{"slug":"vitamin-b12","name":"Vitamin B12","aliases":["Serum cobalamin","Methylmalonic acid (MMA)"],"kind":"biomarker","system":"epigenetics","unit":"pg/mL","unitNote":"Some reports print pmol/L: multiply pg/mL by about 0.738. NICE prints ng/L, which is the same as pg/mL.","answer":"A serum B12 below about 200 pg/mL is deficient on most definitions, and about 200 to 350 is an indeterminate zone in which roughly half of older adults showed metabolic deficiency on methylmalonic acid, so a low-normal value with symptoms is a reason to check MMA rather than a reassurance; above about 350 deficiency is unlikely, no trial has shown a benefit from raising a low-normal level in people without symptoms, and very high values without supplements have tracked with liver, kidney and blood disorders rather than with better health.","summary":"Vitamin B12 is the cofactor the body needs for methylation, DNA synthesis and part of mitochondrial fuel handling, and the serum level is a rough proxy for how much reaches the cells. Laboratories call about 200 to 900 pg/mL normal, but deficiency with nerve damage has been documented inside that range, and a methylmalonic acid is what settles a value between about 200 and 350. Above that the evidence thins: trials that raised low-normal levels did not improve nerves or memory, and very high values without supplements have tracked with illness.","whatItIs":[{"text":"On a metabolomic panel B12 is read as a node in a network rather than as a level. Human cells use it in only two enzymes: methionine synthase, which hands a methyl group to homocysteine to make methionine and keeps the folate cycle turning, and methylmalonyl-CoA mutase, which feeds odd-chain fats and some amino acids into the mitochondrial citric-acid cycle. When B12 runs short, the substrate of each enzyme backs up: homocysteine on the methylation side and methylmalonic acid on the mitochondrial side. That is why methylmalonic acid, not the serum B12, is the functional readout, and why a raised homocysteine with a normal methylmalonic acid points at folate instead.","cites":["banerjee2003","gherasim2013","allen2018"],"grade":"A"},{"text":"Vitamin B12, also called cobalamin, is a vitamin with a central role in cellular metabolism, especially in DNA synthesis, methylation and mitochondrial metabolism. When the cells run short, two intermediates back up and can be measured: homocysteine, from the methylation side, and methylmalonic acid, from the mitochondrial side. Clinical deficiency with the classic blood and nerve findings is relatively uncommon, but subclinical deficiency affects between 2.5 and 26 percent of the general population depending on the definition used, and the cut-offs that separate the two remain debated.","cites":["green2017"],"grade":"A"},{"text":"The serum number is an imperfect measure of that supply. Only about a quarter of circulating B12 is bound to transcobalamin, the carrier that delivers it to cells; the rest rides on haptocorrin and is not available to tissues. The British haematology guideline is explicit that there is no gold-standard test for deficiency: serum cobalamin stays the first-line test, methylmalonic acid is the second-line test for uncertain results, and homocysteine is helpful but less specific. The international BOND expert review calls methylmalonic acid the most sensitive and specific marker of status and says serum B12 lacks sensitivity and specificity for diagnosing deficiency.","cites":["nexo2011","devalia2014","allen2018"],"grade":"A"},{"text":"Dietary B12 comes only from animal foods and from foods fortified with it. It is released from food protein in the stomach, bound to intrinsic factor made by the stomach's parietal cells, and absorbed in the last part of the small intestine (the terminal ileum). That pathway saturates: about 50 percent of a 1 microgram dose is retained, about 5 percent of a 25 microgram dose, and about 1 percent of a 500 microgram dose, the last by passive diffusion that needs no intrinsic factor. In older adults, the commonest cause of deficiency is not diet but the failure to free B12 from food (more than 60 percent of cases in one review), with pernicious anemia, the autoimmune loss of intrinsic factor, accounting for 15 to 20 percent.","cites":["allen2018","andres2004"],"grade":"A"}],"origins":[{"text":"No animal, plant or fungus can make B12. Only bacteria and archaea carry the roughly thirty-step pathway, and the best reading of its distribution is that it evolved early, to let anaerobic microbes ferment small molecules before oxygen filled the atmosphere; the pathways for siroheme, chlorophyll and heme branched off it later. As oxygen rose, many organisms dropped the fermentative uses and kept the newer B12-dependent ones.","cites":["roth1996"],"grade":"C"},{"text":"Plants have no B12-dependent enzymes at all, which is why strict plant diets run short. Across 326 algal species, more than half need an outside supply, and that dependence arose many separate times in evolution, each time by losing the B12-independent form of methionine synthase; the algae get the vitamin from bacteria they live with. Animals are in the same position: we kept the dependent enzymes and never had the pathway.","cites":["croft2005"],"grade":"B"},{"text":"Because the cofactor is both reactive and scarce, animals evolved an elaborate relay to capture it from food and carry it to the two enzymes that use it: intrinsic factor in the stomach, a receptor in the last part of the small intestine, transcobalamin in the blood and a chain of handling proteins inside the cell, several of them identified only in the last two decades. Some gut bacteria make B12, but the evidence suggests they keep it as a prized resource for the microbial community rather than provisioning the host in meaningful amounts; what reaches our cells comes from animal