When insulin falls, the kidney dumps sodium. That is the ‘keto flu’.

Insulin isn't only a glucose hormone. At the kidney it's a sodium-retaining peptide — it tells the tubule to hold salt. It increases epithelial sodium-channel activity in the distal nephron, it stimulates the basolateral sodium–potassium pump that every segment leans on, and in the proximal tubule it supports NHE3, the sodium–hydrogen exchanger that reclaims the bulk of filtered sodium. Ralph DeFronzo, Cooke, Andres, Faloona and Davis, Journal of Clinical Investigation 1975, infused insulin into healthy men under a euglycaemic clamp and watched urinary sodium, potassium, calcium and phosphate fall. Forty years earlier, Atchley, Loeb, Richards, Benedict and Driscoll withdrew insulin from diabetic patients on a metabolic ward and documented a brisk natriuresis that reversed when the hormone returned. Those papers are still the ones I'd start you with. Carbohydrate restriction — ketogenic, carnivore, a very-low-calorie reset of the DiRECT sort — drops insulin. The kidney lets sodium go. Water follows. Magnesium and potassium hitch a ride. The headache, the cramp, the light-headedness that internet culture filed as keto flu is, in large part, that cation loss plus a brain still waiting on MCT and BDH1 induction. Toxins leaving the body isn't a mechanism we can assay. Natriuresis is.

In short. Insulin tells the kidney to keep salt. When carbohydrate and insulin fall, salt and water leave in the urine. That's the so-called keto flu.

A pair of human kidneys filters about 180 litres of plasma a day and, with it, something like 25 moles of sodium — roughly 575 grams of the ion, or a kilogram and a half of table salt if you insist on weighing chloride too. Almost all of it's taken back. Fractional excretion of sodium in a healthy adult on a mixed diet sits around 1 percent or less. The work is segmented: proximal tubule two-thirds, thick ascending limb another quarter, distal convoluted tubule and collecting duct the remainder, with the last few percent under hormonal control so that intake and output can match across a ten-fold range of dietary salt. Guyton put the long-term blood-pressure set-point on that match. Lifton's monogenic hypertensions — Liddle, Gordon, Gitelman, Bartter — are lesions of the same transporters. The so-called keto flu isn't a new organ. It's a change in the hormonal inputs to a machine that was already running, and the machine's job, before anyone invented a diet tribe, is to keep extracellular sodium near 140 millimolar and extracellular volume compatible with a filling pressure the heart can use.

In short. Your kidneys filter a huge amount of salt each day and take almost all of it back. Hormones decide the last few percent. That last slice is where a diet shows up.

This page is the nephron, the peptide hormones that write on it, and the first fortnight of a hard carbohydrate cut, told at the length you'd want before treating a comments-thread stack as physiology. Insulin, angiotensin II, atrial natriuretic peptide and arginine vasopressin are peptides. Aldosterone is a steroid that happens to sit on the same distal cells. GLP-1, from the incretin family, is natriuretic at the proximal tubule; SGLT2 inhibitors dump glucose and some sodium for a different reason; retatrutide occupies GLP-1, GIP and glucagon receptors and will move insulin, appetite and therefore the kidney's sodium ledger whether or not anyone wrote electrolytes in the protocol. NAD+ is the hydride coin the proximal-tubule mitochondria spend to keep Na+/K+-ATPase running. Sharing a shelf isn't sharing a mechanism. Neighbouring pieces take ketones as fuel and signal, carnivore as ketosis plus elimination, blood sugar as a control system, liver fat as the type-2 variable, and the triple agonist as occupancy. This page is salt, potassium, magnesium and water, named, with the set-point change written down so that a salt shaker can be a correction rather than a personality test.

In short. This page is how the kidney handles salt when insulin falls, and why potassium, magnesium and water then have to be renegotiated. It's physiology, not a supplement plan.

