Vitamin B6: a water-soluble vitamin that can still damage nerves permanently

People assume water-soluble vitamins are harmless at any dose because the excess leaves in urine. Vitamin B6 is the exception that breaks that rule, and the damage it causes at high doses is the same kind of nerve damage its deficiency causes.

What it does in the body

• Vitamin B6 in its coenzyme forms is involved in more than 100 enzyme reactions, mostly in protein metabolism. — general adults Evidence E source

How much you need

• The RDA for vitamin B6 is 1.3 mg a day for both men and women aged 19 to 50. — general adults Evidence E source
• The Tolerable Upper Intake Level for vitamin B6 is 100 mg a day for adults aged 19 and over. — general adults Evidence E source
• The banana is the food everyone names for vitamin B6, and one banana delivers 18% of the daily reference value. A baked potato delivers three times more. — banana raw, potato baked, chicken breast roasted Evidence E source
• The share of American women falling below the vitamin B6 requirement rises tenfold across a lifetime: 5% of girls aged 4-8, 53% of women over 70. — US population by sex and age group Evidence E source
• The 1.7 mg on a label is one of four legal figures for vitamin B6, and the one for pregnancy is higher than the one for adults. — four label populations defined by FDA Evidence E source

How well your body absorbs it

• Most of the vitamin B6 in plant food is bound to a sugar, and in that form the body uses only about 58% of it. — healthy adult men aged 20-35 Evidence B source
• And it is not only a laboratory ratio: women fed a diet where more of the B6 came as the plant-bound form measurably lost B6 status, at the same milligram intake. — healthy women Evidence B source
• Simmering beef loses roughly half its vitamin B6, and a third of what is left goes down the sink with the cooking liquid. — beef and organ meats, simmered, braised or baked Evidence E source

What too little does

• Vitamin B6 deficiency causes a particular anaemia, cracked lips, a swollen tongue, depression and confusion, and weakens the immune system. — people with vitamin B6 deficiency Evidence E source
• When malnutrition puts someone in hospital with acute nerve damage, vitamin B6 is the second most common culprit after thiamine, and it is behind about one case in six. — 83 patients in high-income countries, mean age 39.8 years, 70 per cent women; 72 per cent had had bariatric surgery Evidence C source
• In Parkinson's disease, where levodopa eats into vitamin B6, about four patients in ten have an abnormal B6 level, and it is far more often too low than too high. — people living with Parkinson's disease, in a small set of observational studies focused on polyneuropathy and epilepsy Evidence C source
• The blood level used to decide who is short of vitamin B6 may be set too low: a more sensitive marker starts moving while the standard test still reads normal. — general adult population Evidence E source

Who is most likely to fall short

• People with poor kidney function, including those on dialysis, often have low vitamin B6 levels. — people with impaired renal function Evidence E source