foods, fortified foods or supplements.","cites":["gherasim2013","degnan2014"],"grade":"C"}],"labRange":{"typical":"About 200 to 900 pg/mL on the classic reference range; the lower limit is printed anywhere from about 160 to 245 depending on the assay, and many reports flag about 200 to 350 as a grey zone","claims":[{"text":"The classic reference range, 200 to 900 pg/mL, is the one the landmark oral-versus-injection trial used in 1998. Across assays, cut-offs for deficiency typically range from 120 to 180 pmol/L (about 160 to 245 pg/mL), with 148 pmol/L (about 200 pg/mL) the most common; adults at 150 to 250 pmol/L (about 200 to 340 pg/mL) are assumed to be depleted, some of them deficient. In the US national survey, 2.8 percent of the population fell below 148 pmol/L, 10.5 percent below 200 pmol/L (about 270 pg/mL) and 25.6 percent below 258 pmol/L (about 350 pg/mL), so where the line is drawn changes the prevalence tenfold.","cites":["kuzminski1998","allen2018"],"grade":"A"},{"text":"The 2024 NICE guideline sets the thresholds most clearly: a total B12 below 180 ng/L (pg/mL) confirms deficiency, 180 to 350 is an indeterminate result that means possible deficiency, and above 350 makes deficiency unlikely; for the 'active B12' test (holotranscobalamin) the equivalents are below 25, 25 to 70 and above 70 pmol/L. It asks for a serum methylmalonic acid when a person with symptoms lands in the indeterminate zone, reads MMA against the laboratory's own range, and notes that people of Black ethnicity may have a higher reference range than people of White or Asian ethnicity.","cites":["nice2024"],"grade":"A"},{"text":"The low end is contested because deficiency has been documented inside it. Among 141 consecutive patients with nerve or psychiatric disease caused by B12 deficiency, 2 had a serum B12 above 200 pg/mL and 16 had 100 to 200, and in a later series the serum level was normal in 9 (5.2 percent) of 173 patients with recognized deficiency; in early relapse of pernicious anemia, methylmalonic acid was abnormal more than twice as often as the serum B12. A printed 'normal' therefore tells a reader less than the number, the symptoms and the MMA together.","cites":["lindenbaum1988","lindenbaum1990"],"grade":"B"}]},"evidence":{"headline":"Metabolic deficiency was common below about 350 pg/mL and was settled by methylmalonic acid, not by the serum value; above that, one small cohort saw the least brain-volume loss above about 420, a large Dutch cohort saw the lowest mortality below about 455, trials that raised low-normal levels changed nothing measurable, and values far above the range without supplements marked disease rather than health.","scale":{"min":0,"max":1200,"ticks":[0,200,400,600,800,1000,1200]},"bands":[{"label":"Common laboratory reference range","low":200,"high":900,"tone":"lab","source":"Typical US laboratories; Kuzminski 1998"},{"label":"Deficient on most definitions","low":0,"high":200,"tone":"caution","source":"NICE 2024 (below 180); Allen 2018 (148 pmol/L)"},{"label":"Indeterminate zone: raised MMA or homocysteine in about half of older adults here","low":200,"high":350,"tone":"neutral","source":"NICE 2024; Pennypacker 1992; Lindenbaum 1994"},{"label":"Deficiency unlikely; lowest-mortality quartiles in the Dutch cohort","low":350,"high":455,"tone":"evidence","source":"NICE 2024; Flores-Guerrero 2020"},{"label":"Least brain-volume loss above about 420 in one small cohort; top mortality quartile in the Dutch cohort, probably reverse causation","low":455,"high":815,"tone":"neutral","source":"Vogiatzoglou 2008; Flores-Guerrero 2020"},{"label":"Above the Danish upper limit without supplements: cancer diagnosed 3.4 to 6.3 times as often in the next year","low":815,"high":1200,"tone":"caution","source":"Arendt 2013"}],"claims":[{"text":"Deficiency does not wait for anemia. Among 141 consecutive patients whose nerve or psychiatric disease was caused by B12 deficiency, 40 (28 percent) had neither anemia nor large red cells; their features were tingling, sensory loss, unsteadiness, dementia and psychiatric disorders with markedly raised methylmalonic acid and homocysteine, and 39 of 39 treated improved. In a separate series of 434 episodes of confirmed deficiency, the hematocrit was normal in 28 percent and the mean cell volume in 17 percent.","cites":["lindenbaum1988","savage1994"],"grade":"B"},{"text":"The grey zone is where most metabolic deficiency lives. In 152 outpatients aged 65 to 99, 14.5 percent had a serum B12 of 300 pg/mL or below with methylmalonic acid or homocysteine more than 3 standard deviations above normal, and the share with raised metabolites was the same whether the B12 was 201 to 300 (56 percent) or 200 and below (62 percent); treatment corrected the metabolites in every patient treated. In 548 survivors of the original Framingham cohort, 40.5 percent had a B12 below 258 pmol/L (about 350 pg/mL) against 17.9 percent of younger controls, methylmalonic acid was markedly raised in 11.3 percent, and the authors put the prevalence of true deficiency at 12 percent or more, noting that many older people with 'normal' serum values are metabolically deficient.","cites":["pennypacker1992","lindenbaum1994"],"grade":"B"},{"text":"Methylmalonic acid is the tie-breaker. In 434 episodes of clear-cut deficiency, MMA was raised in 98.4 percent and homocysteine in 95.9 percent, only one patient had both normal, and the authors concluded that normal levels of both rule out clinically significant deficiency with virtual certainty; homocysteine was also raised in 91 percent of folate-deficiency episodes, which is why MMA is the more specific of the two. NICE asks for MMA when a person with symptoms has an indeterminate result, and clinical studies comparing holotranscobalamin with total B12 against an MMA standard suggest the active fraction performs