What follows is the physiology rather than a protocol. Named transporters, named receptors, named papers, named numbers, and a legal class stated once at the close for the catalogue objects that sit next door. People with chronic kidney disease, heart failure, mineralocorticoid antagonists, ACE inhibitors, angiotensin-receptor blockers or a clinician-imposed salt limit already have an appointment; this page isn't that appointment. For the rest, the first week of a well-formulated low-carbohydrate diet is a natriuretic experiment that Phinney and Volek have been trying to get people to salt on purpose for twenty years. The kidney changed its set-point. The shaker is the correction, and it looks undignified, and it works. Random capsules from a comments thread are how you spend money without naming a transporter. Serum electrolytes, a blood-pressure cuff and a waist tape are how you tell natriuresis from a mystery. Order of operations is most of the game, and week one isn't a test of character.

In short. People with kidney or heart disease already have a clinician. For everyone else, the first week of cutting carbohydrate is a salt-and-water shift, and adding salt is the usual fix.

DeFronzo's 1975 clamp remains the cleanest human demonstration. Insulin was infused; glucose was held; urinary sodium fell. The antinatriuresis wasn't a side-effect of falling glucose, because glucose didn't fall. Potassium and phosphate fell with the sodium, which is why a first insulin dose in a depleted diabetic can still drop plasma potassium hard enough to be a ward event. Saudek, Boulter, Knopp and Arky, in the mid-1970s, had already watched sodium retention accompany insulin treatment of diabetes. Tiwari, Riazi and Ecelbarger later mapped the tubular targets in the American Journal of Physiology: NHE3 proximally, the Na+/K+-ATPase along the nephron, ENaC distally, with serum- and glucocorticoid-regulated kinase 1 as a shared downstream node. Brands and Manhiani spent a decade arguing that the sodium-retaining effect is real even when the hypertension story is messier than a brochure would like. The clinical object is simpler than the blood-pressure argument. When insulin is high, the kidney holds sodium. When insulin is low, it doesn't — and that's the switch you're moving. Carbohydrate is the dietary input that moves insulin most. That's why a bread cut is a diuretic, and why a fat-and-protein plate isn't the same diuretic even at the same calories.

In short. In clamp studies, insulin itself makes the kidney keep salt, even when blood sugar is held steady. Cut the insulin signal and the kidney stops holding.

The distal target is ENaC, the amiloride-sensitive epithelial sodium channel on the apical membrane of principal cells in the late distal convoluted tubule and collecting duct. Three subunits, αβγ, a degenerin/ENaC family member, cloned by Canessa, Rossier and colleagues in the early 1990s. Liddle syndrome is a gain-of-function in which Nedd4-2 can no longer ubiquitylate the channel and pull it off the membrane; the patient holds sodium, dumps potassium, and runs a high blood pressure with a suppressed renin and aldosterone. Insulin's distal effect uses some of the same furniture. Occupancy at the insulin receptor, a receptor tyrosine kinase, recruits PI3K. PIP3 rises. PDK1 and Akt fire. SGK1 is transcribed and activated. SGK1 phosphorylates Nedd4-2, Nedd4-2 lets go of ENaC, and the channel stays at the apical membrane longer. Aldosterone writes on the same SGK1–Nedd4-2–ENaC sentence from a nuclear receptor. Two hormones, one channel, a shared kinase — the same furniture we already named. That's why a low-insulin diet and a high-aldosterone compensation can fight over the same principal cell, and why potassium then becomes a negotiation rather than a footnote.

In short. A sodium channel in the last part of the tubule is where insulin and aldosterone both act. Insulin keeps that channel on the cell surface. Less insulin means less channel.

Proximally the target is NHE3, SLC9A3, the apical sodium–hydrogen exchanger that reclaims most filtered sodium and, with carbonic anhydrase, most filtered bicarbonate. Insulin increases NHE3 activity. So does angiotensin II at AT1. Atrial natriuretic peptide and, relevantly for a later heading, GLP-1, decrease it. The proximal tubule isn't a hormone-deaf bulk handler. It's the segment that decides how much sodium is delivered downstream, and downstream is where ENaC, NCC and the potassium channels live. Inhibit SGLT2 and you also inhibit a slice of proximal sodium uptake, because SGLT2 cotransports glucose with sodium; the natriuresis of the gliflozins starts here, then is partly offset by tubuloglomerular feedback and by downstream compensation. The point for a diet essay is cruder. Falling insulin takes a brake off natriuresis at more than one segment. The proximal tubule lets more sodium through. The collecting duct, unless aldosterone has already arrived, lets more of that extra load leave. Water follows the sodium. Extracellular volume contracts by a litre or two. The scale applauds. The inner ear and the muscles don't always agree.

