B12 Normal Range: What Your Lab Results Actually Mean (And Why "Normal" Often Isn't)

The Short Answer: B12 Normal Range at a Glance

Standard serum B12 reference range (pg/mL and pmol/L)

Adults: 200 to 900 pg/mL, which equals 148 to 664 pmol/L.

That’s the number most people came here for. Some labs report a top end of 1,100 pg/mL, some stop at 900, and a few run their lower limit down to 180.

Conversion is simple and worth memorizing if you ever compare results across countries:

• pg/mL × 0.738 = pmol/L
• pmol/L × 1.355 = pg/mL

So 300 pg/mL is 221 pmol/L. And 148 pmol/L is 200 pg/mL. European, Australian, and Canadian labs almost always report in pmol/L, which is why forum threads about B12 turn into a mess of mismatched numbers.

The four-tier interpretation table I actually use

Serum total B12	pmol/L	My interpretation
Under 150 pg/mL	Under 111	Deficient. Treat, then find the cause.
150 to 300 pg/mL	111 to 221	Gray zone. Functional deficiency is common here. Order MMA.
300 to 900 pg/mL	221 to 664	Adequate for most people, though symptoms plus high-risk factors still justify MMA.
Above 900 pg/mL	Above 664	Elevated. If you’re not supplementing, this needs investigation.

Now my position, up front: a result of 250 pg/mL that your lab flagged as “normal” is not reassuring. It’s a question, not an answer. I’ll explain exactly why in the gray zone section, and I’ll show you the data that changed my mind.

Why your lab’s range may differ from the one below

Two reasons.

First, reference ranges are statistical, not physiological. A lab takes a “healthy” reference population, measures everyone, and calls the middle 95% normal. Nobody checked whether the bottom 2.5% of that population had functioning methionine synthase enzymes. The range describes what’s common, not what’s optimal.

Second, assay platforms disagree. Roche, Abbott, Siemens, and Beckman all use slightly different competitive-binding chemiluminescent immunoassays, calibrated against different standards. That’s why you’ll see lower cutoffs of 180, 193, 200, or 232 pg/mL depending on which machine ran your sample. Move labs, and your “normal” can shift by 15% without your body changing at all.

What the B12 Test Is Actually Measuring (And What It Misses)

Total serum B12: haptocorrin vs transcobalamin bound

This is the part almost every article on the B12 normal range skips, and it’s the single most useful thing on this page.

Vitamin B12 doesn’t float freely in your blood. It travels bound to two carrier proteins. Roughly 70 to 80% is bound to haptocorrin (also called transcobalamin I), a protein your cells largely cannot unload. It’s essentially in transit to the liver. The remaining 10 to 30% is bound to transcobalamin II, and that fraction, called holotranscobalamin or “active B12,” is the only portion your tissues can actually take up via the CD320 receptor.

Standard serum B12 measures both. All of it. Indiscriminately.

The 20/80 problem

Think of it like this. Your total serum B12 tells you how much money is sitting in the bank building. Holotranscobalamin tells you how much is in your account. A vault full of somebody else’s cash doesn’t buy groceries.

Which means you can have a total B12 of 400 pg/mL that looks perfectly respectable while your active fraction sits at 22 pmol/L and your cells are running on fumes. Haptocorrin levels rise with liver disease, alcohol use, certain myeloproliferative conditions, and even some cancers, inflating total B12 while the usable fraction stays flat or falls.

This is the core gap. Total serum B12 can look fine while your cells are starving.

Chemiluminescent immunoassay quirks and false normals

Three specific ways the test lies to you:

Biotin interference. High-dose biotin (over 5 mg/day, and the hair-skin-nails crowd routinely takes 10,000 mcg) interferes with streptavidin-biotin based immunoassays. Depending on platform architecture, it can push B12 falsely high or falsely low. The FDA issued a safety communication about biotin interference across immunoassays in 2017 and updated it in 2019. Practical fix: stop biotin 48 to 72 hours before your draw. Nobody tells patients this. I tell every patient this.

Anti-intrinsic factor antibodies. Here’s the nastiest one. In pernicious anemia, circulating anti-intrinsic factor antibodies can bind the intrinsic factor reagent used in competitive-binding assays and produce a falsely normal or even falsely elevated B12 result. Carmel and Agrawal described exactly this failure mode in the New England Journal of Medicine in 2012. The patients most likely to be deficient are the ones whose test is most likely to look fine.

Sensitivity is mediocre. Depending on which cutoff you choose, a single serum B12 has a sensitivity somewhere in the range of 65 to 95% for true deficiency. Set the cutoff low enough to avoid false positives and you miss a large chunk of genuinely deficient people. False reassurance isn’t rare. It’s the default failure mode of this test.

So what do you do with a normal number? You look at who’s holding it. That’s where age and life stage come in.

B12 Normal Range by Age, Sex, and Life Stage

Adults (18 to 64)

200 to 900 pg/mL (148 to 664 pmol/L).

There’s no clinically meaningful sex difference in the standard adult range, though population data consistently show women running slightly higher on average, likely a mix of body size, oral contraceptive effects on binding proteins, and supplement use patterns. Labs don’t split the range by sex, and they don’t need to.