What the evidence does and does not show

• A systematic review of 14 randomised trials found insufficient evidence that vitamin B6 supplements change cognitive function. — people with normal cognition, dementia or ischaemic vascular disease, 14 RCTs Evidence A source
• Pooling 18 studies, vitamin B6 eased nausea in pregnancy both on its own and alongside another medicine. — pregnant women with nausea and vomiting, 18 studies Evidence A source
• In hyperemesis gravidarum, the severe form that puts women in hospital, vitamin B6 did not shorten anything: the stay was if anything about three quarters of a day longer than on placebo. — pregnant women up to 20 weeks with hyperemesis gravidarum; the review covers 25 trials, the vitamin B6 comparison rests on one of them Evidence A source
• In the largest recent trial, doxylamine with pyridoxine cut the nausea score by one point more than placebo on a scale that runs from 3 to 15. — women in early pregnancy with moderate to severe nausea and vomiting, at 13 tertiary hospitals in mainland China Evidence B source
• In the trial that licensed the delayed-release combination, symptom scores fell by 4.8 points on the drug against 3.9 on placebo over two weeks. — pregnant women with nausea and vomiting of pregnancy, treated for 14 days Evidence B source
• Head to head against ondansetron, pyridoxine with doxylamine came out clearly second for nausea. — pregnant women with nausea and vomiting of pregnancy, randomised to 4 mg ondansetron or 25 mg pyridoxine plus 12.5 mg doxylamine Evidence B source
• Low mood is the one deficiency sign that has been tested head on: a month of 100 mg of B6 a day cut depression scores in people who already had symptoms, and did nothing for anxiety or stress. — 267 adults recruited from the general population, 215 of them women, split by baseline Depression Anxiety Stress Scales score Evidence B source
• Oral contraceptives lower vitamin B6, and in a small crossover trial replacing it cut depression scores by a fifth while they rose on placebo. — healthy college women aged 18 to 25 using combined oral contraceptives consistently for at least a year and taking no other supplements Evidence B source
• Low blood B6 tracks with how severe a neurological disease is, which is a reason to measure it but not proof that it caused anything. — people with multiple sclerosis split by disability score, 47 with a score of 4.5 or above and 52 below 4 Evidence C source
• Across 121 observational studies covering nearly two million people, those eating the most vitamin B6 from food had about a fifth lower cancer risk. — 121 observational studies and nine randomised controlled trials Evidence C source
• The same review found the cancer benefit vanished twice over: it weakened once supplements were counted in, and the randomised trials showed no protection at all. — participants of nine randomised controlled trials of vitamin B6 Evidence A source
• In one rare inherited epilepsy, the form of vitamin B6 is the whole treatment: nine in ten of the children who responded to the active form had already failed on ordinary pyridoxine. — patients with pyridox(am)ine 5'-phosphate oxidase (PNPO) deficiency, a neonatal-onset B6-responsive epileptic encephalopathy Evidence C source

What too much does

• Taking 1 to 6 g of oral pyridoxine a day for 12 to 40 months can cause severe and progressive sensory neuropathy with ataxia. — general adults Evidence E source
• A review of the nerve damage literature concluded that taking more than 50 mg of B6 a day for a long stretch may do harm and should be discouraged. — people with peripheral neuropathy attributed to vitamin B6 deficiency or excess Evidence A source
• The same trial recorded a safety signal nobody was looking for: babies of women on doxylamine-pyridoxine were more often small for gestational age. — the same 352 women in early pregnancy with moderate to severe nausea and vomiting, followed to delivery Evidence B source
• Among women taking vitamin B6 for pregnancy nausea, roughly one in seven of those described in the case literature developed nerve symptoms such as tingling, numbness or unsteadiness. — pregnant women using vitamin B6 above the tolerable upper intake for nausea and vomiting, described in 19 reports found in MEDLINE Evidence C source
• Europe cut its safe upper limit for vitamin B6 to 12 mg a day in 2023, eight times lower than the 100 mg limit our own page still quotes from the American authority. — adults, including pregnant and lactating women; separate scaled limits for infants and children Evidence E source
• Long-term high-dose P5P carried a liver cost: one in five patients had raised liver enzymes, and two developed biopsy-confirmed cirrhosis. — the 49 PNPO-deficient patients on pyridoxal-5'-phosphate therapy Evidence D source
• The Netherlands did not wait for a new upper level: it capped vitamin B6 in supplements at 21 mg a day in 2018, and reports of nerve damage fell away after it. — people in the Netherlands who reported neuropathy or its symptoms while taking vitamin B6 Evidence D source
• Two uncomfortable details sit in the same data: the dose barely predicts the blood level, and neuropathy reports kept coming in at low doses after the cap. — the subset of the 224 Dutch reports that recorded both a dose and a plasma level Evidence D source

Medicines it interacts with

• Several antiepileptic drugs speed up the breakdown of vitamin B6, lowering blood levels and raising homocysteine. — people taking valproic acid, carbamazepine or phenytoin Evidence E source

What to do with this

Meat, fish, poultry, potatoes, chickpeas and bananas carry vitamin B6, and the daily requirement is small enough that an ordinary diet passes it without trying.

The reason to be careful is at the other end. B6 is sold in doses of 50 and 100 mg, often for premenstrual symptoms or nerve complaints, and grams a day taken over a year or more cause severe and progressive nerve damage in the arms and legs. That is the same organ system the deficiency harms, which is what makes the mistake so easy to miss.