In short. Further up the tubule, insulin also helps a sodium–hydrogen exchanger take salt back. When insulin falls, more salt is delivered downstream and more of it's lost.

The pump underneath both stories is Na+/K+-ATPase, the basolateral enzyme Skou described, three sodium out, two potassium in, one ATP, the reason a proximal-tubule cell is a mitochondrion with a brush border attached. Insulin stimulates the pump. Catecholamines do. Thyroid status does. A cell that can't mint ATP can't reabsorb sodium, which is why ischaemic acute tubular injury is, among other things, a pump failure, and why the kidney's resting oxygen consumption is out of proportion to its mass. Peter Rich's whole-body ATP turnover, forty to sixty kilograms a day, has a renal share that textbooks put near a tenth of resting oxygen use, concentrated in cortex. NAD+ is the hydride coin Complex I wants oxidised so that the chain can pump the protons ATP synthase spends. The neighbouring cofactor essay is that topology. It belongs in a sodium essay because the nephron's day job is ATP-gated sodium transport, not because a 1000 milligram cake of β-NAD+ is a salt substitute. Two jobs. Same campus.

In short. Every sodium-taking step leans on a pump that spends ATP. Kidney cells are packed with mitochondria for that reason. Energy failure is salt-handling failure.

Carbohydrate is the dietary insulin driver that matters here. A glucose load, particularly a rapidly absorbed one, raises insulin. Protein raises it less, via incretins and amino-acid sensing at the β-cell, and fat barely does. That's why a ketogenic macronutrient ratio and a carnivore plate can both drop insulin, and why they aren't the same experiment: carnivore is also an elimination trial and a protein load, and a protein load still feeds hepatic gluconeogenesis and a residual insulin. People who stall in ketosis on steak often have enough insulin left to blunt natriuresis and ketogenesis together. HMGCS2 is competing with a busy TCA cycle; ENaC is still seeing some of the peptide it likes. A well-formulated ketogenic diet, in Phinney's usage, is low enough in carbohydrate that insulin stays down, high enough in fat that energy doesn't collapse, and salted on purpose because the kidney's set-point has moved. The salt isn't a cheat. It's the correction for a hormone that's no longer writing 'hold' on the collecting duct you still have.

In short. Sugar and starch raise insulin the most, protein less, fat hardly at all. A low-carbohydrate plate therefore changes the kidney's salt orders, and that's why salt is added on purpose.

Let's name the segments, because nicknames skip them. Bowman's capsule is the filter: fenestrated endothelium, glomerular basement membrane, podocyte slit diaphragm, a size- and charge-selective barrier that lets water, sodium, glucose, amino acids and urea through and keeps albumin mostly in. The proximal tubule reabsorbs about 65 percent of filtered sodium, almost all the glucose (SGLT2 in S1/S2, SGLT1 in S3), amino acids, bicarbonate and phosphate, and a large share of calcium and magnesium, isosmotically, so water follows through aquaporin-1. The descending thin limb lets water out into a hypertonic medulla. The thin ascending limb is impermeable to water and begins to dump salt. The thick ascending limb, NKCC2, ROMK, CLC-Kb, the target of loop diuretics, reabsorbs another 15 to 25 percent of sodium and, because the transport is electrogenic, drags magnesium and calcium paracellularly through claudin-16 and claudin-19. Destroy that segment and you get Bartter physiology: salt wasting, hypokalaemia, hypomagnesaemia, a concentrating defect. Furosemide is a reversible Bartter. A hard natriuresis of any cause will borrow some of that look, which is why we name the segment.

In short. Different parts of the tubule take back different shares of the filtered salt. The loop of Henle also pulls magnesium and calcium along. Break that segment and those minerals fall too.

