Edit Log

Articles on this site are, as a rule, left as they were first published, and where their claims conflict with current evidence a separate review section is added below the article rather than the text being changed.

A small number of statements have been an exception, because they were not merely unsupported but wrong in a way that could lead a reader to harm. Those have been removed and replaced. Every such change is recorded here, with the original wording, so that nothing is altered silently. Earlier versions of every page also remain available through the Internet Archive.

Iron and Manganese

femn.html · first published 12 October 2001 · revised 22 August 2026

Removed

1. The article's opening sentence: “As is the case with most other associated mineral pairs, the absorption of iron is dependent on manganese, however with manganese levels being frequently lower than iron levels, and with many forms of iron causing constipation or gastric upsets when supplemented, the addition of manganese, when low, is vitally important.”
2. “In actual clinical settings, Dr. Ronald Roth had not seen a single case of a patient's iron (ferritin) levels decline as a result of taking manganese, even when doses as high as 150 mg per day were supplemented on an ongoing basis.”
3. “In fact, most minor iron-deficiency situations can be dealt with by using manganese alone - without any iron - which reduces any potential adverse effects that can be part of routine iron supplementation.”
4. “While some sources claim that manganese lowers iron levels, this is mostly a theoretical consideration that would only happen under unusual circumstances.” Together with the sentence that followed it: “This not only prevents further ratio conflicts between the two elements, but also substantially reduces the amount of iron needed when manganese is supplemented at the same time.”
5. “In contrast to Acu-Cell Analysis, routine blood tests do not include manganese measurements, but use a number of iron determinations alone, none of which are very reliable to assess true iron requirements in a patient, with only ferritin levels being more useful, and being closer to actual intracellular values.”
6. “As a result, patients are given far too many false positive and false negative recommendations (especially with polycythemia or thalassemia issues) to take - or not to take extra iron supplements, to the detriment of the patient.”
7. A clause within the paragraph on causes of low iron: “however low iron can very easily result from manganese levels having been low for a long period of time, or from” — the sentence now reads “and other factors such as excessive calcium, zinc, or magnesium levels can reduce iron values as well.”
8. “It has strong estrogenic properties, and as a result is the most important element when nutritionally treating menopausal symptoms, menstrual problems, osteoporosis, and postpartum depression, for which manganese, along with Vitamin B1, is most effective. Manipulating manganese (or DHEA) levels can be an effective way to delay, or (sometimes) hasten menopause, trigger the menstrual cycle again after starting menopause, or along with iron and iodine levels, affect the frequency, duration and heaviness of the menstrual cycle.”
9. “Like iron, manganese can be helpful with some types of asthma, where lung capacity measurably improvers proportional to manganese intake. Extra supplementation of manganese may be helpful in some cases of carpal tunnel syndrome, deafness, epilepsy, infertility, and lack of libido in both sexes. In addition, individuals who regularly dislocate joints (particularly knee joints), frequently present with insufficient manganese levels, so normalizing manganese in those cases will permanently resolve that problem.”

Reason

On the opening sentence. Iron and manganese do not assist one another across the intestinal wall; they compete for the same transporter, DMT1, and which one wins depends on iron status. The opening sentence read that competition as dependence, the reverse of the relationship, and built a supplement recommendation on it. Manganese does not aid iron absorption, and cannot correct an iron deficiency.

On the 150 mg figure. The Tolerable Upper Intake Level for manganese is 11 mg per day, and the EFSA safe level of intake is 8 mg per day. The highest oral doses in any published human trial are in the region of 30 to 40 mg of elemental manganese per day, given as manganese ascorbate in glucosamine and chondroitin osteoarthritis studies. A daily intake of 150 mg is roughly fourteen times the Upper Intake Level and about four times anything ever trialled. At that intake, competition with iron at the shared DMT1 transporter should have been plainly visible, and the reported absence of any effect is not consistent with the compound having been administered as described. Neither the product nor the dose was analytically verified, and the outcome was assessed by a method with no validated reference range.

On correcting iron with manganese. Iron deficiency is a depletion of iron stores. Only iron can restore them, whether from diet, oral supplementation or infusion. No quantity of a different element supplies iron atoms. Because iron and manganese compete for the same intestinal transporter, taking manganese while iron is low may in fact reduce absorption of what dietary iron is available. The advice was also directed at iron-deficient readers, who are precisely the group in whom DMT1 is upregulated and manganese absorption is therefore raised.