better.","cites":["savage1994","nice2024","nexo2011"],"grade":"A"},{"text":"For the brain, the cohorts point one way. In 107 community-dwelling volunteers aged 61 to 87 without cognitive impairment, followed with yearly MRI for 5 years, those in the bottom third of B12 (below 308 pmol/L, about 420 pg/mL) had 6.2 times the odds of faster brain-volume loss after adjustment for age, homocysteine, folate and other factors; the same was true for low holotranscobalamin, while high MMA or homocysteine was not associated. In 1,459 older Americans in the national survey, low B12 status (below 148 pmol/L or MMA above 210 nmol/L) carried 2.7 times the odds of anemia and 2.5 times the odds of cognitive impairment. A small cohort and a cross-sectional survey, not proof of cause.","cites":["vogiatzoglou2008","morris2007"],"grade":"C"},{"text":"The trials point the other way. In 201 people aged 75 and over with a B12 of 107 to 210 pmol/L (about 145 to 285 pg/mL) and no anemia, 1 mg of oral B12 a day for 12 months raised serum B12 from 231 to 641 pmol/L and holotranscobalamin more than fourfold, and changed nothing: no effect on nerve conduction, the clinical neurological examination or cognitive function. In 195 people aged 70 and over with mild deficiency, 1,000 micrograms a day for 24 weeks corrected the biochemistry and did not improve cognition. Across 11 trials of B vitamins including B12 in 22,000 people, there was no effect on any cognitive domain. The one positive trial, VITACOG, slowed brain atrophy (0.76 versus 1.08 percent a year) in 168 people with mild cognitive impairment using folic acid, B6 and 0.5 mg of B12 together, so it cannot be credited to B12 alone.","cites":["dangour2015","eussen2006","clarke2014","smith2010"],"grade":"A"},{"text":"At the high end the association reverses. In 5,571 Dutch adults (mean age 53.5) not taking B12 supplements, followed for a median of 8.2 years, the median B12 was 394 pg/mL; mortality was 33.8 deaths per 10,000 person-years in the lowest quartile and 65.7 in the highest (above 455.41 pg/mL), and after adjustment for age, kidney function, liver enzymes, cancer history and the usual risk factors the top quartile carried 1.85 times the mortality, 1.25 per standard deviation. Higher B12 went with higher liver enzymes at baseline, and the authors' candidate explanation is release of B12 from liver stores or reduced clearance in liver or kidney disease. In 333,667 Danes referred for a B12 test, 6 percent were above the upper reference limit (601 pmol/L, about 815 pg/mL), and in the following year cancer was diagnosed 3.4 times as often as expected at 601 to 800 pmol/L and 6.3 times as often above 800, concentrated in blood cancers and smoking- and alcohol-related cancers. The number was a marker of illness already present, not a cause.","cites":["floresguerrero2020","arendt2013"],"grade":"B"}]},"levers":[{"name":"Animal foods and fortified foods","effect":"raise it; the only dietary sources","claim":{"text":"Dietary B12 is limited to animal-source foods, so strict vegetarianism (veganism) leads to deficiency unless supplements or fortified foods are eaten. Absorption from a meal saturates at about 50 percent of a 1 microgram dose, which is why intake spread across meals beats one large serving. Across 18 studies that defined deficiency by methylmalonic acid or holotranscobalamin, vegetarians were deficient at rates of 62 percent in pregnancy, 21 to 41 percent in adolescence and 11 to 90 percent in old age, with higher rates in vegans and in people vegetarian since birth.","cites":["allen2018","pawlak2013"],"grade":"A"}},{"name":"Oral supplements, and the route","effect":"raise it; high oral doses work even without intrinsic factor","claim":{"text":"Because about 1 percent of a large oral dose crosses the gut wall passively, oral B12 at a high enough dose bypasses the intrinsic-factor system. In 38 newly diagnosed deficient patients randomized to 2 mg by mouth daily or 1 mg by injection on a tapering schedule, blood counts and neurological findings recovered equally, and at 4 months the oral group's serum B12 had risen from 93 to 1,005 pg/mL against 95 to 325 in the injection group. The 2018 Cochrane review (3 trials, 153 people) found no clinically relevant difference in serum B12 between 1,000 micrograms a day by mouth and injections, at low certainty, with oral treatment costing less. NICE prescribes at least 1 mg a day by mouth when malabsorption is the cause, and lifelong injections when autoimmune gastritis, total gastrectomy or complete terminal ileal resection is the cause. Over-the-counter products vary; NICE advises a product containing cyanocobalamin, methylcobalamin or adenosylcobalamin.","cites":["allen2018","kuzminski1998","wang2018","nice2024"],"grade":"A"}},{"name":"Metformin","effect":"lowers it by about a fifth over years","claim":{"text":"In a randomized trial of 390 people with type 2 diabetes, 4.3 years of metformin lowered serum B12 by 19 percent against placebo and raised the absolute risk of deficiency (below 150 pmol/L) by 7.2 percentage points, one extra case for every 14 people treated, with a further 11.2 points in the low range (150 to 220 pmol/L, about 200 to 300 pg/mL). In the Diabetes Prevention Program's long follow-up, low or borderline-low B12 (298 pg/mL or below) affected 19.1 percent of metformin users against 9.5 percent on placebo at 5 years and 20.3 versus 15.6 percent at 13 years, with each year of use raising the odds of deficiency by 13 percent. NICE lists metformin among the medicines that count as a risk factor and suggests replacement, by mouth or injection, while the medicine continues. Whether a medicine should change is the prescriber's call, never a reason to stop on one's own.","cites":["dejager2010","aroda2016","nice2024"],"grade":"A"}},{"name":"Acid-suppressing medicines","effect":"lower absorption from food over