The distal convoluted tubule is NCC, the sodium–chloride cotransporter, the thiazide target, a WNK–SPAK phosphorylated machine that also sets how much sodium is left for ENaC and how much magnesium TRPM6 can save. Gordon syndrome, PHA type II, is WNK or cullin-3 mischief that overactivates NCC: hypertension, hyperkalaemia, the opposite of Gitelman. Gitelman is loss of NCC, a lifelong thiazide: salt craving, low magnesium, low potassium, low blood pressure. The collecting duct then splits the remaining load between principal cells (ENaC, ROMK, aquaporin-2 under vasopressin) and intercalated cells (acid–base, some potassium recovery via H+/K+-ATPase, chloride via pendrin). Aldosterone writes on both. Vasopressin writes on the principal cell's water channel. Atrial natriuretic peptide writes on inner-medullary collecting-duct sodium transport and on the vasculature. The late nephron is where a diet, a diuretic, a peptide hormone and a steroid can still change the day's balance by a few grams of sodium, which is the difference between a headache and a working Tuesday for you.

In short. The last parts of the tubule are under hormonal control. That's where a diet, a water-retaining hormone and aldosterone still decide a few grams of salt a day.

Magnesium's renal life is mostly not ENaC. About 10 to 25 percent of filtered magnesium is reabsorbed proximally, 50 to 70 percent in the thick ascending limb via the paracellular route that NKCC2's lumen-positive potential drives, and the last 5 to 10 percent in the distal convoluted tubule through TRPM6, an apical channel whose abundance is set by EGF, oestrogen, and magnesium status itself. Loop diuretics waste magnesium because they collapse the TAL voltage. Thiazides waste it because they hyperpolarise the DCT cell and because chronic NCC shutdown remodels the segment. Insulin-withdrawal natriuresis isn't a loop diuretic, but volume contraction plus secondary aldosteronism plus a higher distal sodium load will still move magnesium, and a Western intracellular magnesium that was already marginal will unmask as a cramp. That's why the triad in the molecular stub is sodium, potassium and magnesium, not sodium alone. If you salt and ignore the other two, you've named one transporter and left two cations to chance.

In short. Most magnesium is reabsorbed in the loop of Henle, not at the insulin channel. A salt-losing fortnight still drags magnesium out, which is why cramp is part of the picture.

Potassium is the cation aldosterone will spend to save sodium. Giebisch spent a career on the distal potassium secretory path: ROMK on principal cells, BK channels when flow is high, a lumen-negative potential that ENaC's sodium entry writes, and an aldosterone-driven increase in both ENaC and the basolateral pump that refills the cell with potassium so ROMK has something to dump. McDonough and Youn have more recently put the muscle and the kidney in one potassium homoeostasis, with the skeletal-muscle Na+/K+-ATPase as a buffer that insulin itself drives — which is the other DeFronzo observation, the phosphate and potassium falling in the clamp as potassium enters cells. Two insulin effects, opposite directions, different timescales: acute insulin moves potassium into muscle; chronic insulin tells the kidney to hold sodium and, indirectly, to hold the volume that keeps aldosterone quieter. Drop insulin hard and you get both a transcellular potassium shift back out of cells and, over days, an aldosterone-driven renal potassium leak. Plasma potassium can look 'fine' on a morning blood test while total-body potassium isn't. That's a measurement problem, not a reassurance.

In short. Insulin pushes potassium into muscle in minutes. Over days, low insulin and high aldosterone make the kidney leak potassium. A single blood test can miss the second story.

The ATP bill is why a mitochondria diagram belongs on a salt page. Proximal-tubule epithelium is among the most mitochondrion-dense tissues a histologist will show you. The basolateral membrane is folded to hold the pumps. Oxygen extraction is high. A brief hypotension that a muscle would shrug off will kill proximal cells, dump tubular proteins, and present as acute kidney injury. Complex I wants oxidised NAD+. SIRT3 deacylates the neighbourhood if the cofactor is there to spend. MOTS-c is a 16-mer a mitochondrion translated from 12S rRNA, AMPK-adjacent, a different invoice on the same campus. None of that's a licence to treat a lyophilised NAD+ cake as a treatment for keto flu. It's a licence to remember that natriuresis is work the organelle was doing, and that a diet which changes sodium transport has changed an ATP customer. Isolated-tubule oxygen consumption, ouabain-sensitive, is how a renal physiologist would show you the invoice. A homogenate NAD+ kit isn't that experiment. The neighbouring mitochondria essay owns the cristae. This paragraph is the reason they're in the picture.