On the interaction being theoretical. The competition between the two elements at DMT1 is not an edge case; it is the principal determinant of how much of either is absorbed, and it has been demonstrated in people.

On the three passages about testing. The removed sentences presented Acu-Cell Analysis as the comparator against which routine blood work falls short, and stated that ferritin is unreliable and that it approximates intracellular values. Ferritin is in fact the recommended first-line test for iron status; its real limitation is that it rises with inflammation, infection, malignancy and liver disease, which is handled by interpreting it alongside transferrin saturation and soluble transferrin receptor — not by substituting an intracellular measure that has no validated reference range. The replacement text says this, and keeps the article's own warning about polycythemia and thalassemia.

On manganese as a cause of low iron. The removed clause reversed the established direction of the interaction. Low iron does not follow from long-standing low manganese; competition at the shared DMT1 transporter runs the other way, and iron deficiency raises manganese absorption rather than lowering it.

On blood sugar. Manganese supplementation improved glucose tolerance in mice on a high-fat diet, but that finding does not carry over cleanly to people: plasma manganese shows a U-shaped association with type 2 diabetes in humans, with risk raised at both low and high levels, so more manganese is not simply better for blood sugar.

On the estrogenic and treatment claims. Manganese is not estrogenic. It acts on the hypothalamus to stimulate luteinizing hormone secretion, an effect studied in animals as an endocrine-disruption hazard — manganese-induced precocious puberty — not as a therapy. Higher blood manganese is associated with lower bone mineral density, most pronounced in postmenopausal women: the opposite of what would help with osteoporosis, the exact group the passage recommended it to. No human evidence supports manipulating manganese levels to delay or hasten menopause, or to restart a menstrual cycle. None of asthma, carpal tunnel syndrome, deafness, epilepsy, infertility, low libido, or recurrent joint dislocation has controlled human trial evidence behind manganese supplementation as a treatment. Manganese is a genuine cofactor in several enzyme systems, but an enzyme that requires a cofactor is saturated at ordinary intake, and being a cofactor does not establish that more of it treats a condition in someone who isn't deficient.

Added in their place

Referenced text describing the DMT1 relationship in both directions, a table of the four iron/manganese status combinations and the conditions that produce each, a passage on the absence of any routine clinical test for manganese status, and an editorial recommendation on supplement dosing. The full reasoning and sources are in the evidence review on that page.

Three further passages were added to the article at the same time, none of which replaced anything that was there before:

• An editor's note arguing that the established position — that manganese deficiency is uncommon — should be read against the absence of any routine test. The claim is widely asserted but does not rest on population testing, because population testing does not exist. Until a validated status marker is in general clinical use, the defensible position is that the prevalence is unknown rather than low. This is an editorial position taken by this site, not a finding from the literature.
• A note on the products sold today, observing that manganese supplements are commonly sold at 10 mg per capsule and some at 50 mg — the first delivering the whole Tolerable Upper Intake Level in a single soluble dose, the second exceeding the highest amount given in any published human trial. The position that such doses are experimental at best is editorial, and the accompanying observation about overload symptoms is offered as an observation rather than as trial evidence.
• A passage on how people test this in practice, describing what people commonly do when no blood test for manganese status exists, and how the outcome is generally read. It is a description of common practice, not a recommendation to undertake it.

The replacement text is checked line by line against its own citations. The claim that sustained iron intake reduces manganese uptake is sourced to a cell-culture study, so it is described as what that study actually shows — restoring intracellular iron protects cells against manganese's toxic effects, rather than iron reducing how much manganese is absorbed. The table's “High Fe – Low Mn” row presents that direction as an expectation that follows from the DMT1 competition established elsewhere on the page, not as a separately documented finding. The claim that manganese absorption falls in the presence of dietary fibre, calcium and copper is narrowed to calcium alone, the only one of the three actually tested in the cited study. Heme iron is described as considerably more absorbable than plant iron, rather than with a specific multiple its citation does not give, and manganese's role is described in brain function only, matching what its citation covers, rather than brain and muscle function.

Bismuth and Lithium

bili.html · first published 12 October 2001 · revised 22 August 2026

Removed

1. “Magnesium can be used to treat lithium overdose, while calcium can be used to treat bismuth overdose.”
2. “In addition to treating patients with Manic-Depressive Illness, lithium has been used with some success for Ménière's disease, Huntington's Chorea, and alcoholism.”
3. “In animal models, lithium has been reportedto be beneficial for brain injury, spinal cord injury, stroke, Parkinson's disease, and ALS (amyotrophic lateral sclerosis), whereby recent clinical trials suggest hat lithium may stop the progression of ALS.” Quoted exactly as published, including the two typographical errors — “reportedto” and “suggest hat”.

