The Real History of B Vitamins
In the early 1960s, a paediatrician at the Cleveland Clinic was looking at a group of children whose problems his profession could not explain.
They had episodes of unexplained collapse, autonomic instability, gastric symptoms that did not fit the usual diagnostic categories, sleep disturbances, and a constellation of features that mainstream paediatrics in that period would have called functional or psychosomatic. The paediatrician was Derrick Lonsdale, and what he noticed in his patients would become the basis for a sixty-year clinical tradition that the mainstream still has not absorbed.
Lonsdale had been trained in classical medicine, but he had also been reading the older biochemistry literature on thiamine. What he saw in his patients, when he tested them carefully, was something that did not match the textbook picture of thiamine deficiency.
Their serum thiamine levels were often normal. They did not have classical beriberi. But when he gave them thiamine in higher doses than the textbook required, they recovered. The pattern was not what mainstream paediatrics was looking for. It was thiamine deficiency that hid behind normal blood tests, that produced symptoms unlike those described in nutritional textbooks, and that responded to therapeutic doses given by a clinician who knew what to look for.
Lonsdale would spend the next six decades writing about it. He would publish hundreds of papers, co-author a major textbook with Chandler Marrs in 2017, contribute regularly to the Hormones Matter platform, and as late as 2021, at the age of ninety-seven, publish a peer-reviewed paper in Cells with Marrs titled “Hiding in Plain Sight: Modern Thiamine Deficiency.”
He died in June 2024, having spent almost all of a hundred-year life arguing for something the medical mainstream had refused to see.
Part 2 of this series told the story of the niacin clinical tradition: how Hoffer, Osmond, Kaufman and Bill Wilson had observed that high-dose B3 produced therapeutic effects in conditions that had nothing to do with classical pellagra, and how the medical mainstream had refused to engage with what they reported. Part 3 is about the parallel story for thiamine. The clinicians changed, the disease shifted from psychiatric to dysautonomic, the institutional response stayed the same.
This article is also about something Part 2 only gestured at. The thiamine tradition is not one lineage but two. Lonsdale and his successors in the United States represent one. A completely independent Italian clinical tradition, beginning with Antonio Costantini in 2010, represents the other. Both arrived at high-dose thiamine as a therapeutic intervention for serious chronic disease.
Both observed the same characteristic finding: blood thiamine levels were normal, but the patients responded to gram doses that the textbook had no framework for. The convergence of two separate clinical lineages on the same intervention, from different starting points, in different countries, in different disease contexts, is the kind of independent confirmation that ought to make mainstream medicine pay attention. It has not.
The Functional Deficiency Tradition
The classical picture of thiamine deficiency in mainstream medicine is beriberi: the disease of polished-rice diets that killed Japanese sailors and Indonesian prisoners in the late nineteenth century, the disease of cardiac failure and peripheral neuropathy, the disease whose chemistry was finally worked out by Robert Williams when he synthesised thiamine in 1936. Once thiamine was available as a supplement, classical beriberi disappeared from most of the developed world. By the 1950s, the medical mainstream considered thiamine deficiency essentially solved.
What Lonsdale saw in his clinic was something the mainstream picture could not accommodate. His patients did not have cardiac beriberi or polyneuropathy. They had what would now be called dysautonomia — autonomic nervous system dysfunction producing wildly varying symptoms across body systems. Postural tachycardia, gastric dysmotility, dizziness, fatigue, anxiety, sleep disturbance, temperature dysregulation, cognitive fogginess. The textbook description of thiamine deficiency from the beriberi literature did not describe any of this. And yet his patients responded to high-dose thiamine.
The mechanism Lonsdale proposed has aged well. Thiamine is required as a cofactor for several enzymes essential to the production of cellular energy, particularly in tissues with high metabolic demand. The brain and the autonomic nervous system are among the most metabolically active tissues in the body. A functional thiamine deficiency, in which the vitamin is present at normal blood levels but the cellular machinery cannot use it effectively, would produce dysautonomic symptoms before it produced classical beriberi.
The classical picture would only emerge in extremis, when the deficiency had become severe enough to damage cardiac and peripheral nerve function. Functional deficiency, in this picture, would be far more common than classical deficiency, would present with subtle and shifting symptoms across body systems, and would be missed entirely by standard serum testing.