In short. Kidney tubules burn a lot of fuel to pump salt. That's why they're full of mitochondria. Changing salt handling changes that energy bill.

Every January a cohort of people cut bread, lose two to five kilograms in a week, and either feel reborn or feel wrecked. Both groups lost glycogen-bound water and a pile of sodium. Liver glycogen is on the order of 100 grams in a fed adult, muscle glycogen several hundred more depending on training; each gram of glycogen binds something like three grams of water, a number the older literature keeps using because it's the right size. Empty the glycogen and that water leaves. The sodium is the other half, and it's the half a salt-free 'clean eating' week makes worse. Insulin down, ENaC quieter, NHE3 quieter, a few hundred millimoles of sodium in the urine that used to be in the extracellular space. Water follows osmoles. Haematocrit ticks up. Blood pressure often falls, which is a feature if you were holding too much volume and a bug if you stand up and the room goes grey. The reborn group usually salted their food. The wrecked group usually didn't, and then bought a detox tea. The kidney was the organ. Insulin was the signal. Twitter wasn't invited.

In short. The first kilograms off on a low-carbohydrate week are mostly glycogen, water and salt, not fat. People who salt the food usually feel better. People who don't often feel wrecked.

Ketone anions add a third natriuretic force on top of the insulin withdrawal. β-Hydroxybutyrate and acetoacetate are strong-ish acids whose conjugate bases, once filtered, obligate a cation if they aren't fully reabsorbed. Early ketosis, before the kidney has upregulated the relevant transporters and before the brain and muscle are taking the ketones as fuel, dumps BHB and AcAc with sodium or potassium attached. Cahill's starvation studies already had this in the electrolyte tables. A well-adapted ketogenic liver is making ketones as a fuel; a week-one ketogenic liver is also making an anion the urine has to pair. That pairing isn't a toxin leaving. It's charge balance. The neighbouring ketones essay takes BHB as HDAC inhibitor, NLRP3 dampener and GPR109A ligand. This paragraph takes the same metabolite as a filtered anion with a sodium partner. Fuel and ligand and osmole are allowed to be the same carbon skeleton. The cell didn't pick a lane. We shouldn't either, but we should name which job we're measuring.

In short. Ketone bodies in urine take a positive ion with them to balance charge. Early on, that ion is often sodium or potassium. That's chemistry, not a cleanse.

The other half of feeling awful is the brain. Monocarboxylate transporters MCT1 and MCT2, and the mitochondrial BDH1 that converts BHB back to acetoacetate so SCOT can make it acetyl-CoA, take days to up-regulate in neurons. Cahill showed that a fully starved human brain will run mostly on ketones; he didn't show that it does so on Tuesday of week one. Glucose is falling because glycogen is gone and because insulin is low and glucagon is letting the liver make ketones rather than hold glycogen. The MCT lag is a transporter problem. The sodium lag is a volume problem. Together they produce headache, fog, irritability, and the internet's favourite attribution error. Salt the food, wait for the transporters, then decide whether you like the diet. Judging a ketogenic or carnivore pattern on day three, unsalted, is like judging endurance training on the first hill. The physiology hasn't finished installing the enzymes. The comments thread has often already written the review.

In short. The brain also needs a few days to turn on the machinery that burns ketones. Headache in week one is often salt loss plus that lag, not proof the diet is working or failing.

Carnivore and keto both raise the same natriuretic flag, and they're still not the same diet. Carnivore is an elimination plus a protein load; keto is a macronutrient ratio. Both can drop insulin hard enough to diurese. Carnivore's protein can blunt ketosis and leave a little more insulin on the table, which can mean a little less natriuresis and a little more gluconeogenesis. Keto's fat can raise ketones faster and the anion-paired cation loss with them. Fibre-free carnivore also collapses some of the colonic potassium and magnesium handling that plants were contributing, which is a small term and not the main story, and is mentioned so that a steak-only week isn't compared with a green-leaf ketogenic week as if the cation intakes were equal. Organ meat changes the micronutrient table again. The neighbouring carnivore essay owns the elimination and the microbiome. This page is the nephron. When someone on either pattern reports day-three misery, the first question is sodium intake, the second is potassium and magnesium, and the third is whether they're drinking a gallon of water because a comments thread said so. Toxins aren't on our list.