Older adults (65+): why the same range is a problem

Applying the adult range to a 78-year-old is where I part ways with standard practice.

Absorption declines with age for structural reasons. Atrophic gastritis affects a substantial fraction of adults over 60, reducing gastric acid and pepsin, which are required to cleave B12 off the food proteins it arrives bound to. Intrinsic factor output drops. Add decades of H. pylori exposure and years of acid suppression, and the machinery is worn down even when dietary intake looks adequate.

NHANES-derived analyses have repeatedly shown a meaningful proportion of older adults with serum B12 in the “normal” range who nonetheless have elevated methylmalonic acid, the metabolite that accumulates when B12-dependent enzymes stall. Their number is fine. Their biochemistry isn’t.

In my view, anyone over 65 with a B12 under 400 pg/mL and any cognitive complaint, gait change, or paresthesia deserves an MMA. Full stop.

Infants and children

Kids run higher, and it surprises people.

• Newborns: roughly 160 to 1,300 pg/mL
• Infants and children: roughly 200 to 1,200 pg/mL

Two drivers: active maternal-fetal transfer across the placenta concentrates B12 in cord blood (often 1.5 to 2× maternal levels), and children have higher cellular turnover with different binding protein distributions. A toddler at 210 pg/mL is not comfortably normal in the way an adult at 210 might appear to be.

Pregnancy and breastfeeding: the physiological dip

Serum B12 falls 20 to 30% by the third trimester in healthy, well-nourished women. This is expected. Plasma volume expands by around 45%, and transcobalamin and haptocorrin distributions shift. A third-trimester result of 180 pg/mL does not automatically mean deficiency.

Holotranscobalamin holds up much better across gestation, which is one of the few settings where I’ll argue the active B12 test earns its extra cost outright.

That said, don’t wave off a genuinely low result. Maternal B12 below roughly 200 pg/mL is associated with lower infant stores, and exclusively breastfed infants of B12-deficient mothers are one of the best-documented high-risk groups in the entire literature, presenting with irritability, feeding difficulty, hypotonia, and developmental regression that can be dramatic and, if caught late, incompletely reversible.

Vegans, vegetarians, and long-term plant-based eaters

B12 comes from bacteria, and in the human food supply it arrives essentially only via animal products or fortification. No reliable plant source exists. Spirulina, nori, and tempeh contain corrinoid analogues that can occupy binding sites without doing the job.

The numbers are stark. Gilsing and colleagues, publishing in the European Journal of Clinical Nutrition in 2010, compared British omnivores, vegetarians, and vegans, and found 52% of vegans were B12 deficient by serum criteria, with mean concentrations roughly half those of omnivores. Vegetarians landed in between at around 7%. Subsequent reviews have put deficiency prevalence in unsupplemented vegans anywhere from 40 to 80% depending on cohort and cutoff.

Supplemented vegans do fine. Unsupplemented ones frequently don’t.

Life stage	Typical reference range (pg/mL)	pmol/L	My practical floor
Newborn	160 to 1,300	118 to 959	Under 300 warrants attention
Infant/child	200 to 1,200	148 to 886	Under 300 warrants attention
Adult 18 to 64	200 to 900	148 to 664	400
Adult 65+	200 to 900	148 to 664	400, with lower threshold to order MMA
Pregnancy (T3)	Falls 20 to 30%	Variable	Use holoTC, not total B12
Vegan, unsupplemented	Same range applied	Same	400, plus annual MMA

The Gray Zone: Why 200 to 400 pg/mL Is the Most Dangerous Result You Can Get

The cutoff was built for the wrong disease

Let me make the argument plainly, because I think it’s the most important thing to understand about the B12 normal range.

The 200 pg/mL lower cutoff was calibrated to detect megaloblastic anemia, not to prevent neurological injury. Those are two different thresholds in the same person, and the hematologic one is later.

B12 does two jobs. It’s the cofactor for methionine synthase (methylation, DNA synthesis, red cell production) and for methylmalonyl-CoA mutase (branched-chain amino acid and odd-chain fatty acid metabolism). Myelin maintenance depends on the methylation arm. Nerves can start failing before the bone marrow visibly complains.

Neurological damage in the “normal” range

The classic evidence is Lindenbaum et al., New England Journal of Medicine, 1988. They studied 141 consecutive patients with neuropsychiatric abnormalities caused by cobalamin deficiency and confirmed by response to treatment. The finding that reset the field: 28% had no anemia and no macrocytosis. Their blood counts were clean. Their nerves weren’t.

A meaningful fraction had serum B12 values in the low-normal range. Paresthesias, gait ataxia, memory complaints, and mood changes preceded any hematologic clue. That paper is nearly four decades old and clinical practice still hasn’t fully absorbed it.

Japan and parts of Europe use a higher cutoff

You’ll see this claim circulating: Japan treats anything under 500 pg/mL as deficient, and that’s why Japanese dementia rates are lower. I want to handle this honestly, because it gets badly overstated online.

