Vitamin B6 On The Label: One Dose, Two Upper Limits

What the label prints, and the two ceilings

The Gelagen panel lists vitamin B6 as pyridoxine HCl, 2 mg, per serving, and the serving is two gummies. The reference intake it is normally read against is the Recommended Dietary Allowance of 1.3 mg a day for adults up to 50, so the row is roughly one and a half times the allowance. That is the whole of the label’s side of the story, and it is a short one: a form, an amount and a comparison with an allowance.

The interesting part is on the other side of the ledger, at the ceiling. Vitamin B6 is one of the few nutrients where two well-regarded bodies have looked at broadly the same evidence and published upper limits that sit more than eight times apart. In the United States the Tolerable Upper Intake Level for adults is 100 mg a day. In Europe, the European Food Safety Authority’s 2023 opinion sets it at 12 mg a day. A shopper who reads a US label and a European one side by side can be forgiven for wondering which of them is wrong.

Neither is, exactly. They are answers to slightly different questions, built from different studies, with different amounts of caution added on top. Working out why is a useful way to learn how an upper limit is made, and it puts the 2 mg on this panel in proportion without any drama attached.

Two things are worth noticing before going further. First, the printed amount is under the lower of the two ceilings by a wide margin, so on this line alone nobody has to choose between the American and the European view. Second, both ceilings exist for the same reason, a nerve problem, and that reason is where the story starts.

Where the concern began: high doses, in the thousands of milligrams

Vitamin B6 was long treated as harmless because it is water soluble and the body clears what it does not use. That assumption broke in 1983, when Schaumburg and colleagues described seven adults who developed ataxia and severe sensory nerve dysfunction after taking large daily amounts of pyridoxine. Four were severely disabled and all improved after they stopped. The authors called it a new megavitamin syndrome and asked for safe guidelines to be set. The European opinion’s review of the literature adds the detail that these patients had taken roughly 2 to 6 grams a day, several of them for premenstrual symptoms, for periods from a couple of months to more than three years.

The units matter here. Two to six grams is a thousand to three thousand times the printed amount on this panel. It is the difference between a vitamin and a drug, and nobody reading that first report would confuse the two.

A later prospective study is the clearest evidence that the effect is dose related rather than a quirk of a few unlucky patients. Berger and colleagues gave 1 or 3 grams a day to five healthy volunteers and followed them with blood levels, symptoms and sensory testing. They stopped the pyridoxine at the first sign of an abnormality. Volunteers on the higher dose became symptomatic sooner than those on the lower one, and the symptoms kept progressing for two to three weeks after the vitamin was stopped, a pattern the authors called coasting, even though blood levels had returned to normal. That is a small study of five people at very high doses, and it should be read as a demonstration of mechanism and direction, not as a tolerance threshold.

Then came the reports that pulled the numbers down. Parry and Bredesen described sixteen patients with neuropathy linked to pyridoxine at 0.2 to 5 grams a day and noted that the time to symptoms was inversely related to the daily amount: the more taken, the sooner the trouble. A case report from Waterston and Gilligan in 1987 suggested that long-term intake could cause neuropathy at doses as low as 200 mg a day. So by the late 1980s the lowest documented harmful doses had fallen from grams to a couple of hundred milligrams, and the obvious question was how far further down they went.

That question is what separates the two upper limits.

The US limit: a dose where nothing got worse

The American figure comes from the Institute of Medicine’s 1998 report on the B vitamins. According to the description in the European opinion’s own review of earlier assessments, the American committee set a no-observed-adverse-effect level of 100 mg a day for adults. It took that level from two studies of patients with diabetic neuropathy or carpal tunnel syndrome, in whom neuropathy did not get worse on intakes of 100 to 300 mg a day. It then applied no uncertainty factor at all, so the level for adults became 100 mg.

That reasoning is worth restating in plain terms, because it explains the number without endorsing or attacking it. The US approach found a group of people who took 100 milligrams or more and did not visibly deteriorate, and used the lower end of the range as the ceiling. It is a “did not see harm” ceiling, and the strength of any such ceiling depends on how well the people studied represent everyone else and how long they were watched.