In short. Carnivore and keto both drop insulin and can dump salt. They still differ in protein, ketones and what else is on the plate. Misery on day three is still usually salt first.

A very-low-calorie reset of the DiRECT magnitude is the same natriuretic physics with a smaller energy term. Taylor's twin-cycle work, and the Counterpoint and DiRECT trials, emptied liver fat and returned first-phase insulin in a large fraction of people in the first years of type 2. The formula soup is low in carbohydrate and in energy; insulin falls; natriuresis follows; the early kilograms are water and glycogen again. Clinics that run these programmes already salt, already watch blood pressure, already reduce antihypertensives because the set-point moved. That last clause is the serious version of the shaker. A person on an ACE inhibitor plus a thiazide plus a sudden carbohydrate cut can diurese into a creatinine bump and a potassium problem. The diet isn't the villain. The stacked natriuresis is. I'll keep saying clinician whenever the stack includes a drug that already writes on the nephron. For an otherwise healthy adult who cut bread, the stack is insulin withdrawal plus optional overdrinking, and the correction is salt, then potassium, then magnesium, then patience for MCT induction. Order of operations. Most of the game.

In short. Very-low-calorie diabetes resets dump salt for the same insulin reason. Clinics already watch blood pressure and water pills. Stacked salt loss is the risk, not the soup itself.

Volume contraction is a renin stimulus. Juxtaglomerular cells notice a lower afferent-arteriolar pressure, a lower sodium chloride at the macula densa, and a sympathetic discharge, and they release renin. Renin cleaves angiotensinogen to angiotensin I; angiotensin-converting enzyme makes angiotensin II; angiotensin II constricts, thirsts, and tells the adrenal zona glomerulosa to make aldosterone. Angiotensin II is an octapeptide. Aldosterone is a steroid. Together they're the axis that tries to save the sodium insulin just stopped saving. Secondary hyperaldosteronism of a low-carbohydrate fortnight isn't a disease. It's the compensation. The cost is potassium. Aldosterone increases ENaC, increases the basolateral Na+/K+-ATPase, increases ROMK, and the lumen-negative potential that sodium entry writes drives potassium out. Palmer and Clegg have been writing this sentence for a decade in the electrolyte-and-diabetes literature: the symptoms people blame on carbohydrate withdrawal often track potassium and magnesium, not a moral failure to 'push through'. RAAS will try. It won't save you from a salt-free steak week. It will spend potassium attempting to.

In short. When volume falls, the renin–angiotensin–aldosterone system tries to save sodium. Aldosterone does that in part by dumping potassium. The compensation has a bill.

Angiotensin II itself is a proximal and distal sodium-retaining peptide, AT1-coupled, Gq in many of its vascular beds, with NHE3 among the tubular effectors. It's also a thirst hormone and a vasopressin-releasing hormone. The first days of a hard cut therefore stack: insulin down (natriuretic), angiotensin II up (antinatriuretic, but starting from behind), aldosterone up (antinatriuretic at ENaC, kaliuretic), vasopressin up if plasma osmolality ticks or if volume receptors fire. The net urinary sodium is still a loss until intake rises to the new set-point, because the insulin term was large and because ketone anions are still leaving. The net urinary potassium is often a loss even after sodium stabilises, because aldosterone stays up while insulin stays down. That's the renegotiation in the dek. It isn't a mystery. It's an axis doing the job Guyton assigned it, against a dietary input the axis didn't evolve to see every January — and we can measure that. Measuring plasma renin and aldosterone on day five of an unsalted ketogenic week would be a legitimate assay. Almost nobody does it. They buy a 'keto electrolyte' sachet whose label can't name ENaC.

In short. Several hormones move at once when you cut carbohydrate: insulin down, angiotensin and aldosterone up. Sodium still falls until you eat more of it. Potassium often keeps falling.