The first half is roughly true. Japanese clinical practice and several European centers do use a substantially higher functional cutoff, in the neighborhood of 500 pg/mL, and they act on gray-zone results rather than filing them under normal. The rationale is metabolic, not cultural.

The second half, the dementia claim, is where I’d pump the brakes. There’s no controlled evidence that a national B12 cutoff drives dementia incidence. Diet, cardiovascular risk profile, diagnostic patterns, and life expectancy differ enormously between populations. Anyone presenting that as settled causation is selling something. The higher cutoff is defensible on biochemical grounds alone. It doesn’t need a dementia story propped on top of it.

What the functional markers reveal

Here’s the physiological argument, and it’s the strong one.

When B12 becomes functionally insufficient, methylmalonyl-CoA can’t convert to succinyl-CoA, so methylmalonic acid accumulates. Simultaneously, homocysteine can’t be remethylated to methionine, so homocysteine rises. These metabolites are the direct readout of enzyme function.

And in population studies, MMA and homocysteine begin climbing in a substantial number of people below roughly 400 pg/mL, not below 200. The relationship isn’t a cliff at the reference limit. It’s a curve that starts bending well inside the “normal” range.

Framingham Offspring Study data (Tucker and colleagues, American Journal of Clinical Nutrition, 2000, with related cognitive analyses following) mapped plasma B12 against homocysteine across a large community cohort and showed exactly this: metabolic evidence of insufficiency in people whose serum B12 would never have been flagged. Subsequent Framingham work linked low-normal B12 to poorer cognitive performance measures.

My practical stance: anything under 400 pg/mL is a prompt to order MMA, not a green light. If MMA is normal and you feel fine, great, we’re done. If MMA is elevated, we found something your serum B12 was never going to show you.

So which tests should you actually be asking for?

The Tests That Beat Serum B12: MMA, Homocysteine, and Holotranscobalamin

Methylmalonic acid (MMA): normal range and interpretation

Serum MMA reference: typically under 0.40 µmol/L, with some labs using under 0.27 µmol/L. Urine MMA is also available, usually reported per gram creatinine.

MMA is the closest thing we have to a direct B12 function test. Since methylmalonyl-CoA mutase absolutely requires adenosylcobalamin, a rise in MMA means the enzyme is starved. It’s B12-specific in a way homocysteine is not (folate deficiency doesn’t touch it).

But it has confounders, and I see these missed constantly:

• Renal impairment. MMA clears renally. Reduced eGFR raises it independent of B12 status. Always interpret MMA alongside creatinine.
• Small intestinal bacterial overgrowth. Gut bacteria produce propionate, which feeds into the same pathway and can elevate MMA.
• Volume depletion. Hemoconcentration nudges it up.
• Very early infancy and pregnancy shift the interpretation slightly.

An MMA above 0.75 µmol/L with normal renal function is strong evidence of functional B12 deficiency. Values between 0.40 and 0.75 are suggestive and I’d treat and recheck.

Total homocysteine: normal range and confounders

Reference: roughly 5 to 15 µmol/L, and I’d target under 10.

Homocysteine is sensitive. It rises early in B12 deficiency. The problem is specificity, because it also rises with:

• Folate deficiency
• B6 deficiency
• Hypothyroidism
• Renal disease
• The MTHFR C677T variant (particularly TT homozygotes)
• Smoking, heavy coffee intake, and certain medications

So a high homocysteine tells you something in the one-carbon pathway is off. It doesn’t tell you it’s B12. Order it alongside folate, or don’t bother trying to interpret it in isolation. Sample handling matters too: homocysteine rises in whole blood left at room temperature, so it needs prompt separation or a chilled tube.

Holotranscobalamin (active B12): normal range

Typical range: 25 to 165 pmol/L. Under 35 pmol/L suggests deficiency. 25 to 70 pmol/L is a gray zone worth chasing with MMA.

HoloTC measures the fraction actually available to your cells, and it has a genuine advantage: it has a shorter half-life than the haptocorrin-bound pool, so it falls earlier than total B12 when intake or absorption drops. It’s the first marker to move.

Downsides: availability varies by region, it costs more, and it’s not immune to confounding (renal impairment raises it, and transcobalamin polymorphisms shift baseline values). It’s also affected by the same immunoassay quirks as total B12.

The cB12 combined indicator

Sergey Fedosov’s group developed something I find genuinely clever. The cB12 (combined indicator of B12 status) integrates all four markers, total B12, holoTC, MMA, and homocysteine, into a single logarithmic score adjusted for age and folate status.

The scale runs roughly from -3 to +3. Values below -0.5 indicate low B12 status, below -1.5 indicate probable deficiency, and above +1.5 suggest elevated status. It resolves the exact problem that makes single-marker interpretation frustrating: what to do when total B12 says one thing and MMA says another.

It’s underused. Not every lab computes it, but if yours reports all four values, the calculation is published and reproducible.

