Does thiamin (B1) interact with any medications?

The diuretic furosemide flushes thiamin out in urine and has been linked to levels falling into the deficient range.

What this means for you

One diuretic is worth knowing about by name. Furosemide increases how much thiamin you lose in urine, and research has linked its use to levels falling into the deficient range.

That combination matters most in people who already have reason to be low. Heart failure is treated with furosemide, and it often comes with poor appetite, which means the losses land on someone who is eating less to begin with.

Not for everyone. The other group where thiamin runs low is people with long-term alcohol use disorder, where deficiency is common and the consequences reach the nervous system. There, thiamin is a medical treatment rather than a supplement choice.

Neither of these is a reason to stop a diuretic or to start a supplement unsupervised. It is a reason to mention thiamin at the next appointment if furosemide is long-term.

Weight-loss drugs belong on this page now

Not for everyone. Semaglutide taken for obesity has produced published cases of Wernicke's encephalopathy, which is thiamin deficiency reaching the brain. A review found six of them: four women, two men, average age 47, on the drug for about five months on average. Every one had prolonged gut symptoms and large weight loss first. One died. One recovered well. The remaining four were left with chronic Korsakoff syndrome, and one of them also lost the use of their legs for good. Six published cases sit against millions of prescriptions, so this is not an argument against the drug. It is an argument for naming thiamin out loud if vomiting or barely eating has gone on for weeks.

Kidney disease works the same way, and the published reports describe the same run-up: appetite gone, vomiting, weight loss, abdominal pain, diarrhoea. In those reports the size of the treating dose is what separated outcomes. Thiamine into a vein at 500 mg three times a day often brought full recovery. Korsakoff's syndrome, the permanent version, turned up among people given less. The semaglutide cases were treated at the same 500 mg three times a day, which is roughly four hundred times the daily requirement. That number is here so that nobody reads this section and reaches for a supplement bottle. It has no part in this.

Thiamin as a hospital drug, and a contradiction that is still open

If you have read that thiamin saves lives in sepsis, you have read one of several published answers, and the newest one is the weakest.

Unsettled. Given on its own in septic shock, thiamine first looked useful. Six randomised trials in 438 patients showed a fifth fewer deaths in hospital, a risk ratio of 0.80, and less than half as much need for dialysis. A 2026 meta-analysis then pooled nine trials and 520 patients, close to twice the base, and the survival result did not hold. Short-term mortality came out at an odds ratio of 0.78 with a p value of 0.18, which means the difference is no longer distinguishable from chance. The dialysis finding did survive, and survived strongly, at an odds ratio of 0.26. Lactate at 24 hours came down a little. So the part that replicated is the kidneys, and the part that gets quoted is the part that did not.

Unsettled. Given inside the hydrocortisone, vitamin C and thiamine cocktail, it produced nothing. Nine trials and nine cohorts later, the odds ratio for death at 28 to 30 days landed on exactly 1.000, and the figures for hospital deaths and for dialysis missed significance too. That study was testing three drugs at once rather than thiamine by itself, so it does not settle the question either.

There is one thread worth pulling. Inside the 2026 analysis, two studies looked only at patients who were genuinely thiamine deficient, 51 people in total, and there the odds ratio for death was 0.18. Fifty-one patients is a very thin base for a result that large, so nobody should build anything on it. It does point the same way as everything else on this site, which is that thiamine helps people who are short of thiamine and does not appear to do much for people who are not.

A third analysis sits in the middle and is much the largest. Across 35 randomised trials of thiamine given into a vein to critically ill patients, deaths came out at a risk ratio of 0.89, with a range from 0.75 to 1.06 that includes no effect at all. Shock was about eleven hours shorter, lactate and organ failure scores came down a little, and the stay in intensive care ran about half a day longer. The reviewers rated every one of those findings low certainty. None of this is a decision you or a relative make. It is here so that a claim made at you in a waiting room has a shape you recognise.

Written from the evidence cards and the data on this page, then checked back against the sources. Bioma Learn has no human medical reviewer at this time and we say so rather than invent one. Information, not medical advice. How we verify.

Full guide: Thiamin: a deficiency that is mostly about alcohol, not about diet