Lonsdale’s clinical practice gave him decades of evidence for this picture. So did the work of others. The Hormones Matter platform, which Marrs founded and which became the institutional home for the modern thiamine clinical tradition, has published hundreds of articles documenting the same pattern. Chandler Marrs’s own clinical and research contributions, including the 2017 Elsevier textbook on thiamine deficiency and dysautonomia and the 2021 Cells paper, have anchored the tradition in published peer-reviewed work. Elliot Overton, a more recent contributor whose practical guides to the protocol have reached a wide patient audience, has documented what he calls thiamine’s role as a Great Imitator: the vitamin deficiency that looks like everything else and gets diagnosed as everything else.
The list of conditions that turn out, in the clinical experience of this tradition, to involve unrecognised functional thiamine deficiency is long. Chronic fatigue syndrome. Fibromyalgia. POTS and other dysautonomias. Irritable bowel syndrome and SIBO. Anxiety disorders with somatic features. Histamine intolerance. Mast cell activation syndrome. Some forms of treatment-resistant depression.
Many of these patients have spent years cycling through specialist clinics, accumulating diagnoses and prescriptions, and not improving. When they encounter a clinician who tests for functional thiamine deficiency and treats it appropriately, many of them do.
This is, of course, the kind of claim that ought to be tested in randomised trials. It has not been, at least not at the scale the conditions deserve. The clinical evidence base remains observational and consists of decades of case experience accumulated by serious physicians working largely outside mainstream institutions. The same charge that was levelled at Hoffer’s niacin work has been levelled at Lonsdale’s: not enough trials, not enough mechanism, not enough engagement with the mainstream’s preferred standards of evidence. The same answer applies. The clinicians have been there, doing the work, with patients in front of them. The mainstream has not.
The Forms of Thiamine
The Lonsdale clinical tradition has also done substantial work on something the textbook ignores: the question of which form of thiamine is given. Thiamine hydrochloride is the form supplied in most multivitamins and the form used in classical beriberi treatment. It is water-soluble, absorbed through active transport in the small intestine, and limited in how much can be taken up at a time. For correcting classical deficiency it is adequate. For treating functional deficiency in tissues that are not effectively using thiamine, it often is not.
Three other forms of thiamine have distinct clinical applications. Benfotiamine is a fat-soluble derivative developed in Japan, more bioavailable than thiamine HCl and able to reach tissues that water-soluble thiamine cannot. Thiamine tetrahydrofurfuryl disulphide, known as TTFD, was developed for the treatment of beriberi in postwar Japan and has the unique property of crossing the blood-brain barrier through passive diffusion, which makes it potentially useful for neurological conditions where the brain is not getting enough thiamine despite normal serum levels. Sulbutiamine is a synthetic disulphide derivative used clinically in Europe for asthenia and fatigue, with documented effects on memory and energy.
The forms make a clinical difference because functional deficiency is not corrected by simply increasing the dose of the standard form. Patients who do not respond to high doses of thiamine HCl sometimes respond dramatically to benfotiamine or TTFD. Patients who tolerate one form poorly tolerate another well. The clinical tradition has worked these distinctions out across decades and the published literature now supports much of what the clinicians knew.
The cofactor relationships are equally important. Thiamine requires magnesium to be activated in the cell. The active enzyme form, thiamine pyrophosphate, will not work without magnesium. People who are deficient in magnesium will not respond well to thiamine supplementation, regardless of the dose or the form.
This was a central point in our previous magnesium series, and it returns here in a different context. The thiamine tradition has documented for decades what biochemistry now confirms: the B vitamins do not act in isolation. They are a system of interlocking cofactors that work together or fail together.
A reasonable concern at this point in the article is the question of paradoxical reactions. Some patients who begin thiamine supplementation experience initial worsening of their symptoms before improving. This is documented in the Lonsdale tradition and in the Italian Parkinson’s literature alike. The mechanism appears to involve the sudden activation of mitochondrial energy production in tissues that have been chronically energy-starved, which produces transient symptoms while the cellular machinery adjusts.
The clinical answer is to start at a low dose, increase gradually, and persist through the initial period. Patients who get past the paradoxical phase often improve substantially. Patients who panic and stop sometimes never try again. This is not a side effect in the pharmaceutical sense; it is the kind of adjustment phenomenon that distinguishes a nutrient correcting a long-standing deficiency from a drug producing a foreign effect.