Reason

On treating an overdose with a mineral. Lithium poisoning is a medical emergency. Management is intravenous fluid resuscitation and, in severe cases, haemodialysis, which the international EXTRIP workgroup recommends as the treatment of choice. Magnesium is not a treatment for it. Bismuth encephalopathy is managed by withdrawing the bismuth; calcium is not a treatment for it either, and chelation - the intervention that does remove bismuth - has been reported to increase urinary excretion while the patient deteriorated clinically. A reader who acted on the removed sentence would be taking a supplement instead of seeking emergency care, which is the specific harm this site's removal policy exists for.

On Huntington's chorea and alcoholism. Both were tested after the article was written, and both failed. A double-blind crossover trial of lithium carbonate, haloperidol, the two combined and placebo found no significant effect on chorea from any treatment, and a placebo-controlled trial agreed. In alcoholism, a Department of Veterans Affairs cooperative study randomised 457 men over 52 weeks and found no difference between lithium and placebo, in depressed and non-depressed drinkers alike. No adequate controlled evidence was identified for Ménière's disease in either direction.

On ALS. The statement was accurate about the state of knowledge when it was written: an Italian pilot study in 2008 did report that lithium slowed progression. It did not replicate. Four randomised controlled trials totalling 469 patients were subsequently pooled in a meta-analysis which found no benefit for survival, functional rating or forced vital capacity, and the phase 3 LiCALS trial found no survival advantage at 18 months. Because the sentence speaks in the present tense about an incurable disease, leaving it in place would tell ALS patients in 2026 something that stopped being true in 2013.

Added in their place

Referenced text on the actual management of lithium and bismuth poisoning, on the trial results for each of the proposed indications, on the ecological drinking-water literature and what it can and cannot show, and on the 2025 finding that lithium is depleted in the prefrontal cortex in Alzheimer's disease. The reference table was annotated to state which substance each figure refers to, after a reader wrote in asking whether the 400 mg in the Therapeutic Range row was elemental lithium. It is not - it is lithium carbonate, the prescription salt, containing about 75 mg of lithium.

Two entries in the reference table were also rewritten. Bipolar disorder was listed as a symptom of low lithium; it is not a lithium deficiency, and lithium's effectiveness as a prescription medicine for it does not make it one. Bismuth was given a list of deficiency symptoms although bismuth is not an essential element and no deficiency state has been established.

Two further passages were annotated rather than removed, because in each case the original wording could be answered directly from the literature. The article refers in several places to patients being lithium or bismuth deficient and to low lithium being common; no validated biochemical marker of lithium status exists, so those are readings against an assumed range rather than diagnoses. And the article states that lithium raises stomach acid in the lower stomach; the animal evidence points the other way, with lithium chloride reducing gastric acid output in both dogs and rats. No human trial was found in either direction, and the annotation says so.

A note was added beneath the synergist and antagonist rows to the effect that an antagonist is not an antidote. Calcium and magnesium remain listed as antagonists of bismuth and lithium, which is the biochemistry the removed overdose sentence was built on, and without that note a reader could reconstruct the removed advice from the table alone.

Two paragraphs were also added to the closing section on Helicobacter pylori, neither of which replaced existing text. The first states that bismuth's antimicrobial action against the organism is real and is the best-supported claim on the page, but that it raises eradication rates only when added to a multidrug regimen alongside a proton pump inhibitor and antibiotics — bismuth on its own is not a treatment for the infection. The second records that the article's statement that bismuth and lithium are “invariably always low” during active infection was not addressed by any study identified, that no evidence was found for either element indicating infection, gastric acidity or treatment response, and that bismuth is not an essential element, so there is no established deficiency state for a reading to indicate.

The replacement text is checked line by line against its own citations. The passage on the 2025 Alzheimer's finding cites that paper only for its human measurement — lithium was the one metal significantly reduced in post-mortem prefrontal cortex tissue in mild cognitive impairment and Alzheimer's disease — and not for a mouse experiment reported in the same paper, which does not appear on this page. The Department of Veterans Affairs cooperative study of 457 men over 52 weeks is cited to the paper that actually describes it, Dorus et al., JAMA 1989. The claim that chelating bismuth poisoning with DMPS has been reported to worsen the clinical picture is cited to the case report that documents it, alongside the general bismuth-toxicity review for the point that review does support: that bismuth toxicity is usually reversible once the bismuth is stopped.