years of use","claim":{"text":"Stomach acid frees B12 from food protein. In 25,956 people newly diagnosed with B12 deficiency and 184,199 matched controls in a US health system, two or more years of a proton pump inhibitor carried 1.65 times the odds of deficiency and two or more years of a histamine-2 blocker 1.25 times, with a stronger association at higher daily doses (1.95 above 1.5 pills a day). A case-control association, consistent with the physiology, and both drug classes appear on NICE's list of risk factors.","cites":["lam2013","nice2024","allen2018"],"grade":"B"}},{"name":"Age, atrophic gastritis and pernicious anemia","effect":"lower it by cutting absorption from food, then from everything","claim":{"text":"Deficiency occurs in more than 20 percent of older people, most often because atrophic gastritis and reduced acid stop B12 being freed from food (food-cobalamin malabsorption, more than 60 percent of cases) and less often because autoimmune destruction of the parietal cells removes intrinsic factor altogether (pernicious anemia, 15 to 20 percent). In Framingham, 40.5 percent of the surviving original cohort had a B12 below 258 pmol/L against 17.9 percent of younger controls. The intrinsic factor antibody test confirms pernicious anemia with more than 95 percent specificity but only 50 to 70 percent sensitivity, so a negative test does not rule it out; NICE then suggests parietal cell antibodies, gastrin, an absorption test or gastroscopy, and lifelong injections once autoimmune gastritis is the cause.","cites":["andres2004","lindenbaum1994","allen2018","nice2024"],"grade":"A"}},{"name":"Stomach and bowel surgery, nitrous oxide, other medicines","effect":"lower it; some causes are permanent","claim":{"text":"The odds of B12 deficiency after Roux-en-Y gastric bypass are more than 3 times higher, and NICE treats deficiency as highly likely after total gastrectomy or complete terminal ileal resection without replacement. Its risk-factor list also names coeliac disease and other autoimmune conditions (thyroid disease, type 1 diabetes), a family history of deficiency, previous abdominal or pelvic radiotherapy, recreational nitrous oxide, and the medicines colchicine, phenobarbital, primidone, topiramate and pregabalin alongside metformin and the acid suppressants.","cites":["allen2018","nice2024"],"grade":"A"}}],"caveats":[{"text":"Serum B12 misses tissue deficiency, and methylmalonic acid has a confounder of its own. The serum level was normal in 5.2 percent of recognized deficiency in one series and above 200 pg/mL in 12 of 419 patients, each with raised MMA and homocysteine. But MMA rises when the kidneys clear it slowly: in the folate-deficient comparison group, MMA was raised in 12.2 percent, all but one attributable to renal insufficiency or low blood volume, which is why NICE reads MMA against the laboratory's own range and the haematology guideline says the clinical picture, not any single test, decides.","cites":["lindenbaum1990","savage1994","nice2024","devalia2014"],"grade":"A"},{"text":"Supplements and injections change the number within days. NICE asks that blood be drawn before replacement starts, that the clinician ask about over-the-counter B12 in tablets, injections or patches because they can raise total or active B12 without fully treating a deficiency, and that the test not be repeated at all in someone on injections. Assays also fail in specific ways: intrinsic factor antibodies can interfere with immunoassays, so the BOND review says those methods should not be used to measure B12 in pernicious anemia, and lithium-heparin tubes can produce falsely raised results.","cites":["nice2024","allen2018"],"grade":"A"},{"text":"Pregnancy and the combined oral contraceptive pill lower the total number without lowering supply. Concentrations are 25 to 30 percent lower in mid-pregnancy from altered binding to haptocorrin rather than dilution, and holotranscobalamin is unaffected, which is why NICE uses active B12 as the initial test in pregnancy; it also cautions that the combined pill can lower total B12 without causing a deficiency.","cites":["allen2018","nice2024"],"grade":"A"},{"text":"Folate and B12 interact. In 1,459 older Americans surveyed after folic acid fortification, a serum folate above 59 nmol/L (the 80th percentile) in people with low B12 status went with 3.1 times the odds of anemia and 2.6 times the odds of cognitive impairment, while the same high folate with normal B12 went with less cognitive impairment (odds 0.4). High-dose folic acid alongside an unrecognized B12 deficiency is the combination to avoid; folate also raises homocysteine's dependence on B12, so a raised homocysteine with ample folate points at B12.","cites":["morris2007","nice2024"],"grade":"C"},{"text":"Week to week, a healthy person's serum B12 varies by about 7 percent, so two results need to differ by about 20 percent before the change is more than noise; between-person variation (24 percent) is far larger, which makes a person's own trend more informative than a population range. Holotranscobalamin varies more (13 percent within a person). A study of 15 volunteers over 10 weeks.","cites":["brokner2017"],"grade":"C"},{"text":"A normal blood count rules nothing out, and homocysteine is the companion number. NICE says not to exclude deficiency on the absence of anemia or large red cells, and the series above put the figure at 28 percent of neurological deficiency without either. For the methylation side of the story, the homocysteine entry on this shelf covers what that number adds and what lowering it did and did not do in trials.","cites":["nice2024","lindenbaum1988"],"grade":"A"}],"seeAPhysician":["Numbness or tingling in the hands or feet, trouble with balance or walking, memory or mood change, or vision change together with a B12 below about 350 pg/mL, which NICE