The Italian Confirmation
The independent confirmation of the thiamine tradition from outside the Lonsdale lineage began in 2010, in a clinic in Italy, with a neurologist named Antonio Costantini. The story is documented carefully on the website of the High Dose Thiamine Foundation, the institutional home of what is now called the B1 Protocol.
Costantini started high-dose intramuscular thiamine in patients with ulcerative colitis. His reasoning was clinical, not theoretical. Colitis patients commonly experience profound fatigue, and fatigue is a classical feature of thiamine deficiency. He gave intramuscular thiamine to a group of his patients, expecting to improve their fatigue. The fatigue improved. So did, unexpectedly, several other symptoms. The patients’ blood thiamine levels had been normal throughout. The mechanism, whatever it was, was not correction of classical deficiency.
In 2011, Costantini extended the protocol to multiple sclerosis patients, who also commonly experience fatigue. Again the response was striking. In 2013, the team tested the same protocol in spinocerebellar ataxia type 2, a hereditary degenerative condition for which mainstream neurology offered little. Again, improvement. Later in 2013, noting that hereditary Parkinson’s disease shared genetic features with multiple sclerosis, Costantini and his colleagues tested high-dose intramuscular thiamine in Parkinson’s. The results were remarkable.
The first published case report came in BMJ Case Reports in August 2013. Three Parkinson’s patients treated with 100 mg of intramuscular thiamine twice weekly. After fifteen days, their scores on the Unified Parkinson’s Disease Rating Scale, the standard clinical measure of Parkinson’s severity in which higher scores reflect more severe symptoms, improved by between thirty-one and seventy-one per cent. The hypothesis Costantini proposed in the paper sets out the clinical model directly.
Parkinson’s disease, in his model, involves a thiamine metabolism dysfunction that creates a focal severe deficiency in the brain regions most affected by the disease, despite normal blood thiamine. The dysfunction can be reversed by high doses of thiamine, which increase diffusion transport into the affected tissues. The model is identical in structure to what Lonsdale had been describing for dysautonomia: functional deficiency in a specific tissue, invisible to standard testing, responsive to high-dose treatment.
A second paper followed in 2015. This was an open-label pilot study in fifty Parkinson’s patients, followed for up to two and a third years. The mean UPDRS score fell from 38.5 to 18.2 within three months and remained stable across the follow-up. Motor scores improved from 22.0 to 9.9. Some patients with mild symptoms showed full clinical recovery. The authors concluded that thiamine appeared to have both restorative and neuroprotective effects in Parkinson’s. This was an open-label study, which means both patients and researchers knew what was being given, and the placebo effect could not be excluded for the early period. But the stability of the improvement over years is not easily explained by placebo.
Independent confirmation from the United States came from Luong and Nguyen in 2012, who reported in the Journal of Neurological Research on five male Parkinson’s patients aged sixty-five to eighty-two, treated with 100 to 200 mg of intramuscular thiamine daily. All five improved within days. In three of them, the conventional Parkinson’s medication carbidopa/levodopa was discontinued for ten to fourteen days without worsening of motor symptoms, suggesting that the thiamine was doing therapeutic work the medication was not.
In 2018, Costantini and Fancellu published a careful paper on dosing and overdose. They observed that doses had to be individually titrated. Patients with longer disease duration, higher symptom severity, and higher body weight needed more thiamine. Patients of Anglo-Saxon and African origin required smaller doses than Italian patients, which suggested underlying metabolic variation in thiamine handling. Overdose presented as initial improvement followed by symptom worsening, with discomfort, restlessness and migraine; stopping for a week resolved these and treatment could be resumed at half the dose. The correct individual dose was the one that produced symptom improvement without side effects.
By Costantini’s death in 2020, his team had treated more than 2,500 Parkinson’s patients with high-dose thiamine. The work continues through Sergio Pièche and the High Dose Thiamine Foundation, which publishes safety reviews, supports ongoing research, and provides resources in multiple languages. Daphne Bryan’s book on the protocol has become the guiding reference for patients and clinicians.
Patient testimonies, posted publicly on the foundation’s website, describe the kind of life-restoring recovery that the orthomolecular tradition has been documenting for sixty years. One Parkinson’s patient, whose name is Daphne, posted that the protocol had given her back the ability to play the piano. The reader of Part 2 will recognise the echo. Kaufman’s elderly arthritis patient in Bridgeport in 1949 had regained the same skill on the same instrument from the same family of vitamins.