The claim that no validated human biomarker of lithium status exists does not cite the study of lithium content across 160 beverages and a fruit-fly feeding experiment, which does not address human biomarkers. That source instead supports a fact in the intake section: mineral water is a markedly more concentrated dietary source of lithium than wine, beer, soft drinks, tea or coffee.

ALS / Lou Gehrig's Disease

dis-als.html · first published 12 August 2003 · revised 21 August 2026

Removed

1. “...whereby recent clinical trials suggested that lithium (which is used to treat Manic Depressive Illness), may stop the progression of ALS.”

Reason

The same statement appeared on the Bismuth and Lithium page and was removed for the same reason, but it matters more here, because this is the page a person newly diagnosed with a fatal disease is likely to reach.

The claim was reasonable when written. A pilot study published in 2008 did report that lithium delayed progression in ALS patients. It did not replicate. Four randomised controlled trials totalling 469 patients were subsequently pooled in a meta-analysis which found no benefit for survival, functional rating or forced vital capacity, and the phase 3 LiCALS trial found no survival advantage at 18 months. Leaving a present-tense sentence in place would tell ALS patients in 2026 something that stopped being true in 2013.

Added in its place

The 2008 pilot study is now cited explicitly, so that readers can see what the original claim rested on, followed by the trial results that overturned it and a note that lithium is not a treatment for ALS. The one question still open - a possible survival benefit confined to patients homozygous for the C-allele at rs12608932 in UNC13A - is described as the trial question it is, rather than as an available treatment.

Prostatitis and Benign Prostatic Hypertrophy

dis-pro.html · first published 19 October 2001 · revised 22 August 2026

Reversed

This article stated that Prostatitis is a condition of excess zinc, and built its treatment advice on that. The direction is the wrong way round, and every passage carrying it has been corrected. The central statements as they previously read:

1. “...however only an enlarged prostate (Benign Prostatic Hyperplasia), or BPH, is at times linked to below-normal levels of zinc.” The word “only” is gone: Prostatitis is consistently linked to below-normal zinc in prostatic fluid, and BPH's own zinc picture is unsettled — of three tissue-zinc studies checked directly, one found zinc 61% lower in BPH tissue, one found it higher, and one found no difference from normal, so the article states no direction for it.
2. “Subsequently, with some exceptions, nutritional supplements used in the treatment of BPH will generally not work for Prostatitis, or may have to be avoided.” It now reads that BPH supplements don't automatically apply to Prostatitis, and names the one that does: phytotherapy, which a Cochrane review of prostatitis treatments found produces a modest symptom improvement.
3. “Prostatitis presents with above-normal, to typically very high cellular levels of zinc, and symptoms are invariably worsened by either supplementing extra zinc, consuming foods that are high in zinc, or by any other remedial aspect or dietary lifestyle that synergistically promotes greater zinc retention.”
4. “Lowering cellular Zinc Levels through diet or nutritional supplementation provides the most effective, safest, and cheapest treatment approach for the great majority of Chronic Prostatitis sufferers, with the most common zinc antagonists consisting of Vitamin C, Calcium + Vit D, Selenium, Folic Acid, Vitamin A, Vitamin B1, and at times Iron - but rarely Copper, which should be individually matched.”
5. “Dietary treatments include the avoidance of foods high in zinc (shellfish, herring, organ meats, wheat germ, soybeans, and some seeds and nuts) ... so supplementing zinc antagonists may be a more convenient approach.”
6. “From personal, clinical experience, lowering zinc levels through dietary or supplemental intervention has proven to be the most effective method to bring Prostatitis quickly under control.”
7. “Prostatitis and Benign Prostatic Hypertrophy go hand in hand with abnormally high cellular levels of zinc (prostatitis), and often below-normal cellular levels of zinc (BPH).”

Removed

1. “Some individuals claim that their condition improved after supplementing zinc, but it always turns out that they either did not have Prostatitis in the first place - i.e. they self-diagnosed their condition, or they were misdiagnosed by an unqualified practitioner.”
2. “While zinc is pro-inflammatory, Copper is anti-inflammatory and also a zinc antagonist, so for some of those suffering from Prostatitis, a higher intake of copper-rich foods, or short-term copper supplementation can be helpful. Unfortunately, a considerable percentage of the population suffers from copper overload, which would therefore prohibit additional copper intake for most Prostatitis patients.”
3. “With a history of Prostatitis, zinc should be used with caution when treating an enlarged prostate, otherwise one may trade one set of symptoms (BPH) for another (Prostatitis).”