treats as a reason to start replacement without waiting for a methylmalonic acid result.","Anemia, especially with large red cells, with any B12 value: the treatment response and the search for the cause both belong with a physician.","A B12 far above the reference range in someone taking no B12 supplement or injection, which in large cohorts marked liver disease, kidney disease and blood cancers more often than good nutrition.","An indeterminate value (about 200 to 350) in someone taking metformin or long-term acid suppression, after bariatric or bowel surgery, with an autoimmune condition or a family history of pernicious anemia, or who is pregnant or breastfeeding.","Symptoms that persist 3 to 6 months after a reassuring result, which NICE says is a reason to look for other causes and to repeat the test."],"openQuestions":["No trial has treated a symptom-free adult with a B12 in the middle of the reference range to a higher target and measured outcomes; the two trials that raised low-normal levels in people without symptoms found nothing to measure, and an optimal level is undefined.","Whether low-normal B12 causes brain atrophy, as one small cohort suggests, or merely accompanies it, has not been tested in a trial with brain volume as the outcome using B12 alone.","Which second-line test should be standard, methylmalonic acid or holotranscobalamin, and at what cut-off, is unsettled; both guidelines ask laboratories to validate their own ranges.","Why high serum B12 predicts cancer and mortality in people not taking supplements is unexplained beyond release from liver stores and reduced clearance, and whether a supplement-driven high value carries any risk is unknown.","Whether the different oral forms (cyanocobalamin, methylcobalamin, adenosylcobalamin) differ in any clinical outcome has not been tested head to head in an adequately sized trial."],"questions":[{"q":"Is a B12 of 300 low?","a":"Most laboratories print it as normal, and NICE calls 180 to 350 pg/mL indeterminate rather than low. In older adults with a value of 201 to 300, 56 percent had raised methylmalonic acid or homocysteine, the same share as those at 200 and below, so a 300 with fatigue, tingling, poor balance or memory change is a reason to check methylmalonic acid, and with neurological symptoms NICE starts replacement without waiting. A 300 without symptoms is a number to understand (diet, metformin, acid suppression, stomach health) and to recheck if symptoms appear."},{"q":"What is an optimal vitamin B12 level?","a":"No trial defines one. The evidence is firm that below about 200 pg/mL is deficiency and that about 200 to 350 needs a methylmalonic acid to interpret. Above 350 the cohorts disagree: one small study saw the least brain shrinkage above about 420, and a large Dutch cohort saw the lowest mortality below about 455 with higher mortality in its top quartile, most likely because illness raises the number. Trials that raised low-normal levels in people without symptoms improved nothing measurable. A physician's zone above the grey zone is a judgment, and this page says so."},{"q":"Can B12 be too high?","a":"From food, no: absorption saturates. From supplements, a high serum value is expected and no toxicity threshold has been set. The caution is a high value in someone not taking B12: in 333,667 Danes, a level above the reference range was followed by a cancer diagnosis 3.4 to 6.3 times as often as expected over the next year, and in 5,571 Dutch adults the top quartile carried 1.85 times the mortality. The number was reporting liver, kidney or blood disease already present, which is why an unexplained high B12 is a question for a physician rather than a reason to stop a supplement."}],"citations":[{"id":"green2017","authors":"Green R, Allen LH, Bjørke-Monsen AL, et al.","title":"Vitamin B12 deficiency","journal":"Nature Reviews Disease Primers","year":2017,"pmid":"28660890","doi":"10.1038/nrdp.2017.40","type":"review","finding":"B12 has a central role in DNA synthesis, methylation and mitochondrial metabolism; subclinical deficiency affects 2.5 to 26 percent of the general population depending on the definition; biomarkers are low total and transcobalamin-bound B12 and raised homocysteine and methylmalonic acid; cut-offs remain debated.","url":"https://pubmed.ncbi.nlm.nih.gov/28660890/"},{"id":"allen2018","authors":"Allen LH, Miller JW, de Groot L, et al.","title":"Biomarkers of Nutrition for Development (BOND): Vitamin B-12 Review","journal":"Journal of Nutrition","year":2018,"pmid":"30500928","doi":"10.1093/jn/nxy201","type":"consensus","finding":"Expert panel review: deficiency cut-offs typically 120 to 180 pmol/L (most common 148); 150 to 250 pmol/L depleted; 1.0 pg/mL = 0.7378 pmol/L; NHANES 2.8, 10.5 and 25.6 percent below 148, 200 and 258 pmol/L; MMA the most sensitive and specific marker; absorption about 50 percent of 1 µg and 1 percent of 500 µg; animal foods the only dietary source; pregnancy lowers B12 25 to 30 percent via haptocorrin; intrinsic factor antibodies interfere with immunoassays; IF antibody test more than 95 percent specific, 50 to 70 percent sensitive; gastric bypass more than 3 times the odds of deficiency.","url":"https://pubmed.ncbi.nlm.nih.gov/30500928/"},{"id":"nice2024","authors":"National Institute for Health and Care Excellence","title":"Vitamin B12 deficiency in over 16s: diagnosis and management (NICE guideline NG239)","journal":"National Institute for Health and Care Excellence, London","year":2024,"url":"https://www.nice.org.uk/guidance/ng239","type":"guideline","finding":"Total B12 below 180 ng/L confirms deficiency, 180 to 350 indeterminate, above 350 unlikely (active B12: 25 and 70 pmol/L); MMA when indeterminate with symptoms; active B12 in pregnancy; do not rule out on absence of anemia or macrocytosis; risk-factor list including metformin, PPIs, H2 blockers, anticonvulsants, bariatric surgery, autoimmune disease, nitrous oxide; oral at least 1 mg a day for malabsorption, lifelong injections for autoimmune gastritis or total gastrectomy or ileal resection; test before replacement; over-the-counter B12 and the combined pill affect the result."