Which test to order and when

My decision tree:

• Total B12 under 200 pg/mL: treat. Then find the cause. Don’t spend two months on confirmatory testing while someone’s nerves demyelinate.
• Total B12 200 to 400 pg/mL: order MMA, plus holoTC if available and affordable. This is the gray zone and this is where the diagnosis lives.
• Total B12 above 400 pg/mL with symptoms: if the person is high-risk (older, metformin, PPI, post-bariatric, vegan, nitrous oxide exposure, autoimmune history), still run MMA. Remember the anti-IF antibody assay failure.
• Total B12 above 900 pg/mL without supplementation: different workup entirely, and it’s not benign.

Marker	Normal range	What it detects	Rough cost	Main limitation
Total serum B12	200 to 900 pg/mL	Total circulating B12, mostly inert	$20 to 60	Includes unusable haptocorrin fraction; assay interference
Holotranscobalamin	25 to 165 pmol/L	Bioavailable B12	$60 to 120	Limited availability; renal confounding
Methylmalonic acid	Under 0.40 µmol/L	Functional B12 at the cellular level	$80 to 150	Raised by renal impairment, SIBO
Homocysteine	5 to 15 µmol/L	One-carbon pathway function	$40 to 90	Non-specific; folate, B6, thyroid, MTHFR
cB12 combined	-0.5 and above	Integrated status	Cost of all four	Not routinely reported

What Causes a Low B12 Result

A low number is a finding, not a diagnosis. The question is always why.

Pernicious anemia and autoimmune gastritis

Autoimmune destruction of gastric parietal cells eliminates intrinsic factor, and without intrinsic factor, ileal B12 absorption collapses. Prevalence runs around 0.1% in the general population and 2 to 3% in adults over 60, with higher rates in people of Northern European and African descent and in anyone with another autoimmune condition (Hashimoto’s, type 1 diabetes, vitiligo).

Testing: anti-intrinsic factor antibodies have high specificity (roughly 95 to 100%) but poor sensitivity (about 50 to 70%). A positive result essentially seals the diagnosis. A negative result rules out nothing. Anti-parietal cell antibodies flip the profile, more sensitive but far less specific, showing up in plenty of people without pernicious anemia. Elevated fasting gastrin plus low pepsinogen I strengthens the case.

And circle back to the assay problem: those same IF antibodies can produce a falsely normal serum B12. Pernicious anemia can hide behind its own diagnostic test.

Medications: metformin, PPIs, H2 blockers, nitrous oxide

Metformin. The cleanest data comes from de Jager and colleagues in the BMJ (2010), a randomized placebo-controlled trial in type 2 diabetes with 4.3 years of follow-up: metformin produced a 19% reduction in serum B12 versus placebo, along with rising homocysteine. The mechanism appears to involve calcium-dependent B12-intrinsic factor uptake in the terminal ileum. The effect scales with dose and duration. If you’ve been on 2,000 mg daily for a decade, check your B12, and check MMA if it’s under 400.

Proton pump inhibitors. Stomach acid liberates B12 from dietary protein. Suppress acid, impair that step. Lam et al. published a large case-control analysis in JAMA (2013) using Kaiser Permanente data and found two or more years of PPI use associated with B12 deficiency at an odds ratio of 1.65. H2 blockers carried a smaller but real signal at 1.25. Occasional use isn’t the concern. Years of daily omeprazole is.

Nitrous oxide. This one deserves emphasis because it’s the most dangerous and the most easily missed. N2O irreversibly oxidizes the cobalt core of cobalamin from Co(I) to Co(III), inactivating methionine synthase. It doesn’t lower your serum B12. The molecule is still there, still measurable, just chemically dead. Patients present with subacute combined degeneration of the spinal cord, numbness ascending from the feet, gait instability, and a completely normal B12 result. With recreational use rising sharply, I now ask about it directly in anyone under 35 with unexplained neuropathy. MMA and homocysteine are the tests that catch it.

Gastric bypass, sleeve gastrectomy, and ileal resection

Roux-en-Y bypass removes most of the stomach from the digestive path, cutting acid and intrinsic factor production and bypassing the site where B12 is released from food. Sleeve gastrectomy reduces parietal cell mass. Ileal resection (Crohn’s surgery, for example) removes the receptor site where the B12-intrinsic factor complex is absorbed.

These patients need lifelong monitoring and supplementation, typically high-dose oral (1,000 mcg daily) or parenteral. Annual B12 with MMA is reasonable. I’ve seen post-bariatric patients five years out with a B12 of 260 who were told everything looked fine.

Dietary intake: vegan, vegetarian, and low-intake older adults

Covered above for plant-based diets, and it’s the most preventable cause on this list. Cyanocobalamin at 50 to 250 mcg daily, or 2,000 mcg weekly, solves it for pennies.

For older adults, here’s the distinction that matters and that most articles miss: food-bound cobalamin malabsorption is far more common than true dietary lack. These people eat meat, eggs, and dairy. They just can’t liberate the B12 from the protein it’s bound to, because their gastric acid output has fallen. The clinical tell is that they absorb supplemental crystalline B12 perfectly well while failing to absorb the B12 in their steak. Which means an oral supplement fixes it. Nobody needs injections for this particular problem.

Crohn’s, celiac, SIBO, and H. pylori

Terminal ileal Crohn’s disease damages the exact stretch of intestine responsible for B12 uptake. Untreated celiac disease causes broad malabsorption, and B12 deficiency shows up in a meaningful fraction of newly diagnosed patients even though the primary damage is proximal.