Crucially, Costantini was clear from the beginning that high-dose thiamine was not a replacement for conventional Parkinson’s medication. It was an adjunct, given in addition to whatever the patient’s neurologist had prescribed. Patients on levodopa continued at the same doses. The protocol was designed to work alongside conventional treatment, not against it.
One of the effects Costantini documented was that thiamine appeared to prevent the development of the dyskinesias that often complicate long-term levodopa therapy. This is the kind of careful clinical observation that protects the protocol against the easy criticism that it asks patients to abandon prescribed medication.
The B-Complex as a System
The Italian protocol does not give thiamine alone. Costantini’s team added small doses of all the other B vitamins twice weekly when patients received the high-dose thiamine. The reason is biochemical. High-dose thiamine increases the body’s demand for other B vitamins, which share metabolic pathways and act as cofactors for each other’s activation.
Adequate levels of all the B vitamins are required for the system to function. Riboflavin is needed to convert B6 to its active form. B6 is needed for the conversion of tryptophan to niacin. B12 and folate work together in the methylation cycle that underlies much of cellular metabolism. The B vitamins are a system. They cannot be sensibly treated as isolated nutrients.
There is one further point worth holding in mind for everything that follows in this series. The form of each B vitamin is as important as the dose. The B-complex that supports thiamine activation is the methylated bioactive forms — methylfolate rather than synthetic folic acid, methylcobalamin rather than cyanocobalamin, P5P rather than pyridoxine hydrochloride.
The reasons for these distinctions, and the evidence around them, are the subject of Parts 4 and 5 of this series. For now, the point is that the supplements which support a high-dose thiamine protocol are not the cheap synthetic forms used in mass-produced multivitamins and fortified foods. They are the active forms that the body can use directly, in the absence of the methylation enzymes that the synthetic forms require.
This point is worth holding in mind because it is where Medicine Girl’s critique of B vitamins, which is otherwise mostly wrong, comes closest to touching the truth. Some of the safety concerns that have been documented in the mainstream literature about high-dose B vitamin supplementation involve synthetic folic acid and cyanocobalamin specifically, not the methylated forms. The responsible end of the orthomolecular tradition has been distinguishing these for decades. Parts 4 and 5 will engage that material directly.
What This Means for Medicine Girl’s Series
In her series, Medicine Girl argues that thiamine has been over-promoted by the supplement industry and that the historical and clinical case for its therapeutic use is thin. We addressed the historical case in Part 1 of this series. The clinical case sits in the work this article has described: Lonsdale and the functional-deficiency tradition, Marrs and Overton and the peer-reviewed literature, Costantini and the independent Italian lineage, the convergence of two separate clinical traditions on the same intervention in different disease contexts.
The clinical tradition is real, well-documented, and continuing. It has been refused engagement by the medical mainstream for sixty years, in the same way the niacin tradition was refused engagement, and for the same reasons. Functional deficiency does not fit the textbook picture. Gram-dose therapeutic supplementation does not fit the pharmaceutical model. Conditions that respond to vitamin therapy are not patentable, and the institutions that benefit from patents have not invested in the trials that would settle the question.
The orthomolecular and functional-medicine traditions have done what they could, with what they had, for as long as they have been allowed to practise. The medical mainstream has had sixty years to engage with the evidence and has not. Medicine Girl’s series asks the reader to dismiss this entire body of work as supplement-industry overpromotion. The work itself, and the clinicians who built it, deserve better.
The fourth article in this series turns from thiamine to folic acid. The story shifts from suppressed clinical traditions to a different problem entirely: a mass nutrient intervention that the mainstream did adopt, that has produced documented harms, and that some clinicians have been arguing against since the 1990s.
Coming up in Part 4: Folate, folic acid, and the methylation cycle. Why synthetic folic acid is not the same as natural folate. What the mass-fortification programmes have produced. The methylfolate alternative and the clinicians who have been arguing for it since the 1990s.
References
On Derrick Lonsdale’s life and clinical work:
Lonsdale D. A Nutritional Interpretation of Disease. Bloomington, IN: AuthorHouse, 2011.
Lonsdale D. Why I Continue to Recommend Thiamine at Age 100. Hormones Matter, June 2024.
https://www.hormonesmatter.com/
Marrs C. In Memoriam: Derrick Lonsdale, 1924-2024. Hormones Matter, June 2024.