},{"id":"devalia2014","authors":"Devalia V, Hamilton MS, Molloy AM; British Committee for Standards in Haematology","title":"Guidelines for the diagnosis and treatment of cobalamin and folate disorders","journal":"British Journal of Haematology","year":2014,"pmid":"24942828","doi":"10.1111/bjh.12959","type":"guideline","finding":"No gold-standard test; serum cobalamin first line, methylmalonic acid second line, homocysteine less specific; holotranscobalamin a potential first-line test with a grey area; definitive cut-offs not possible and local ranges needed; treat without delay when clinical features are strong; oral therapy may be suitable.","url":"https://pubmed.ncbi.nlm.nih.gov/24942828/"},{"id":"nexo2011","authors":"Nexo E, Hoffmann-Lücke E","title":"Holotranscobalamin, a marker of vitamin B-12 status: analytical aspects and clinical utility","journal":"American Journal of Clinical Nutrition","year":2011,"pmid":"21593496","doi":"10.3945/ajcn.111.013458","type":"review","finding":"About one quarter of circulating B12 is bound to transcobalamin and available to cells ('active B12'); studies comparing holoTC with total B12 against an MMA standard suggest holoTC performs better.","url":"https://pubmed.ncbi.nlm.nih.gov/21593496/"},{"id":"andres2004","authors":"Andrès E, Loukili NH, Noel E, et al.","title":"Vitamin B12 (cobalamin) deficiency in elderly patients","journal":"CMAJ","year":2004,"pmid":"15289425","doi":"10.1503/cmaj.1031155","type":"review","finding":"Deficiency occurs in more than 20 percent of older people; food-cobalamin malabsorption causes more than 60 percent of cases and pernicious anemia 15 to 20 percent.","url":"https://pubmed.ncbi.nlm.nih.gov/15289425/"},{"id":"kuzminski1998","authors":"Kuzminski AM, Del Giacco EJ, Allen RH, Stabler SP, Lindenbaum J","title":"Effective treatment of cobalamin deficiency with oral cobalamin","journal":"Blood","year":1998,"pmid":"9694707","type":"rct","finding":"38 deficient patients randomized to 2 mg oral daily or 1 mg intramuscular on a tapering schedule: equal hematologic and neurologic recovery; serum B12 93 to 1,005 pg/mL (oral) versus 95 to 325 (injection) at 4 months; normal range stated as 200 to 900 pg/mL.","url":"https://pubmed.ncbi.nlm.nih.gov/9694707/"},{"id":"lindenbaum1988","authors":"Lindenbaum J, Healton EB, Savage DG, et al.","title":"Neuropsychiatric disorders caused by cobalamin deficiency in the absence of anemia or macrocytosis","journal":"New England Journal of Medicine","year":1988,"pmid":"3374544","doi":"10.1056/NEJM198806303182604","type":"cohort","finding":"Of 141 consecutive patients with neuropsychiatric disease from B12 deficiency, 40 (28 percent) had no anemia or macrocytosis; serum B12 above 200 pg/mL in 2, 100 to 200 in 16, below 100 in 22; 39 of 39 improved with treatment.","url":"https://pubmed.ncbi.nlm.nih.gov/3374544/"},{"id":"lindenbaum1990","authors":"Lindenbaum J, Savage DG, Stabler SP, Allen RH","title":"Diagnosis of cobalamin deficiency: II. Relative sensitivities of serum cobalamin, methylmalonic acid, and total homocysteine concentrations","journal":"American Journal of Hematology","year":1990,"pmid":"2339684","doi":"10.1002/ajh.2830340205","type":"cohort","finding":"Serum cobalamin normal in 9 (5.2 percent) of 173 patients with recognized deficiency and above 200 pg/mL in 12 of 419, each with raised MMA and homocysteine; in early relapse MMA abnormal more than twice as often as serum cobalamin.","url":"https://pubmed.ncbi.nlm.nih.gov/2339684/"},{"id":"pennypacker1992","authors":"Pennypacker LC, Allen RH, Kelly JP, et al.","title":"High prevalence of cobalamin deficiency in elderly outpatients","journal":"Journal of the American Geriatrics Society","year":1992,"pmid":"1447433","type":"cohort","finding":"152 outpatients aged 65 to 99: 14.5 percent deficient by B12 of 300 pg/mL or below plus raised metabolites; raised metabolites in 56 percent of those at 201 to 300 versus 62 percent at 200 or below; metabolites corrected in every treated patient.","url":"https://pubmed.ncbi.nlm.nih.gov/1447433/"},{"id":"lindenbaum1994","authors":"Lindenbaum J, Rosenberg IH, Wilson PW, Stabler SP, Allen RH","title":"Prevalence of cobalamin deficiency in the Framingham elderly population","journal":"American Journal of Clinical Nutrition","year":1994,"pmid":"8017332","doi":"10.1093/ajcn/60.1.2","type":"cross-sectional","finding":"548 Framingham survivors: 40.5 percent below 258 pmol/L versus 17.9 percent of younger controls; MMA markedly raised in 11.3 percent; prevalence of deficiency 12 percent or more; many with 'normal' serum values metabolically deficient.","url":"https://pubmed.ncbi.nlm.nih.gov/8017332/"},{"id":"savage1994","authors":"Savage DG, Lindenbaum J, Stabler SP, Allen RH","title":"Sensitivity of serum methylmalonic acid and total homocysteine determinations for diagnosing cobalamin and folate deficiencies","journal":"American Journal of Medicine","year":1994,"pmid":"8154512","doi":"10.1016/0002-9343(94)90149-x","type":"cross-sectional","finding":"434 episodes of B12 deficiency: MMA raised in 98.4 percent, homocysteine in 95.9 percent, both normal in one patient; hematocrit normal in 28 percent, MCV in 17 percent; homocysteine raised in 91 percent of folate deficiency; MMA raised in 12.2 percent of folate deficiency, almost all from renal insufficiency.","url":"https://pubmed.ncbi.nlm.nih.gov/8154512/"},{"id":"vogiatzoglou2008","authors":"Vogiatzoglou A, Refsum H, Johnston C, et al.","title":"Vitamin B12 status and rate of brain volume loss in community-dwelling