The European authors have reservations on both counts, and they are not new. An earlier European committee, the Scientific Committee on Food, had noted discrepancies in how one of these studies was reported, in the number of patients and in the length of the study, and questioned whether patients with existing nerve disease and a short follow-up were the right population for judging a long-term safe intake in healthy adults. The same opinion records that the American committee, for its part, chose not to use the largest case series on the other side because it judged the administered doses to be insufficiently reliable.

So each side had a reason for setting aside the other’s evidence. That is the root of the gap, and the disagreement is about study quality, not about whether high doses of B6 can hurt nerves. Everyone agrees that they can.

The European limit: the same syndrome, a different study

The study the Europeans lean on is the one the Americans set aside. Dalton and Dalton examined women attending a clinic for premenstrual syndrome who had raised blood B6 levels. Of 172 such women, about 60 per cent had neurological symptoms, among them tingling, numbness, bone pain, muscle weakness and twitching, mostly in the limbs. The symptoms went away when B6 was withdrawn and came back in four women when it was restarted, and recovery was complete within six months of stopping.

The detail that makes this study so awkward, and so useful, is in the averages. The women with symptoms took a mean of 117 mg a day and the women without took 116 mg, essentially the same. What differed was how long they had been taking it: about 2.9 years for the symptomatic group against 1.6 years for the others. In other words, at these intakes the dose alone did not separate those who had trouble from those who did not. Time did, and so, presumably, did individual susceptibility.

The Scientific Committee on Food had used this study in 2000 and taken 100 mg as its starting point, which is the average dose in the case series, then divided by four. Two of those factors of two covered the inverse relationship between dose and time to symptoms, and the limited data. That gave 25 mg a day. When EFSA revisited the question, it worked from the same paper but a different observation. In the lowest dose group, women who had taken supplements of under 50 mg a day for at least six months, neuropathy was reported in 48 per cent. The exact intakes below 50 mg were not recorded in the study, which is a limitation the opinion states openly.

From that and from case reports and national safety-vigilance data, the EFSA panel identified 50 mg a day as a reference point: the lowest intake linked with certainty to neuropathy when taken for more than six months. It stressed that no lowest-observed-effect level could be pinned down from human data and that smaller doses had also been linked with the problem, though not confirmed. It divided the 50 mg by four to allow for the dose and duration relationship and for the thin data, giving 12.5 mg.

It then cross-checked against animals. In a subchronic study in beagle dogs, 50 mg per kilogram of body weight a day for about a hundred days produced loss of myelin in sensory nerve roots without visible symptoms. Applying a factor of 300 and a default adult weight of 70 kg gives about 11.7 mg a day. The panel took the midpoint of the two values and rounded down, and 12 mg a day was the result, for adults including pregnant and lactating women.

Why the numbers sit so far apart

Laid out like this, the eight-fold gap has three ingredients and none of them is a scandal.

• Different evidence chosen. The US level rests on patients who did not get worse on 100 mg or more. The European level rests on a clinic series where people did get worse, and on animals. One asks what was seen to be safe, the other what was seen to be harmful.
• Different starting points. The US reasoning starts from 100 mg and stays there. The European reasoning starts from 50 mg, because the case-control series reported neuropathy at intakes below 100 mg.
• Different caution added. The US level applies no uncertainty factor. The European level divides by four, then blends with an animal-based estimate. The panel’s stated reason for the factor is the inverse relationship between dose and time to onset, together with limited data on where a lowest-effect level truly lies.

The two European figures are not the only ones on the record either. The European opinion tabulates several national and international bodies, and the numbers scatter: 25 mg from the earlier European committee, 10 mg from the UK expert group, 50 mg from the Australian and New Zealand authority, and 100 mg both from the US committee and from the joint WHO and FAO group. A scatter that wide, from serious bodies, tells you the science of the low end is uncertain, and that is the honest reading of it.

A systematic review by Ghavanini and Kimpinski lands in the middle. It notes that neurology practitioners commonly discourage patients from taking more than 50 mg a day because of the risk of a toxic sensory neuronopathy, that the evidence for that practice had never been systematically reviewed, and concludes that very low doses prevent deficiency neuropathy and that pyridoxine above 50 mg a day for extended periods may be harmful and should be discouraged. That is a sensible thing to hold in the mind beside both official figures: a widely followed clinical rule of thumb that falls between them.