Two groups deserve a closer look: people with reduced kidney function, who often run low, and people taking valproic acid, carbamazepine or phenytoin, which speed up the breakdown of the vitamin. Both are reasons to ask a clinician rather than to self-prescribe.

Nausea in early pregnancy is the exception where B6 earns its place. Eighteen studies pooled together show a clear improvement, alone or alongside doxylamine, and obstetric guidelines have used it for years. Doses there are set by a clinician, which also keeps you well away from the level where the nerve damage starts. A review of that literature put the caution at more than 50 mg a day taken for months.

Unsettled. That exception has edges. In hyperemesis gravidarum, the severe form treated in hospital, a Cochrane review found B6 shortened nothing, and the stay ran about three quarters of a day longer than on placebo. Against ondansetron it came second. The product sold for sickness pairs pyridoxine with an antihistamine, so the trials test the pair, and the largest of them beat placebo by one point on a scale running from 3 to 15. That same trial found more babies small for gestational age on the combination, an odds ratio of 3.8 with an interval from 1.0 to 14.1. Among women in the case literature who went above the limit, 164 of 1,226 had nerve symptoms.

One more thing is worth saying plainly about this page. Most of the deficiency evidence here is rated low, and that is what the question allows rather than what the search managed. To show what a shortage of B6 does to a person, someone would have to keep a group short of it until they fell ill, and nobody will ever run that trial. What exists instead is case series and counts inside groups already at risk.

The safe ceiling depends on which continent you ask

Unsettled. The 100 mg upper limit on this page comes from the American authority. Europe has moved away from it. The EU committee concluded in 2000 and again in 2012 that harm was unlikely below 100 mg a day, then proposed 25 mg, and in 2023 EFSA cut the limit to 12 mg a day for adults, pregnancy and breastfeeding included. Eight times lower than the number this page prints. The Netherlands did not wait for that review and capped supplements at 21 mg a day from October 2018. The underlying observations are not in dispute: adverse effects have been seen above 50 mg a day taken for months to years, and clear neurotoxicity from 500 mg. What the two authorities disagree about is how much room to leave below the harm. Until they agree, read a 50 mg tablet as a dose two regulators would question and one would not.

The Dutch cap looks like it worked. After October 2018 only one report involved a dose much above the new limit, a 2020 case taking 200 mg a day, and from the second half of 2019 the reports of nerve damage fell away.

Not for everyone. Two details in that same Dutch data are harder to live with. Reports of neuropathy at low doses kept arriving after the cap. And across the cases where both were recorded, the daily dose explained only 15 per cent of how much B6 ended up in the blood. So the number on the tablet is a poor guide to what is happening inside any particular person. If your hands or feet start tingling or going numb while you are taking B6, the dose being a modest one is not reassurance. Stop it and say so to a doctor.

The cancer claim, and why the two halves of it disagree

Myth. Vitamin B6 lowers your cancer risk. This is as clean a worked example as this site has. Across 121 observational studies covering nearly two million people, those eating the most B6 from food had about a fifth less cancer, a risk ratio of 0.78, and for cancers of the digestive tract 0.68. Measured in blood instead of on a questionnaire it looked stronger still, 0.66 overall. Then the same review looked at nine randomised trials, 34,911 people assigned to B6 or not, and found no protection at all. The association also weakened or disappeared once supplement intake was counted alongside food. People who eat a lot of B6 from food are eating vegetables, fish and whole foods, and they differ from everyone else in a hundred ways a questionnaire never records. Giving them the vitamin on its own reproduces none of it.

Plant B6 is worth less than the label suggests

Most of the B6 in plant food is stuck to a sugar, as pyridoxine glucoside, and your gut has to cut it loose before you can use it. In healthy men, the bound form was used at about 58 per cent of the rate of the free vitamin. That is a laboratory ratio, so somebody checked whether it shows up in people. Women fed two diets carrying almost identical B6, 1.52 against 1.44 mg a day, but with 27 per cent against 9 per cent of it in the bound form, came out measurably worse off on the high-bound diet. Urinary and plasma B6 fell, red-cell PLP fell, and more B6 left in the stool.