SIBO plays a nastier double game: bacteria in the small intestine consume B12 before you absorb it, and they produce propionate that elevates MMA. So you get real deficiency plus an exaggerated marker.

H. pylori contributes through chronic atrophic gastritis and reduced acid output, and eradication has been shown in several small trials to improve B12 status. If you’ve got low B12, dyspepsia, and no other explanation, it’s worth testing for.

Symptoms at Each Level: What Low B12 Actually Feels Like

Early and easily-dismissed signs

The first symptoms of B12 deficiency are so generic that almost nobody connects them to a vitamin. Fatigue that sleep doesn’t fix. Brain fog. A short fuse with people you like. Exercise intolerance where the same 3-mile run suddenly feels like 6.

Then there’s glossitis: a smooth, beefy-red, sore tongue where the papillae have flattened out. It’s one of the few genuinely specific signs, and I’d bet most people who’ve had it assumed they burned their mouth on coffee.

Neurological symptoms (and why they come first for some people)

Here’s the part that changes how you should think about screening. In a meaningful subset of patients, neurological damage shows up before any blood abnormality. Lindenbaum’s classic work at Columbia documented patients with clear neuropsychiatric deficits from B12 deficiency and completely normal hematocrit and MCV.

The progression is fairly predictable. Paresthesias first, usually symmetric, usually feet before hands. Then loss of vibration sense and proprioception, because the dorsal columns of the spinal cord are exquisitely sensitive to cobalamin deficiency. Then an unsteady, wide-based gait that gets worse in the dark (when you lose visual compensation for absent position sense).

The end stage has a name: subacute combined degeneration of the spinal cord. Demyelination of both dorsal and lateral columns. It’s ugly, and past a certain point it doesn’t fully reverse.

Hematologic changes: macrocytosis, MCV, and the folate masking problem

Classic B12 deficiency enlarges red cells. MCV above 100 fL is the textbook flag, along with hypersegmented neutrophils (five or more lobes) on the smear, and in advanced cases pancytopenia as the bone marrow’s DNA synthesis grinds down.

Now the problem. Folate fortification of grain products has been mandatory in the US since 1998, and plenty of people are also taking a multivitamin with 400 to 800 mcg of folic acid on top. Folate rescues the hematologic pathway. Your MCV stays a comfortable 89. Your CBC looks pristine.

And the neurological damage keeps going.

That’s the whole argument against using a CBC to screen for B12 deficiency. It was never a great screen. In a folate-fortified population, it’s close to useless.

When symptoms don’t match the number

Your liver holds 2 to 5 mg of B12, roughly a thousand times your daily requirement. That’s 3 to 5 years of reserve. Which is why deficiency arrives like a slow leak, not a burst pipe, and why every symptom along the way gets filed under “getting older” or “stress at work.”

The clock that matters: neurological damage becomes progressively less reversible after roughly 6 to 12 months of untreated deficiency. That’s my entire case for testing aggressively and early rather than waiting for the MCV to budge. A $60 methylmalonic acid test is cheap next to permanent sensory loss.

High B12: When Your Result Is Above the Normal Range

Supplement-driven elevation (the boring, common cause)

If you take 1,000 mcg of B12 daily and your serum level comes back at 1,800 pg/mL, that’s not a finding. That’s arithmetic. There’s no established tolerable upper intake level for B12 precisely because the excess gets cleared renally without incident.

I see people panic about this constantly. Don’t. Stop the supplement for a few weeks if you want a clean baseline, but a high number in a supplementing person tells you nothing except that you swallowed the pill.

Liver disease, renal failure, and myeloproliferative disorders

Unexplained elevation in someone not supplementing is different, and it’s underinvestigated in primary care.

The mechanism is haptocorrin. Roughly 80% of circulating B12 is bound to haptocorrin, which is produced by granulocytes and cleared by hepatocytes. Damage the liver and clearance falls. Ramp up granulocyte production and release goes up. Either way, total serum B12 inflates without a single extra molecule reaching your cells.

The usual suspects: alcoholic and non-alcoholic liver disease, hepatocellular carcinoma, chronic myeloid leukemia, polycythemia vera, and chronic kidney disease.

The malignancy association nobody warns you about

Arendt and colleagues published a Danish population cohort in the Journal of the National Cancer Institute in 2013 that I think deserves far more attention than it gets. They tracked more than 300,000 people referred for B12 testing who weren’t on supplements. Cancer risk climbed with cobalamin level, and the association was strongest in the first year after testing, with the highest category showing several-fold elevated risk compared to the reference group.

Is high B12 causing cancer? Almost certainly not. It’s a marker, probably reflecting haptocorrin dynamics in liver and hematologic malignancies. But as a marker in an unexplained case, it earns a workup: liver panel, CBC with differential, renal function, and a look at whatever else the history suggests.

Is high B12 toxic?

Not in any meaningful sense from oral or injected cobalamin in people with functioning kidneys. Decades of 1,000 mcg daily dosing haven’t produced a toxicity syndrome.