Where two milligrams sits against both

Now the practical question. If the printed 2 mg is compared with the cautious European ceiling of 12 mg, it is one sixth of it. Against the American 100 mg, it is one fiftieth. Against the 50 mg reference point from which the European figure was derived, it is one twenty-fifth. On every one of those comparisons the amount is small, and the neuropathy literature is about doses that are far larger, both in amount and in time.

It would be wrong to turn that into a promise, and the sources do not offer one. What the numbers support is narrow and can be said without decoration: the B6 amount on this panel is above the daily allowance, below both published upper limits, and far below any dose at which the cases described above were documented. That is a statement about a number on a label. It is not a claim that B6 does anything for skin, hair or nails, and the panel does not make one.

The reason the ceiling still matters, even with a comfortable margin, is addition. The European opinion’s conclusion is instructive: European populations are unlikely to exceed the level except for regular users of food supplements containing high doses of vitamin B6. In other words the ceiling is approached by people who stack products, not by people who eat and take one modest gummy. The stacking usually happens without anyone deciding it, because B6 appears in multivitamins, in B-complex tablets, in products sold for energy or stress, and in some fortified foods and drinks, and each label lists its own amount.

The second-product amounts in that table are illustrations chosen to show how the sum behaves, not descriptions of any particular product. The arithmetic is the point. A single line that is well inside both ceilings can sit alongside another line that is not, and no label can see the other one.

Notice too that the US ceiling is far more forgiving of the stacked cases than the European one, which is why it matters which you are reading. A person who takes 27 mg in total would be at a quarter of the US level and over twice the European one. Neither figure says that person will be harmed, and neither says they will not. Both say the question deserves a look at how long it has been going on and whether anything odd, such as tingling or numbness in the hands or feet, has appeared.

What this comparison does not say

It is worth being blunt about three limits, because a post about upper limits can easily sound like a warning or like an all-clear when it is neither.

First, an upper limit is not a target. Nobody is meant to aim at 12 mg or 100 mg, and staying far below either is the ordinary, sensible place to be. Second, the upper levels above are for healthy adults. They do not address anyone who is pregnant, breastfeeding or under treatment for a condition, and Gelagen is not intended for pregnant or nursing women in any case. Third, the case reports and the case series describe real but uncommon events, at doses far above this label, in people who had usually been taking B6 for many months. A short-lived scare is not what the literature shows, and neither is a routine event at low doses.

A fourth limit is about what the label cannot tell you. The panel gives an amount per serving of a single form. It does not say what else you take, and it cannot. The sum in the section above is one only you can do, with the labels in your own cupboard. If two of your products list B6, add the numbers. If the total is well under 12 mg, there is nothing more to work out. If it is not, that is a fair thing to raise with a pharmacist or the clinician who looks after your care.

If you ever notice persistent numbness, tingling or unusual sensations in the hands or feet while taking any supplement that carries B6, the literature above suggests that stopping the source and getting it checked is the reasonable response. In the series described earlier the symptoms generally improved after the vitamin was withdrawn, sometimes with a lag of weeks, which is exactly why they are worth catching early.

How to read a B6 line on any label

The method is short enough to use in a shop aisle.

• Find the milligram figure and note the form. Pyridoxine HCl is the everyday supplemental form, and it is the one the studies above used.
• Compare it with the allowance of 1.3 mg for adults, to see whether the line is a small contribution or a large one.
• Add every other B6 figure in your routine, from tablets, powders, drinks and multivitamins, before comparing the total with the ceiling you prefer. The European 12 mg is the cautious choice. The US 100 mg is the permissive one.
• Remember that time is part of the risk. The case-control series suggests that a few years of a moderate intake looked different from a year or two, so a total that has been steady for a long time deserves more attention than the same total last month.

The disagreement between two upper limits is not a reason to distrust either. It is a reasonable illustration of how honest the science is about what it does not know at the low end of a dose range. The Gelagen line sits comfortably beneath both, and that is all the arithmetic can say for it.

Related reading on this journal. See also how a percent Daily Value is worked out and two other small rows on the same panel. The full printed panel is on the supplement facts page.

Sources for this article

Sourced from the product label and PubMed-indexed research. Gelagen amounts are read from the printed supplement facts panel. Every study above is listed with its PubMed number, and no claim here goes beyond what those records state.