elderly","journal":"Neurology","year":2008,"pmid":"18779510","doi":"10.1212/01.wnl.0000325581.26991.f2","type":"cohort","finding":"107 volunteers aged 61 to 87, 5 years of MRI: bottom tertile of B12 (below 308 pmol/L) had 6.17 times the odds of greater brain-volume loss after adjustment; same for low holoTC; MMA and homocysteine not associated.","url":"https://pubmed.ncbi.nlm.nih.gov/18779510/"},{"id":"morris2007","authors":"Morris MS, Jacques PF, Rosenberg IH, Selhub J","title":"Folate and vitamin B-12 status in relation to anemia, macrocytosis, and cognitive impairment in older Americans in the age of folic acid fortification","journal":"American Journal of Clinical Nutrition","year":2007,"pmid":"17209196","doi":"10.1093/ajcn/85.1.193","type":"cross-sectional","finding":"NHANES 1999 to 2002, 1,459 seniors: low B12 status (below 148 pmol/L or MMA above 210 nmol/L) carried odds of 2.7 for anemia and 2.5 for cognitive impairment; with low B12, serum folate above 59 nmol/L carried odds 3.1 for anemia and 2.6 for cognitive impairment; with normal B12, high folate was protective (0.4).","url":"https://pubmed.ncbi.nlm.nih.gov/17209196/"},{"id":"dangour2015","authors":"Dangour AD, Allen E, Clarke R, et al.","title":"Effects of vitamin B-12 supplementation on neurologic and cognitive function in older people: a randomized controlled trial","journal":"American Journal of Clinical Nutrition","year":2015,"pmid":"26135351","doi":"10.3945/ajcn.115.110775","type":"rct","finding":"201 people aged 75 and over with B12 107 to 210 pmol/L and no anemia: 1 mg oral B12 daily for 12 months raised B12 from 231 to 641 pmol/L and holoTC from 56 to 240 with no effect on nerve conduction, neurological examination or cognition.","url":"https://pubmed.ncbi.nlm.nih.gov/26135351/"},{"id":"eussen2006","authors":"Eussen SJ, de Groot LC, Joosten LW, et al.","title":"Effect of oral vitamin B-12 with or without folic acid on cognitive function in older people with mild vitamin B-12 deficiency: a randomized, placebo-controlled trial","journal":"American Journal of Clinical Nutrition","year":2006,"pmid":"16895884","doi":"10.1093/ajcn/84.1.361","type":"rct","finding":"195 people aged 70 and over with mild B12 deficiency: 1,000 µg B12 daily, with or without 400 µg folic acid, for 24 weeks corrected the biochemistry and did not improve cognitive function.","url":"https://pubmed.ncbi.nlm.nih.gov/16895884/"},{"id":"clarke2014","authors":"Clarke R, Bennett D, Parish S, et al.; B-Vitamin Treatment Trialists' Collaboration","title":"Effects of homocysteine lowering with B vitamins on cognitive aging: meta-analysis of 11 trials with cognitive data on 22,000 individuals","journal":"American Journal of Clinical Nutrition","year":2014,"pmid":"24965307","doi":"10.3945/ajcn.113.076349","type":"meta-analysis","finding":"B vitamins lowered homocysteine 26 to 28 percent with no effect on any cognitive domain or on global cognition across 11 trials and 22,000 people.","url":"https://pubmed.ncbi.nlm.nih.gov/24965307/"},{"id":"smith2010","authors":"Smith AD, Smith SM, de Jager CA, et al.","title":"Homocysteine-lowering by B vitamins slows the rate of accelerated brain atrophy in mild cognitive impairment: a randomized controlled trial","journal":"PLoS One","year":2010,"pmid":"20838622","doi":"10.1371/journal.pone.0012244","type":"rct","finding":"VITACOG: 168 people over 70 with mild cognitive impairment completing MRI; folic acid 0.8 mg, B12 0.5 mg and B6 20 mg daily for 24 months gave brain atrophy of 0.76 versus 1.08 percent a year, 53 percent slower when baseline homocysteine was above 13 µmol/L.","url":"https://pubmed.ncbi.nlm.nih.gov/20838622/"},{"id":"floresguerrero2020","authors":"Flores-Guerrero JL, Minovic I, Groothof D, et al.","title":"Association of Plasma Concentration of Vitamin B12 With All-Cause Mortality in the General Population in the Netherlands","journal":"JAMA Network Open","year":2020,"pmid":"31940038","doi":"10.1001/jamanetworkopen.2019.19274","type":"cohort","finding":"5,571 Dutch adults not taking B12 supplements, median 8.2 years: median B12 394 pg/mL; 33.8 versus 65.7 deaths per 10,000 person-years in the lowest and highest quartiles; fully adjusted HR 1.85 for the top quartile (above 455.41 pg/mL) and 1.25 per SD; higher B12 tracked with higher liver enzymes; mechanism proposed as release from liver stores or reduced clearance.","url":"https://pubmed.ncbi.nlm.nih.gov/31940038/"},{"id":"arendt2013","authors":"Arendt JF, Pedersen L, Nexo E, Sørensen HT","title":"Elevated plasma vitamin B12 levels as a marker for cancer: a population-based cohort study","journal":"Journal of the National Cancer Institute","year":2013,"pmid":"24249744","doi":"10.1093/jnci/djt315","type":"cohort","finding":"333,667 Danes referred for a B12 test without cancer or B12 treatment: 6 percent above the upper reference limit (601 pmol/L); first-year cancer incidence 3.44 times expected at 601 to 800 pmol/L and 6.27 times above 800, highest for hematological and smoking- and alcohol-related cancers.","url":"https://pubmed.ncbi.nlm.nih.gov/24249744/"},{"id":"pawlak2013","authors":"Pawlak R, Parrott SJ, Raj S, Cullum-Dugan D, Lucus D","title":"How prevalent is vitamin B(12) deficiency among vegetarians?","journal":"Nutrition Reviews","year":2013,"pmid":"23356638","doi":"10.1111/nure.12001","type":"review","finding":"18 studies defining deficiency by MMA or holoTC: deficiency in 62 percent of pregnant vegetarians, 25 to 86 percent of children, 21 to 41 percent of adolescents and 11 to 90 percent of the elderly; higher in vegans and in lifelong vegetarians.","url":"https://pubmed.ncbi.nlm.nih.gov/23356638/"},{"id":"wang2018","authors":"Wang H, Li L, Qin LL, Song Y, Vidal-Alaball J, Liu TH","title":"Oral vitamin B(12) versus intramuscular vitamin B(12) for vitamin B(12) deficiency","journal":"Cochrane Database of Systematic