This is not an argument against plants. It is the reason a diet built mostly on plant sources sits closer to the line than the milligram figure implies, and one more reason the people who run low tend to be older women eating little meat and fish.

Who actually runs short

The share of people below the requirement climbs steadily with age, and for women it climbs much further. Among girls aged 4 to 8 it is 5.1 per cent. By 14 to 18 it is 38.1 per cent, at 51 to 70 it is 52.4 per cent, and past 70 it is 52.8 per cent. Men sit lower at every single age band, 4.1 per cent at 4 to 8 and 36.8 per cent past 70. Part of this is that the requirement itself goes up after 50, and part of it is that intake goes down.

For food, the banana has the reputation and the potato does the work. One banana carries 0.3 mg, which is 18 per cent of the 1.7 mg label figure. One medium baked potato carries 0.9 mg, which is 53 per cent, and a large one 1.2 mg. A medium roast chicken breast also comes to 0.9 mg. If you want more B6 without thinking about it, the potato and the chicken breast are the answer and the banana is a decoration.

Cooking takes a share of it. Simmered beef keeps about 35 per cent of its B6 if you throw the liquid away and 50 per cent if you use it, and the same split holds for roast, braised and sliced beef. Liver runs 45 against 55. Half of what survives the heat can end up in the pan, so the gravy is not a garnish.

One label detail confuses people. The 1.7 mg on a bottle is one of four legal figures. Infants under a year have 0.3 mg, children 1 to 3 have 0.5 mg, adults and children over 4 have 1.7 mg, and pregnancy and breastfeeding have 2.0 mg. A percentage on a package is measured against the adult figure whoever is holding it.

Where the form of B6 is the entire treatment

There is one rare inherited epilepsy, PNPO deficiency, where newborns seize and only the active form of the vitamin works. Of 49 patients given pyridoxal-5'-phosphate, 38 responded, and survival improved against untreated siblings with the same condition. Ordinary pyridoxine had already been tried and failed in 30 of the 33 responders, which is nine in ten. This is the one place on this page where the word on the label decides the outcome.

Not for everyone. That treatment has a price, and it is a warning to anyone tempted by high-dose P5P sold as a better B6. Among those same 49 patients, 10 had raised liver enzymes, nine of them after more than a year on it, and two developed cirrhosis confirmed by biopsy. Both of those needed 50 to 100 mg per kilogram of body weight a day to keep the seizures under control, which for a 70 kg adult would be grams. One further patient was poisoned at 74 mg per kilogram because a supplement label was unclear about what was in it.

Unsettled. There is also a question about where the line for deficiency sits. A plasma PLP of 30 nmol/L is treated as sufficient, and the intake recommendations are built on that figure. A more sensitive metabolic marker starts moving when PLP falls below 100 nmol/L and moves faster below 50. If that marker is measuring something real, a person can be biochemically short of B6 while their test result reads normal. Nobody has shown yet that it matters to how anyone feels, so this is a reason to watch the field rather than a reason to act.

Where to get it from food

The bottom line

The requirement is 1.3 mg a day for adults up to 50, and the upper limit is 100 mg. That gap looks generous until you see what sits beyond it: taking 1 to 6 grams a day for a year or more causes severe and progressive nerve damage in the limbs. Deficiency is uncommon and shows up as anaemia, cracked lips, a swollen tongue and confusion. People with poor kidney function are the group most likely to run low. On the popular claim that B6 protects thinking, a systematic review of 14 randomised trials found insufficient evidence of any effect. The one use with decent support is nausea in early pregnancy, where a pooled analysis of 18 studies found a real improvement, and it is also the one dose worth checking with a doctor first. That support has now been mapped at its edges. In hyperemesis gravidarum, the severe form that puts women in hospital, a Cochrane review found B6 shortened nothing, and the hospital stay ran about three quarters of a day longer than on placebo. Head to head with ondansetron it came second, 20 mm of nausea relief against 51 on a 100 mm scale. The product sold for sickness pairs pyridoxine with doxylamine, an antihistamine, so what the trials test is the pair. In the largest of them, 352 women, the pair beat placebo by one point on a scale running from 3 to 15, and the researchers said they do not know whether a difference that size matters to a patient. The same trial found babies small for gestational age more often on the combination, an odds ratio of 3.8 with an interval running from 1.0 to 14.1. Among women described in the case literature who took B6 above the limit for sickness, 164 of 1,226 had nerve symptoms. Two more groups run low than this page used to name. In Parkinson's disease treated with levodopa, 60 of 145 patients had an abnormal B6 level and 52 of those were low. Combined oral contraceptives lower it as well. The one deficiency sign anyone has tested head on is low mood, where 100 mg a day for a month lowered depression scores among the 58 people who already had symptoms and left anxiety and stress untouched. That was a subgroup found after randomisation, at a dose no diet reaches, and the trial its authors called for has not been run. The 100 mg upper limit quoted above is the American one, and Europe no longer agrees with it. EFSA cut its own limit to 12 mg a day in 2023, for adults including pregnant and lactating women, which is eight times lower. The Netherlands had already capped supplements at 21 mg a day in 2018. If you are holding a 50 mg tablet, one authority calls that half the safe ceiling and another calls it four times over it.