On the functional deficiency tradition and modern thiamine clinical practice:
Marrs C, Lonsdale D. Thiamine Deficiency Disease, Dysautonomia, and High Calorie Malnutrition. Amsterdam: Elsevier/Academic Press, 2017. (The standard modern textbook of functional thiamine deficiency.)
Marrs C, Lonsdale D. Hiding in Plain Sight: Modern Thiamine Deficiency. Cells 2021; 10(10): 2595. https://pubmed.ncbi.nlm.nih.gov/34685573/
Lonsdale D. A review of the biochemistry, metabolism and clinical benefits of thiamine(e) and its derivatives. Evidence-Based Complementary and Alternative Medicine 2006; 3(1): 49-59. https://pubmed.ncbi.nlm.nih.gov/16550223/
Overton E. Thiamine Deficiency: The Great Imitator. EONutrition, 2020.
https://www.eonutrition.co.uk/
Overton E. Why Thiamine Supplementation Requires Magnesium. Hormones Matter, July 2022.
https://www.hormonesmatter.com/
On the forms of thiamine:
Bettendorff L, Wins P. Thiamin diphosphate in biological chemistry: new aspects of thiamin metabolism, especially triphosphate derivatives acting other than as cofactors. FEBS Journal 2009; 276(11): 2917-2925. https://pubmed.ncbi.nlm.nih.gov/19490098/
Lonsdale D. Thiamine tetrahydrofurfuryl disulfide: a little known therapeutic agent. Medical Science Monitor 2004; 10(9): RA199-203. https://pubmed.ncbi.nlm.nih.gov/15328496/
Volvert ML, Seyen S, Piette M, et al. Benfotiamine, a synthetic S-acyl thiamine derivative, has different mechanisms of action and a different pharmacological profile than lipid-soluble thiamine disulfide derivatives. BMC Pharmacology 2008; 8: 10. https://pubmed.ncbi.nlm.nih.gov/18549472/
On the B-complex as a system:
Kennedy DO. B Vitamins and the Brain: Mechanisms, Dose and Efficacy — A Review. Nutrients 2016; 8(2): 68. https://pubmed.ncbi.nlm.nih.gov/26828517/
On magnesium as a thiamine cofactor:
Oyanagi K, Hashimoto T. Magnesium in Parkinson’s disease: an update in clinical and basic aspects. In: Magnesium in the Central Nervous System. Adelaide: University of Adelaide Press, 2011. https://www.ncbi.nlm.nih.gov/books/NBK507254/
On Antonio Costantini and the B1 Protocol for Parkinson’s disease:
Costantini A, Pala MI, Compagnoni L, Colangeli M. High-dose thiamine as initial treatment for Parkinson’s disease. BMJ Case Reports 2013; 2013: bcr2013009289. https://europepmc.org/article/MED/23986125
Costantini A, Pala MI, Grossi E, et al. Long-term treatment with high-dose thiamine in Parkinson disease: an open-label pilot study. Journal of Alternative and Complementary Medicine 2015; 21(12): 740-747. https://pubmed.ncbi.nlm.nih.gov/26505466/
Costantini A, Fancellu R. An open-label pilot study with high-dose thiamine in Parkinson’s disease. Neural Regeneration Research 2016; 11(3): 406-407. https://pubmed.ncbi.nlm.nih.gov/27127474/
Costantini A, Fancellu R. Thiamine and Parkinson’s disease: dose, safety, and overdose. Gerontology and Geriatric Studies 2018; 4(1): GGS.000583. https://crimsonpublishers.com/ggs/fulltext/GGS.000583.php
Luong KV, Nguyen LT. The beneficial role of thiamine in Parkinson disease. CNS Neuroscience and Therapeutics 2013; 19(7): 461-468. https://pubmed.ncbi.nlm.nih.gov/23462281/
On the High Dose Thiamine Foundation and continuation of the protocol:
The B1 Protocol. High Dose Thiamine Foundation.
Bryan D. Parkinson’s and the B1 Therapy. Independently published, 2019.
Further reading:
Lonsdale D. Thiamine and magnesium deficiencies: keys to disease. Medical Hypotheses 2015; 84(2): 129-134. https://pubmed.ncbi.nlm.nih.gov/25542071/
Marrs C. Chandler Marrs on the Hormones Matter Platform. https://www.hormonesmatter.com/author/chandlermarrs/
Overton E. EONutrition Educational Resources on Thiamine.
https://www.eonutrition.co.uk/
source clivedecarle.substack.com