Two honest caveats. First, high-dose B12 (usually B12 plus B6) has been linked to acneiform eruptions in small case series, with a proposed mechanism involving Cutibacterium acnes metabolism. Small case series only. If your skin flared after starting a B-complex, it’s plausible, and stopping is a fair experiment.

Second, and more important clinically: high total B12 and elevated MMA can coexist. Functional deficiency with a reassuring number on paper. Happens in renal impairment and in haptocorrin-heavy states. The number is not the truth.

How to Get Back Into the B12 Normal Range: Dosing, Forms, and Timelines

Oral vs intramuscular: what the trials actually show

This is where I’ll be blunt, because the practice hasn’t caught up with the evidence.

The Cochrane review by Wang et al. (2018) pooled randomized trials comparing oral and intramuscular B12 for correcting deficiency. Oral doses of 1,000 to 2,000 mcg daily performed comparably to injections on serum B12 and hematologic response. Including in pernicious anemia.

How? Passive diffusion. About 1% of an oral dose crosses the intestinal mucosa without intrinsic factor, without receptors, without anything. One percent of 1,000 mcg is 10 mcg, roughly four times the daily requirement. The mechanism you broke simply gets bypassed by brute force.

That said, I still favor IM for three situations: active neurological involvement (I want repletion fast, not eventually), significant ileal resection, and patients whose adherence to a daily pill is genuinely unreliable. Otherwise, oral is fine and it’s cheaper and it doesn’t require a clinic visit.

Cyanocobalamin vs methylcobalamin vs hydroxocobalamin vs adenosylcobalamin

Cyanocobalamin is the most studied, most stable, and cheapest form. The cyanide moiety is a rounding error (micrograms), irrelevant outside of severe renal failure or Leber’s hereditary optic neuropathy.

Hydroxocobalamin has the longest tissue retention and is the standard injectable in the UK, which is why British dosing schedules stretch to every 2 to 3 months.

Methylcobalamin is the one sold at a premium with the boldest claims. I’ll say it plainly: the superiority claims outrun the data. There’s no convincing head-to-head evidence that it corrects deficiency better in humans, and it’s actually less stable (light-sensitive) than cyanocobalamin. The MTHFR argument people make for it doesn’t hold up, since cobalamin methylation happens intracellularly regardless of which form you swallow.

Adenosylcobalamin is the mitochondrial cofactor form. Fine, no advantage demonstrated, usually sold in combination.

Use cyanocobalamin unless you have a specific reason not to.

Dosing by scenario

• Maintenance for vegans and vegetarians: 25 to 100 mcg daily, or 1,000 mcg two to three times weekly, or 2,000 mcg once weekly. All work.
• Documented deficiency, no neurological signs: 1,000 to 2,000 mcg oral daily for 1 to 3 months, then reassess with MMA.
• Pernicious anemia: 1,000 mcg IM weekly for 4 weeks, then monthly for life. Or 1,000 to 2,000 mcg oral daily for life with monitoring. Both are defensible.
• Neurological involvement: IM 1,000 mcg every other day for 1 to 2 weeks, then weekly, then monthly. Don’t be timid here.
• Post-bariatric: 1,000 mcg oral daily indefinitely, or 1,000 mcg IM monthly. Annual MMA.

The RDA (2.4 mcg/day for adults, 2.6 mcg in pregnancy, 2.8 mcg in lactation) looks laughably small next to those numbers. There’s a reason. Intrinsic factor-mediated absorption saturates at roughly 1.5 to 2 mcg per dose. Therapeutic dosing isn’t targeting that pathway at all.

Sublingual and nasal: any real advantage?

Sublingual tablets test equivalently to swallowed tablets. The “bypasses the gut” pitch is marketing; most of what you take sublingually is swallowed in saliva anyway. If you prefer the taste or the ritual, fine, but don’t pay extra.

Intranasal cyanocobalamin (Nascobal) is real and works. It’s also expensive and mostly reserved for maintenance in people who hate needles.

How long until levels and symptoms improve

• Reticulocyte response: 3 to 5 days. This is the earliest objective sign it’s working.
• Hemoglobin normalization: 6 to 8 weeks.
• Energy and cognitive symptoms: often 1 to 4 weeks, sometimes dramatically.
• Neurological recovery: 3 to 12 months, and sometimes incomplete. Duration of deficiency before treatment is the main predictor.

Two safety points worth knowing. During rapid hematologic recovery in severe megaloblastic anemia, a burst of new cell production can drive potassium into cells and produce clinically significant hypokalemia. Severe cases need potassium monitored in the first week.

And always check folate and iron at the same time. Correcting B12 in an iron-deficient person unmasks the iron problem fast, because you’ve just given the marrow permission to build cells it doesn’t have raw material for. The MCV drops, the ferritin was low all along, and now you’re chasing a second deficiency.

Retesting and Monitoring: When and What to Recheck

Why you shouldn’t retest serum B12 while supplementing

Serum B12 rises within days of your first dose regardless of whether tissue stores have refilled. Recheck it at week four on 1,000 mcg daily and you’ll get a beautiful number that means precisely nothing about your functional status.