Sources

1. b6-01 · Position of an expert body · expert body statement
   Vitamin B6 in coenzyme forms performs a wide variety of functions in the body and is extremely versatile, with involvement in more than 100 enzyme reactions, mostly concerned with protein metabolism.
   NIH Office of Dietary Supplements, VitaminB6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
2. b6-02 · Position of an expert body · expert body statement
   19–50 years | 1.3 mg | 1.3 mg (value from the source's table, not a sentence)
   NIH Office of Dietary Supplements, VitaminB6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
3. b6-03 · Position of an expert body · expert body statement
   19+ years | 100 mg | 100 mg (value from the source's table, not a sentence)
   NIH Office of Dietary Supplements, VitaminB6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
4. b6-a-10 · Position of an expert body · reference dataset
   Banana, raw · Vitamin B-6 0.212 MG на 100 г [fdc_id 2709224] · 1 banana 126 г 0.3 MG || Potato, baked, NFS · Vitamin B-6 0.3 MG на 100 г [fdc_id 2709383] · 1 medium 285 г 0.9 MG · 1 large 400 г 1.2 MG || Chicken breast, baked, broiled, or roasted, skin eaten, from raw · Vitamin B-6 0.699 MG на 100 г [fdc_id 2705955] · 1 medium breast 130 г 0.9 MG
   USDA FNDDS 2021-2023 and 21 CFR 101.9(c)(8)(iv) · checked 2026-09-21 · independently re-checked
5. b6-a-11 · Position of an expert body · national dietary survey
   vitamin_b6_mg · усі 19+ n=4886 median 1.51 mean 1.94 · EAR (DRI, National Academies) floor: ж 4-8 (0.5) 5.1% · ж 14-18 (1.0) 38.1% · ж 19-30 (1.1) 32.9% · ж 51-70 (1.3) 52.4% · ж 71+ (1.3) 52.8% · ч 71+ (1.4) 36.8%
   NHANES 2021-2023 (DR1TOT/DEMO), weighted; EAR from DRI, National Academies · checked 2026-09-21 · independently re-checked
6. b6-a-12 · Position of an expert body · regulatory reference value
   Vitamin B6 [RDI, Milligrams (mg)] 21 CFR 101.9(c)(8)(iv) — 1.7 Adults and children ≥ 4 years · 0.3 Infants 1 through 12 months · 0.5 Children 1 through 3 years · 2.0 Pregnant women and lactating women
   21 CFR 101.9(c)(8)(iv), Reference Daily Intakes · checked 2026-09-21 · independently re-checked
7. b6-a-03 · Randomised controlled trial(s) · controlled isotopic bioavailability trial
   A dual-label study designed to examine the bioavailability of these compounds when administered simultaneously indicated that the utilization of deuterated PN-glucoside was 58 +/- 13% (mean +/- SEM) relative to that of deuterated PN.
   Journal of Nutrition, 1991 · checked 2026-09-21 · independently re-checked
8. b6-a-04 · Randomised controlled trial(s) · randomised controlled crossover feeding trial
   Women consuming a diet containing a higher percentage of the total vitamin B-6 intake as PNG exhibited a decrease in vitamin B-6 status indicators, consistent with the reduced bioavailability of PNG demonstrated in other studies, equal to a loss of 15-18% of the total vitamin B-6 intake.
   Journal of Nutrition, 1996 · checked 2026-09-21 · independently re-checked
9. b6-a-09 · Position of an expert body · reference dataset
   BEEF,GROUND,SIMMERED,WO/DRIPPINGS — vitamin_b6_mg 35.0 · BEEF,GROUND,SIMMERED,W/DRIPPINGS — vitamin_b6_mg 50.0 · LIVER,SIMMERED,WO/DRIPPINGS — vitamin_b6_mg 45.0 · LIVER,SIMMERED,W/DRIPPINGS — vitamin_b6_mg 55.0
   USDA Table of Nutrient Retention Factors, Release 6 (2007) · checked 2026-09-21 · independently re-checked
10. b6-05 · Position of an expert body · expert body statement
    Vitamin B6 deficiency is associated with microcytic anemia, electroencephalographic abnormalities, dermatitis with cheilosis (scaling on the lips and cracks at the corners of the mouth) and glossitis (swollen tongue), depression and confusion, and weakened immune function
