemia and the neurological signs, with no meaningful difference in outcome ^[9].

Most B12 absorption needs intrinsic factor, a protein from the stomach, and that is the step that fails in pernicious anemia and after gut surgery. But roughly 1 to 2% of an oral dose crosses passively, needing no intrinsic factor at all. At ordinary doses that is a rounding error. At 1,000 mcg it is 10 to 20 mcg, which is more than a day's requirement, which is why high-dose oral works even when the normal route is broken.

Sublingual lozenges do not change this either. A 2025 review comparing sublingual, oral and injected B12 found no significant difference between any of them ^[10]. Dissolving it under your tongue is a preference, not an absorption advantage, whatever the label suggests.

Injections still have a place, mostly for severe deficiency with neurological signs where nobody wants to rely on adherence. For everything else, tablets or lozenges, whichever you will remember to take.

Homocysteine is a signal, not a target

High homocysteine is reliably associated with cardiovascular risk, which made it look like a target worth hitting. Eight randomized trials covering 37,485 people lowered homocysteine by about 25% with B vitamins. They found no reduction in major coronary events, no reduction in stroke, and no reduction in overall mortality ^[11].

So a raised homocysteine is worth understanding, because it usually points at a B-vitamin shortage you can correct, and it is not worth treating as a number to push down for its own sake. Find out why it is high, fix that if it is fixable, and put your cardiovascular effort into the markers that change outcomes when they move.

The folate trap

Folate and B12 both correct the same anemia. So a generous folate intake can normalize the blood count of someone whose actual problem is B12, while the nerve damage underneath carries on unaddressed. High-dose folic acid was used deliberately for this in the 1940s and 50s, before the neurological consequence was understood ^[12].

The practical consequence: a normal blood count does not clear you of B12 deficiency, particularly if you eat a lot of leafy greens and legumes, or take a B-complex, or live somewhere with folic acid fortification. Which is most people.

The risk is folate without B12. A small amount of folic acid alongside a B12 tablet is not the trap, since you are correcting the deficiency that would have been hidden. It is folate on its own, in someone whose B12 is short and unmeasured, that does the damage.

What to test, and what the test misses

A serum B12 is the standard test and it is weaker than it looks. It measures all the B12 in your blood, including a large fraction bound to a protein that cannot deliver it to your cells, so the number can come back normal while the usable supply is short ^[13].

If a result is borderline and the picture does not fit, MMA is the tie-breaker. It rises when B12 is missing from the reaction that needs it, which makes it a functional read rather than a headcount. It is expensive and it is not a first test.

• Vitamin B12 · the standard test and how to read it
• MMA · the confirmatory one
• Folate · usually plentiful, read alongside B12
• Homocysteine · the marker this guide argues against chasing

The bottom line

If you are vegan, past sixty, or on long-term metformin or acid-reducing medication, take B12. Either form, 1,000 mcg a day or 2,000 mcg a couple of times a week. The form is close to a coin toss. The manufacturer is not, and it is the part worth spending thought on: independent testing keeps finding supplements under their stated potency, some of them far under ^[14]. Buy from a company that tests its own product and has a name to lose.

Where you buy it is a second question. Researchers bought 30 supplements on Amazon and sent them for laboratory analysis: 17 did not match their labels, and 9 held ingredients the label never mentioned ^[15]. A listing can be filled by a seller the brand has nothing to do with, so the name on the page is not by itself a guarantee of what arrives. Check the sold by line rather than the shipping line: sold by the manufacturer is what you want, and shipped by Amazon on top of that is ordinary warehousing, not a warning sign.

If you are none of those and eat animal foods, you almost certainly do not need any of this.

Nobody needs to lower their homocysteine, and almost nobody needs methylfolate.