I’ve watched clinicians treat, recheck serum B12, see 900 pg/mL, and declare victory while the patient’s paresthesias haven’t moved. The marker responded. The patient didn’t.

The right retest interval

Recheck MMA or homocysteine at 4 to 8 weeks. These reflect what’s happening inside cells, and they normalize only when the enzymatic bottleneck actually clears.

If MMA hasn’t normalized on adequate B12, stop and rethink. Three things to check: renal function (impaired clearance keeps MMA elevated independent of B12 status), SIBO (bacterial propionate production inflates MMA), and the possibility that B12 was never the cause of the symptoms in the first place.

What a good result looks like after treatment

MMA under 0.30 µmol/L. Homocysteine under 10 µmol/L (under 12 if you’re being generous). Serum B12 comfortably above 400 pg/mL, though as noted, that one’s decoration at this point.

Screening cadence by risk group: annual B12 plus MMA for vegans, metformin users on long-term therapy, post-bariatric patients, anyone with pernicious anemia or ileal disease, and adults over 65. Every 2 to 3 years for vegetarians and PPI users.

Cost reality, because this matters: total serum B12 typically runs $15 to $40 out of pocket. MMA runs $50 to $90. HoloTC is frequently not covered in the US and can be hard to even order. That price gap is the single biggest reason MMA gets skipped, and it’s a bad trade when neurological damage is on the table.

Food Sources and Realistic Intake

Highest-density animal sources

Food	B12 per 3 oz (85 g)
Clams, cooked	~84 mcg
Beef liver	~70 mcg
Rainbow trout	~5.4 mcg
Salmon, sockeye	~4.8 mcg
Tuna, light, canned	~2.5 mcg
Beef, ground	~1.5 mcg
Milk (1 cup)	~1.2 mcg
Egg (1 large)	~0.5 mcg
Cheddar (1 oz)	~0.5 mcg

One serving of clams covers a month of RDA on paper. It doesn’t work that way in practice, which brings me to the ceiling.

The absorption ceiling nobody accounts for

Intrinsic factor-mediated absorption saturates at roughly 1.5 to 2 mcg per meal. Eat 70 mcg of liver in one sitting and you’ll absorb maybe 2 mcg through the normal pathway plus about 1% of the rest by diffusion.

Which means frequency beats magnitude for dietary B12. Three modest servings across the day outperform one enormous one. It also means the “I eat liver once a month” strategy is weaker than it sounds.

For older adults with reduced gastric acid, add another layer: food-bound B12 requires acid and pepsin to be liberated from protein, while crystalline B12 in a supplement or fortified food doesn’t. Same person, two very different absorption efficiencies depending on the source.

Fortified foods and what works for plant-based eaters

Fortified nutritional yeast, fortified plant milks, and fortified breakfast cereals are the only reliable non-supplement plant sources. Check the label, because fortification is not universal and “nutritional yeast” on its own is not automatically fortified.

The spirulina, chlorella, and seaweed myth

I’ll take a clear position here because the ambiguity has caused real harm.

Spirulina, chlorella, and most seaweeds (nori being a partial and debated exception) contain pseudo-cobalamin, corrinoid analogues that are structurally similar to B12 but biologically inactive in humans. Worse, they register on standard serum B12 assays, so a spirulina user can show a “normal” B12 level while being functionally deficient. There’s also evidence these analogues compete with true cobalamin at binding sites, potentially making things worse rather than neutral.

If you’re plant-based, take cyanocobalamin. It costs about $5 for a year’s supply. That’s the whole solution.

My Bottom Line on the B12 Normal Range

“Normal” on your lab report is a statistical statement, not a clinical one. It means your value sits within two standard deviations of whatever population that lab sampled, which in the US includes a lot of people with subclinical deficiency. It does not mean your cells have enough cobalamin to run methionine synthase and methylmalonyl-CoA mutase properly.

Here’s my personal target, and I’ll own it as an opinion informed by the functional marker data rather than a consensus guideline: serum B12 between 400 and 700 pg/mL (295 to 515 pmol/L) with MMA under 0.30 µmol/L. Not 210 pg/mL with a green checkmark next to it. Not 250 with “your bloodwork looks great.” Those numbers sit squarely in the zone where a third or more of people have elevated MMA.

If you have symptoms and a borderline number, ask for MMA. If your clinician resists, the argument to bring is the haptocorrin one: 80% of your serum B12 is bound to a carrier protein that can’t deliver it to tissues, so total serum B12 measures mostly the useless fraction. HoloTC measures the deliverable fraction. MMA measures whether the delivery worked. That framing tends to land.

The decision tree, condensed: Below 200 pg/mL, treat. Between 200 and 400 with any symptoms, get MMA (and homocysteine if it’s cheap) before deciding anything. Above 400 with symptoms, still consider MMA if you have renal impairment, liver disease, or take spirulina, then look elsewhere for the cause. Above the reference range without supplementing, get a liver panel and CBC. And whatever you do, don’t let a normal MCV talk you out of testing.