Reviews","year":2018,"pmid":"29543316","doi":"10.1002/14651858.CD004655.pub3","type":"meta-analysis","finding":"3 trials, 153 people: no clinically relevant difference in serum B12 between 1,000 µg a day by mouth and intramuscular B12 (low-quality evidence); oral treatment cost less; no trial reported clinical symptoms.","url":"https://pubmed.ncbi.nlm.nih.gov/29543316/"},{"id":"dejager2010","authors":"de Jager J, Kooy A, Lehert P, et al.","title":"Long term treatment with metformin in patients with type 2 diabetes and risk of vitamin B-12 deficiency: randomised placebo controlled trial","journal":"BMJ","year":2010,"pmid":"20488910","doi":"10.1136/bmj.c2181","type":"rct","finding":"390 patients, 4.3 years: metformin lowered B12 19 percent; absolute risk of deficiency (below 150 pmol/L) 7.2 points higher (number needed to harm 13.8) and of low B12 (150 to 220 pmol/L) 11.2 points higher; homocysteine 23.7 µmol/L in those deficient versus 14.9 with normal B12.","url":"https://pubmed.ncbi.nlm.nih.gov/20488910/"},{"id":"aroda2016","authors":"Aroda VR, Edelstein SL, Goldberg RB, et al.; Diabetes Prevention Program Research Group","title":"Long-term Metformin Use and Vitamin B12 Deficiency in the Diabetes Prevention Program Outcomes Study","journal":"Journal of Clinical Endocrinology and Metabolism","year":2016,"pmid":"26900641","doi":"10.1210/jc.2015-3754","type":"rct","finding":"Low B12 (203 pg/mL or below) 4.3 versus 2.3 percent at 5 years on metformin versus placebo; low or borderline (298 pg/mL or below) 19.1 versus 9.5 percent at 5 years and 20.3 versus 15.6 percent at 13 years; odds of deficiency 1.13 per year of metformin.","url":"https://pubmed.ncbi.nlm.nih.gov/26900641/"},{"id":"lam2013","authors":"Lam JR, Schneider JL, Zhao W, Corley DA","title":"Proton pump inhibitor and histamine 2 receptor antagonist use and vitamin B12 deficiency","journal":"JAMA","year":2013,"pmid":"24327038","doi":"10.1001/jama.2013.280490","type":"cross-sectional","finding":"25,956 cases of B12 deficiency and 184,199 controls: two or more years of PPIs odds 1.65, of H2 blockers 1.25; more than 1.5 PPI pills a day 1.95 versus 1.63 below 0.75.","url":"https://pubmed.ncbi.nlm.nih.gov/24327038/"},{"id":"brokner2017","authors":"Brokner M, Hager HB, Lindberg M","title":"Biological variation of holotranscobalamin and cobalamin in healthy individuals","journal":"Scandinavian Journal of Clinical and Laboratory Investigation","year":2017,"pmid":"28670999","doi":"10.1080/00365513.2017.1335881","type":"laboratory","finding":"15 healthy volunteers over 10 weeks: within-subject variation 6.7 percent for cobalamin and 13.0 percent for holoTC; between-subject 24.1 and 24.6 percent; reference change value 20.1 percent for cobalamin.","url":"https://pubmed.ncbi.nlm.nih.gov/28670999/"},{"id":"banerjee2003","authors":"Banerjee R, Ragsdale SW","title":"The many faces of vitamin B12: catalysis by cobalamin-dependent enzymes","journal":"Annual Review of Biochemistry","year":2003,"pmid":"14527323","doi":"10.1146/annurev.biochem.72.121801.161828","type":"review","finding":"Cobalamin enzymes fall into isomerases (such as methylmalonyl-CoA mutase) and methyltransferases (such as methionine synthase), using the cobalt-carbon bond differently.","url":"https://pubmed.ncbi.nlm.nih.gov/14527323/"},{"id":"gherasim2013","authors":"Gherasim C, Lofgren M, Banerjee R","title":"Navigating the B12 road: assimilation, delivery, and disorders of cobalamin","journal":"Journal of Biological Chemistry","year":2013,"pmid":"23539619","doi":"10.1074/jbc.R113.458810","type":"review","finding":"The reactivity and scarcity of B12 drove higher eukaryotes to evolve an elaborate uptake and intracellular delivery system; new trafficking proteins were still being found after 2009.","url":"https://pubmed.ncbi.nlm.nih.gov/23539619/"},{"id":"roth1996","authors":"Roth JR, Lawrence JG, Bobik TA","title":"Cobalamin (coenzyme B12): synthesis and biological significance","journal":"Annual Review of Microbiology","year":1996,"pmid":"8905078","doi":"10.1146/annurev.micro.50.1.137","type":"review","finding":"B12 synthesis is confined to prokaryotes; the pathway likely evolved for anaerobic fermentation, with siroheme, chlorophyll and heme branching from it later as oxygen rose.","url":"https://pubmed.ncbi.nlm.nih.gov/8905078/"},{"id":"croft2005","authors":"Croft MT, Lawrence AD, Raux-Deery E, Warren MJ, Smith AG","title":"Algae acquire vitamin B12 through a symbiotic relationship with bacteria","journal":"Nature","year":2005,"pmid":"16267554","doi":"10.1038/nature04056","type":"laboratory","finding":"Plants have no B12-dependent enzymes; 171 of 326 algal species need external B12, a dependence that arose repeatedly by losing the B12-independent methionine synthase, with bacteria as the source.","url":"https://pubmed.ncbi.nlm.nih.gov/16267554/"},{"id":"degnan2014","authors":"Degnan PH, Taga ME, Goodman AL","title":"Vitamin B12 as a modulator of gut microbial ecology","journal":"Cell Metabolism","year":2014,"pmid":"25440056","doi":"10.1016/j.cmet.2014.10.002","type":"review","finding":"Some gut microbes synthesize cobalamin, but it is a precious resource in the gut and likely not provisioned to the host in significant quantities.","url":"https://pubmed.ncbi.nlm.nih.gov/25440056/"}],"written":"2026-10-10","updated":"2026-10-10","reviewed":true,"reviewedOn":"2026-10-12","related":{"essays":["diet-and-your-epigenome-eating-to-turn-good-genes-on","what-is-epigenetics-and-why-it-matters-for-health","what-is-metabolomics-a-beginner-s-guide"],"glossary":["Methylmalonic acid (MMA)","Methylation","Homocysteine","MTHFR","Reference range","Optimal range"],"symptoms":["fatigue","brain-fog"],"entries":["homocysteine","ferritin","hba1c","fasting-glucose"]},"url":"https://taggehealth.com/labs/vitamin-b12","author":"Daniel Tagge, MD","publisher":"Tagge Health","license":"Free to read and cite with attribution to the URL above."}