Reason

Zinc in expressed prostatic secretion is lower in men with chronic prostatitis than in controls, and the lower it falls the more severe the pain. The antibacterial activity of normal prostatic fluid, described in the older literature as the “prostatic antibacterial factor”, was identified as free zinc. On that evidence, prostatitis is associated with a shortage of prostatic zinc rather than an excess.

The recommendation to correct low zinc through diet or supplementation cites the randomised trial behind it directly (zinc sulfate, NIH category IIIA Chronic Prostatitis), with a caveat: oral zinc does not raise the specific fluid level found low in Prostatitis, so the benefit is more likely from zinc's antibacterial and immune-modulating action than from restoring that level directly.

That makes the advice actively harmful rather than merely unsupported. A man with prostatitis following the original text would have removed zinc-rich foods from his diet and taken zinc antagonists, lowering the substance responsible for his prostate's own antibacterial defence, while already low in it. The paragraph dismissing men who improved on zinc was removed for the same reason: those are precisely the men the evidence predicts would improve, and the paragraph existed only to explain them away.

The article's separate claim, that an enlarged prostate is linked to low zinc and benefits from supplementing it, was left standing. It is supported, and it is now referenced.

No animal or laboratory study was used to justify any of these changes. The corrections rest on human evidence only, cited in the article's reference list.

Also changed

The article stated that saw palmetto and phytosterols are routinely recommended for an enlarged prostate, treating the two as interchangeable. That sentence was left as written, since it accurately describes common practice, and the evidence for each was added after it: a Cochrane review found saw palmetto (Serenoa repens) produced no improvement in urinary symptoms or peak urine flow against placebo, including at double and triple doses, while a separate Cochrane review found beta-sitosterol did significantly improve urinary symptoms, though without reducing prostate size.

Zinc and Potassium

znk.html · first published 28 December 2001 · revised 22 August 2026

Removed

1. “Red Blood Cell Potassium, or Intracellular Potassium (Acu-Cell Analysis) are more accurate indicators of a patient's true potassium status.”
2. “In contrast to serum potassium, intracellular potassium levels are more diet-dependent and increase on a linear scale according to dietary intake, specific medical conditions, and the levels of interactive minerals such as sodium, magnesium, lithium, calcium, phosphorus, sulfur, chromium and others. Serum and cellular potassium levels coincide only when both are at a low to normal range... If too much potassium is ingested or retained, intracellular measurements readily pick up those higher levels in otherwise healthy individuals, where there would be no changes in serum potassium. These changes can be used for diagnostic, therapeutic, or preventive purposes.”
3. “...although these symptoms would rarely be experienced by patients who are truly potassium deficient - i.e. they exhibit very low cellular levels of potassium - but more so when relied on the limited accuracy of serum potassium measurements.”
4. “many patients' potassium levels appear normal when they actually suffer from hypokalemia (low serum potassium), while others are erroneously treated for hyperkalemia (high serum potassium), when their potassium is perfectly normal.”
5. “The same mixed results can be expected when using zinc for the relief of symptoms from herpes, AIDS, or other infections.”
6. “...those with a tendency for hypoglycemic (low blood sugar) episodes may end up worse as a result of that extra potassium reducing manganese and chromium levels, which help stabilize blood sugar, and partly as a result of potassium reducing sodium, which slows insulin response. In addition, lowering chromium too much as a result of long-term high potassium intake can contribute to osteoporosis of trabecular bone.”

Reversed

1. “Prostatitis is invariably found with cellular zinc levels that range from above-normal, to excessively high... At higher levels, zinc becomes pro-inflammatory, being a main reason why those who supplement extra zinc for prostatitis generally worsen their symptoms. Instead, any approach that lowers zinc will more likely resolve the condition.”
2. “Dr. Ulrich Werth reporting a 96.5% improvement with the symptoms of Parkinson's disease after permanently implanting titanium needles at strategic points in patients' ears.”
3. “Some less common forms of Multiple Sclerosis present with very low cellular zinc and/or potassium levels... symptoms readily improve when normalizing the patient's zinc and potassium status.”