    NIH Office of Dietary Supplements, Vitamin B6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
11. b6-t9 · Observational data · systematic review of case reports and case series
    Thirty-six studies were identified, including 83 patients (mean age 39.8 years; 70% women) with deficient levels of thiamine (54%), pyridoxine (16%), copper (11%) or folate (7%). Common clinical signs were lower extremity weakness (100%), hypesthesia (63%), hypo-/areflexia (54%) and cranial nerve deficits (27%).
    Sørensen et al., Danish Medical Journal, 2025 · checked 2026-09-17 · independently re-checked
12. b6-t10 · Observational data · systematic review of observational studies
    Abnormal B6 levels have been reported in 60 of 145 PwPD (41.4% relative frequency). Low B6 levels were reported in 52 PwPD and high B6 levels were reported in 8 PwPD. There were 14 PwPD, polyneuropathy and low B6. There were 4 PwPD, polyneuropathy and high B6. There were 4 PwPD, epilepsy and low B6. Vitamin B6 level was low in 44.6% of PwPD receiving levodopa-carbidopa intestinal gel and in 30.1% of PwPD receiving oral levodopa-carbidopa.
    Shah et al., Journal of the Neurological Sciences, 2023 · checked 2026-09-17 · independently re-checked
13. b6-a-06 · Position of an expert body · scoping review
    However, the metabolic marker for vitamin B6 status, HK ratio (HKr), starts to increase already when plasma PLP falls below 100 nmol/L and increases more steeply below 50 nmol/L, indicating biochemical deficiency. Consequently, a plasma PLP concentration of 30 nmol/L, may be too low as a marker for an adequate vitamin B6 status.
    Food & Nutrition Research, 2023 · checked 2026-09-21 · independently re-checked
14. b6-06 · Position of an expert body · expert body statement
    People with poor renal function, including those with end-stage renal disease and chronic renal insufficiency, often have low vitamin B6 concentrations
    NIH Office of Dietary Supplements, Vitamin B6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
15. b6-08 · Meta-analysis or systematic review · review of trials cited by expert body
    A systematic review of 14 randomized controlled trials found insufficient evidence of an effect of vitamin B6 supplementation alone or in combination with vitamin B12 and/or folic acid on cognitive function in people with normal cognitive function, dementia, or ischemic vascular disease
    NIH Office of Dietary Supplements, Vitamin B6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
16. b6-t1 · Meta-analysis or systematic review · systematic review and meta-analysis
    Supplementation of pyridoxine alone as well as combined treatment of pyridoxine with an active ingredient as the intervention significantly improved the symptoms of nausea according to Rhode's score [0.78 [95% CI: 0.26, 1.31; p = 0.003; I2 = 57%, p = 0.10)] and PUQE score [0.75 (95% CI: 0.28, 1.22; p = 0.002; I2 = 0%, p = 0.51)], respectively.
    Jayawardena et al., Archives of Gynecology and Obstetrics, 2023 · checked 2026-09-16 · independently re-checked
17. b6-t3 · Meta-analysis or systematic review · Cochrane systematic review of randomised controlled trials
    In a study with 92 participants, women taking vitamin B6 had a slightly longer hospital stay compared with placebo (mean difference (MD) 0.80 days, 95% CI 0.08 to 1.52, moderate-quality evidence). There was insufficient evidence to demonstrate a difference in other outcomes including mean number of episodes of emesis (MD 0.50, 95% CI -0.40 to 1.40, low-quality evidence) or side effects.
    Boelig et al., Cochrane Database of Systematic Reviews, 2016 · checked 2026-09-17 · independently re-checked
18. b6-t4 · Randomised controlled trial(s) · randomised controlled trial