Frequently Asked Questions

Q: What is a normal B12 level by age? Most US labs use the same reference range for all adults: roughly 200 to 900 pg/mL (148 to 664 pmol/L). Age-specific ranges aren’t standard. That’s a problem, because adults over 60 have higher rates of atrophic gastritis and food-bound malabsorption, so a level of 250 pg/mL carries more risk at 75 than at 25. Infants and children generally run higher, often 300 to 1,200 pg/mL.

Q: Is a B12 level of 300 pg/mL good, or is it too low? It’s technically normal and functionally uncertain. Levels between 200 and 400 pg/mL are the grey zone, where roughly 25 to 40% of people show elevated methylmalonic acid, meaning real cellular deficiency. If you have fatigue, brain fog, or any tingling in your hands or feet at 300 pg/mL, get an MMA test before accepting “normal.”

Q: What does the B12 test actually measure, and how does it work? A standard serum B12 test measures total cobalamin bound to both carrier proteins: haptocorrin (about 80%, biologically unavailable to most tissues) and transcobalamin (about 20%, the fraction actually delivered to cells). It uses a competitive binding immunoassay with intrinsic factor. Because it lumps the useful and useless fractions together, it can read normal in a genuinely deficient person.

Q: Is taking high-dose B12 safe, and can you take too much? Yes, it’s safe for people with normal kidney function. B12 has no established tolerable upper intake level because excess is water-soluble and excreted renally. Doses of 1,000 to 2,000 mcg daily have been used for decades without a toxicity syndrome. The rare reported issue is acneiform eruption with high-dose B12 and B6, documented only in small case series.

Q: What is the best B12 dosage to correct a low level? For most people, 1,000 to 2,000 mcg of oral cyanocobalamin daily for 1 to 3 months. About 1% is absorbed by passive diffusion regardless of intrinsic factor, which is why this works even in pernicious anemia. For maintenance afterward, 25 to 100 mcg daily or 1,000 mcg two to three times weekly. Neurological involvement or ileal resection warrants intramuscular dosing: 1,000 mcg weekly for 4 weeks, then monthly.

Q: How long does B12 supplementation take to work? Reticulocytes (new red blood cells) surge within 3 to 5 days. Hemoglobin normalizes over 6 to 8 weeks. Energy and cognitive symptoms often improve within 1 to 4 weeks. Neurological symptoms take 3 to 12 months and may not fully resolve if deficiency went untreated for more than 6 to 12 months.

Q: What does a high B12 level mean if I’m not supplementing? It warrants investigation. Common causes include alcoholic and non-alcoholic liver disease, chronic kidney disease, and myeloproliferative disorders like chronic myeloid leukemia and polycythemia vera. The mechanism is haptocorrin released from damaged liver cells or overproduced granulocytes. Arendt et al. (2013, JNCI) found elevated cobalamin associated with increased cancer diagnosis risk in the following year in a Danish cohort of over 300,000 people. Ask for a liver panel, CBC with differential, and renal function.

Q: Should I get B12 shots or take oral B12? For most people, oral works just as well. The 2018 Cochrane review (Wang et al.) found 1,000 to 2,000 mcg daily oral B12 comparable to intramuscular injections for correcting deficiency, including pernicious anemia. Injections are still preferable with active neurological symptoms, significant ileal resection, or poor pill adherence.

Q: Do I need to fast before a B12 blood test? No. Serum B12 doesn’t require fasting. If your doctor ordered it alongside a lipid panel or glucose, fast for those. Do avoid taking a B12 supplement in the 24 hours before the draw, since a recent dose can transiently inflate the result.

Q: Why is my B12 normal but I still have symptoms? Three possibilities. First, your “normal” is in the 200 to 400 pg/mL grey zone where functional deficiency is common; test MMA. Second, you have a haptocorrin-elevating condition inflating total B12 while the usable fraction stays low. Third, spirulina or chlorella supplements are registering as B12 on the assay when they contain inactive pseudo-cobalamin. And fourth (yes, I said three): B12 may not be the cause. Check ferritin, TSH, and vitamin D.

Q: How do I convert pg/mL to pmol/L for B12? Multiply pg/mL by 0.738 to get pmol/L. To go the other direction, multiply pmol/L by 1.355. So 200 pg/mL = 148 pmol/L, 400 pg/mL = 295 pmol/L, and 900 pg/mL = 664 pmol/L. US labs typically report pg/mL; European, Australian, and Canadian labs usually report pmol/L.

Q: Can biotin supplements affect my B12 test results? Yes, and this is underappreciated. High-dose biotin (the 5,000 to 10,000 mcg doses sold for hair and nails) interferes with immunoassays that use biotin-streptavidin binding, which includes many B12 assays. The FDA issued a safety communication about this in 2017 and reiterated it in 2019. Interference can push results falsely high or falsely low depending on the assay design. Stop biotin for at least 72 hours before any blood test, and tell the lab you were taking it.

---

If you take one thing from all of this, make it this: a checkmark next to your B12 result is a comparison to a population, not a verdict on your physiology. Ask what the actual number is. If it’s under 400 and you feel bad, ask for methylmalonic acid. The test costs less than a month of most supplements, and the window for reversing neurological damage doesn’t stay open forever.

Frequently Asked Questions