Reason

On Acu-Cell Analysis and cellular measurement. Acu-Cell Analysis has published no data of its own. Zinc and potassium can be measured in erythrocytes, in tissue and in the whole body, by whole-body counting, atomic absorption spectrophotometry and X-ray fluorescence, and several of the studies now cited on the page do exactly that. What has never been established is a validated way of reading such a measurement as a clinical status. Reviewing the evidence for the potassium Dietary Reference Intakes, the National Academies were unable to identify any sensitive or specific biomarker of potassium status. Because the measurement techniques plainly exist and are in routine research use, a method that worked would have left a literature behind it. The claims that rested on it are removed rather than qualified, together with the paragraph that made the same claim without naming the method.

On erythrocyte potassium. It has been compared against whole-body potassium directly, and the answer depends on who is measured: it tracked closely in healthy subjects and not at all in untreated hypertensive patients. That is now stated in place of the claim that it is a more accurate indicator.

On the sentence about patients appearing normal. Hypokalemia is defined by low circulating potassium, so a patient cannot have a normal serum potassium and simultaneously “suffer from hypokalemia (low serum potassium)” unless the measurement was wrong, which is a different statement. The sentence ran together three distinct situations - normal serum potassium despite total-body depletion, pseudohyperkalemia from collection or processing, and plain laboratory error - and asserted that “many” patients are mistreated without any prevalence evidence. The preceding sentence on the page already covers preanalytical error accurately, and it stands.

On prostatitis and zinc. The direction was inverted. Chronic prostatitis is associated with lower zinc in prostatic fluid, not higher: in chronic bacterial prostatitis, prostatic fluid zinc averaged roughly a tenth of control values with no overlap between the groups, and a meta-analysis of fourteen case-control studies found prostatic fluid and seminal plasma zinc significantly lower than in controls. A randomised trial of 120 men with NIH-IIIA prostatitis found zinc sulfate 220 mg daily produced better total and pain scores than placebo at twelve weeks, which is the opposite of the article's advice to lower zinc. The replacement is deliberately not categorical either: the correlation with symptom scores is weak, one series found chronic prostatitis fluid zinc close to normal, and benign prostatic hyperplasia is a separate picture in which the tissue measurements disagree outright.

On the Parkinson's figure. The 96.5 % improvement could not be traced to any source. It appears neither in the published trial nor in Dr. Werth's own book. What the published randomised, sham-controlled study reports is that 42 patients were randomised and 32 completed, that the proportion improving by at least 20 % on the UPDRS at twelve months was identical in both arms at 52 %, that overall improvement was greater in the sham group, and that efficacy of the permanent needles was not established.

On multiple sclerosis. Zinc in MS depends entirely on which compartment is measured: meta-analyses find serum and plasma zinc lower than in controls, while some whole-blood and erythrocyte studies report levels several times higher, falling during relapses. No human clinical evidence was identified for an MS subgroup defined by low cellular potassium, or for symptoms improving once zinc and potassium are normalised.

On zinc for infections other than colds. The 2024 Cochrane review of 34 trials and 8,526 participants covers the common cold and upper respiratory infection. It found little or no effect on preventing colds, a possible shortening of an ongoing cold by about two days, and an increase in non-serious side effects, all on low-certainty evidence. It says nothing either way about herpes, AIDS or infections in general, so the sentence extending it to them was removed.

On potassium, chromium and insulin. The mechanism described runs against the published physiology, in which plasma potassium acts as a stimulus for glucose-induced insulin release rather than slowing it. No evidence was found that high potassium intake depletes chromium, or that it causes trabecular osteoporosis by that route. Chromium's own effect on weight is small and inconsistent - about half a kilogram against placebo in one meta-analysis, just over one kilogram in a Cochrane review that found no reliable basis for firm conclusions, and no significant effect on weight, BMI, waist circumference or fat mass in type 2 diabetes.

Added in their place

Referenced text on what the potassium tests can and cannot show, on the dose-response relationship between potassium and blood pressure and how much it differs between hypertensive and normotensive people, on when supplemental potassium is dangerous and food is not, on the zinc findings in chronic prostatitis and benign prostatic hyperplasia, on the chromium weight-loss trials, on the published Parkinson's implant trial, and on zinc in multiple sclerosis by compartment. The reference list grew from two entries to twenty-one.

Retained with an editorial note

Two passages record Dr. Roth's own clinical observations and are kept as written: the laterality associations between cellular potassium and right-sided conditions, cellular zinc and left-sided conditions, and the account of sciatica responding to zinc and potassium ratios with pain shifting from one side to the other. No controlled human research validating either as a diagnostic rule or a treatment was identified, and each now carries a note saying so immediately beneath it, along with the point that no validated reference range interprets a cellular mineral reading as a clinical status.

Corrections to any article can be raised through the contact page.