    Participants receiving acupuncture (mean difference [MD], -0.7 [95% CI, -1.3 to -0.1]), doxylamine-pyridoxine (MD, -1.0 [CI, -1.6 to -0.4]), and the combination of both (MD, -1.6 [CI, -2.2 to -0.9]) had a larger reduction in PUQE score over the treatment course than their respective control groups (sham acupuncture, placebo, and sham acupuncture plus placebo).
    Wu et al., Annals of Internal Medicine, 2023 · checked 2026-09-17 · independently re-checked
19. b6-t6 · Randomised controlled trial(s) · randomised controlled trial
    Diclectin use resulted in a significantly larger improvement in symptoms of nausea and vomiting of pregnancy compared with placebo based on both the pregnancy unique quantification of emesis score (-4.8 ± 2.7 vs -3.9 ± 2.6; P = .006) and quality of life.
    Koren et al., American Journal of Obstetrics and Gynecology, 2010 · checked 2026-09-17 · independently re-checked
20. b6-t7 · Randomised controlled trial(s) · randomised controlled trial
    Patients randomized to ondansetron were more likely to have an improvement in their baseline nausea as compared with those using pyridoxine and doxylamine over the course of 5 days of treatment (median VAS score decreased 51 mm [interquartile range 37-64] compared with 20 mm [8-51]; P=.019).
    Oliveira et al., Obstetrics and Gynecology, 2014 · checked 2026-09-17 · independently re-checked
21. b6-t11 · Randomised controlled trial(s) · randomised controlled trial
    Only within the former group, vitamin B6 produced a selective reduction in depression symptoms compared to placebo or vitamin B12 with a very large effect size (partial η 2 = 0.20). We also found evidence that participants whose DASS depression scores narrowly failed to meet criterion for moderate depression (N = 52) experienced a slightly smaller but still statistically substantial reduction in symptoms (partial η 2 = 0.17).
    Field et al., Depression and Anxiety, 2026 · checked 2026-09-17 · independently re-checked
22. b6-t12 · Randomised controlled trial(s) · randomised crossover trial
    BDI-II scores were reduced 20% by vitamin B6 supplementation in comparison to an 11% rise with placebo ingestion (p = 0.046). POMS scores were not significantly impacted by vitamin B6 supplementation.
    Wilson et al., Journal of Dietary Supplements, 2023 · checked 2026-09-17 · independently re-checked
23. b6-t13 · Observational data · cross-sectional analysis within a randomised controlled trial
    Higher EDSS scores were associated with lower pyridoxal 5'-phosphate (vitamin B-6) concentrations (rs: -0.32; 95% CI: -0.49, -0.12; P = 0.011), with the EDSS≥4.5 group also showing lower baseline pyridoxal 5'-phosphate (vitamin B6) concentrations (β: -0.18; 95% CI: -0.30, -0.07; P = 0.007) than the EDSS<4 group.
    Proschinger et al., American Journal of Clinical Nutrition, 2025 · checked 2026-09-17 · independently re-checked
24. b6-a-01 · Observational data · systematic review and meta-analysis of observational studies and trials
    High intake of dietary (food only) vitamin B6 was statistically significantly associated with lower risk of all cancers (relative risk [RR] = 0.78, 95% CI = 0.73 to 0.84) and specific tumors, with special regard to gastrointestinal carcinomas (RR = 0.68, 95% CI = 0.61 to 0.75).
    Journal of the National Cancer Institute, 2017 · checked 2026-09-21 · independently re-checked
25. b6-a-02 · Meta-analysis or systematic review · systematic review and meta-analysis of randomised trials
    When total (food and supplements) intake was considered, the associations were weaker or null. Findings from RCTs did not support a protective effect of vitamin B6 against cancer, although this evidence was graded as low level.
    Journal of the National Cancer Institute, 2017 · checked 2026-09-21 · independently re-checked
26. b6-a-07 · Observational data · systematic review of case-based evidence
    The majority of PLP-responsive patients responded exclusively to PLP, with PN being attempted but ineffective in most of them (n = 30/33, 90.9%)
    Journal of Inherited Metabolic Disease, 2025 · checked 2026-09-21 · independently re-checked
27. b6-04 · Position of an expert body · expert body statement
    Chronic administration of 1–6 g oral pyridoxine per day for 12–40 months can cause severe and progressive sensory neuropathy characterized by ataxia.
    NIH Office of Dietary Supplements, VitaminB6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted
28. b6-t2 · Meta-analysis or systematic review · systematic review
    Supplementation with pyridoxine at doses greater than 50 mg/d for extended duration may be harmful and should be discouraged.
    Ghavanini and Kimpinski, Journal of Clinical Neuromuscular Disease, 2014 · checked 2026-09-16 · independently re-checked
29. b6-t5 · Randomised controlled trial(s) · randomised controlled trial, secondary outcome
    Compared with placebo, a higher risk for births with children who were small for gestational age was observed with doxylamine-pyridoxine (odds ratio, 3.8 [CI, 1.0 to 14.1]).
    Wu et al., Annals of Internal Medicine, 2023 · checked 2026-09-17 · independently re-checked
30. b6-t8 · Observational data · systematic review of case reports and case series
    Among these sources, 12 reports detailed the adverse effects of excessive vitamin B6 intake in women; 164/1226 individuals experienced neurological symptoms such as burning, tingling, paresthesia, ataxia, or perioral numbness. Additionally, out of 245 women, four experienced miscarriages and one had an intrauterine demise.
    Chen et al., International Journal of Gynaecology and Obstetrics, 2025 · checked 2026-09-17 · independently re-checked
31. b6-a-05 · Position of an expert body · agency position reported in a scoping review
    However, in 2023 this UL was reduced to 12 mg/day vitamin B6 for adults (including pregnant and lactating women).
    Food & Nutrition Research, 2023 · checked 2026-09-21 · independently re-checked
32. b6-a-08 · Mechanistic or laboratory data · systematic review of case-based evidence
    elevated transaminases reported in 10 patients (20.4%). Nine of these patients had been on long-term PLP therapy (> 1 year). In two of these patients, liver cirrhosis occurred and was confirmed by liver biopsy
    Journal of Inherited Metabolic Disease, 2025 · checked 2026-09-21 · independently re-checked
33. b6-a-13 · Mechanistic or laboratory data · pharmacovigilance study of spontaneous reports
    led to a regulatory action to lower the maximum daily dose (DD) of vitamin B6 in supplements to 21 mg/day from October 1, 2018
    Pharmacoepidemiology and Drug Safety, 2025 · checked 2026-09-21 · independently re-checked
34. b6-a-14 · Mechanistic or laboratory data · pharmacovigilance study of spontaneous reports
    in this case, 15% of the variability in plasma levels can be explained by the DD, which is a weak correlation
    Pharmacoepidemiology and Drug Safety, 2025 · checked 2026-09-21 · independently re-checked
35. b6-07 · Position of an expert body · expert body statement
    Some antiepileptic drugs, including valproic acid (Depakene, Stavzor), carbamazepine (Carbatrol, Epitol, Tegretol, and others), and phenytoin (Dilantin) increase the catabolism rate of vitamin B6 vitamers, resulting in low plasma PLP concentrations and hyperhomocysteinemia
    NIH Office of Dietary Supplements, Vitamin B6 - Health Professional Fact Sheet · checked 2026-09-16 · extracted