Potassium from meat is non-trivial. A 250 gram steak might carry a gram of potassium, sometimes more; muscle is a potassium tissue. It isn't always enough, particularly if aldosterone is high, if the person is also training, if they're drinking past thirst, or if they came in depleted. Avocado and leafy greens, on a ketogenic pattern that allows them, are the usual plant potassium. A carnivore plate that's steak-only and unsalted is a potassium-and-sodium double experiment. Potassium salt (potassium chloride, 'lite salt') is how some well-formulated clinics close the gap, with the same caveat the sodium heading will carry: not if you've kidney disease, not if you're on a potassium-sparing diuretic, not if an ACE inhibitor or an angiotensin-receptor blocker is already holding potassium up. Those people have a clinician and a potassium that can rise. This page is for the rest, whose potassium is more likely to be quietly leaving with the aldosterone. Plasma potassium below about 3.5 millimolar is the laboratory flag; cramp, weakness, extra systoles and a U wave on an ECG are the bedside flags. Extra systoles are how a diet week ends up in accident and emergency. Name the ion.

In short. Meat contains potassium, but not always enough when aldosterone is high. People on certain blood-pressure drugs must not add potassium themselves. Everyone else should know the ion by name.

Hypokalaemia isn't a mood. It's a resting-membrane-potential shift. Skeletal muscle becomes less excitable, then frankly weak; smooth muscle in the gut slows; cardiac myocytes pick up afterdepolarisations; the ECG flattens T waves, then grows U waves, then risks arrhythmia. A training session on day four of an unsalted, high-aldosterone week is an experiment on that membrane. So is a sauna. So is a loop-diuretic leftover from an old prescription. The correction, in an otherwise healthy adult, is potassium in food and, if the pattern allows, potassium salt, plus the sodium that will let aldosterone stand down. Giving potassium while still salt-deplete is pushing against a hormone that will keep wasting it. That's why the usual order is sodium first. The set-point is a sodium set-point. Potassium is the currency aldosterone spends to defend it. Replete the sodium, and the steroid often quietens, and the potassium leak slows. Replete only the potassium, and you're topping up a bucket whose tap is still open. McDonough would want the muscle buffer in the same paragraph: insulin down means potassium may have left cells as well. A banana protocol copied from endurance sport isn't a tubular diagnosis.

In short. Low potassium changes how muscle and heart cells fire. Adding sodium so aldosterone can fall often helps more than chasing potassium alone. Food first, and know your drugs.

Secondary hyperaldosteronism also holds a magnesium term. Distal magnesium recovery is fragile, TRPM6 isn't ENaC, and a volume-contracted, aldosterone-high, distal-sodium-loaded nephron will leak magnesium for reasons the TAL and DCT heading already named. Palmer's reviews keep pairing the two cations in diabetes and in diet. A cramp that doesn't answer to potassium sometimes answers to magnesium. A twitch, a poor night, a palpitating evening, the same. Serum magnesium is a poor intracellular readout; the laboratory's normal range isn't a muscle biopsy. Red-cell magnesium and ionised magnesium exist as research measurements and rarely as a GP order. Empiric repletion with a well-absorbed salt — glycinate, citrate, malate — is how most low-carbohydrate clinics close the gap, because the absorption of magnesium oxide is a bad joke and because waiting for a perfect assay is how cramp becomes a brand. Dose is hundreds of milligrams of elemental magnesium, not a 30 milligram marketing tablet you bought because the label was loud. Diarrhoea is the ceiling for citrate. The kidney disease caveat is the same caveat: reduced GFR holds magnesium, and this paragraph isn't for that person.

In short. Magnesium often falls with the same diuresis. Blood tests miss a lot of it. A well-absorbed magnesium salt is the usual practical step, unless the kidneys already can't excrete it.

Arginine vasopressin, nine residues, Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Arg-Gly-amide with a disulphide, is the posterior-pituitary peptide Verney characterised as the antidiuretic hormone. Osmoreceptors in the organum vasculosum of the lamina terminalis notice a 1 percent rise in plasma osmolality; hypovolaemia is a second, less sensitive stimulus via baroreceptors. Occupancy at V2, a Gs-coupled class-A GPCR on principal cells, raises cAMP, protein kinase A phosphorylates aquaporin-2, and the water channel inserts into the apical membrane. Water follows the medullary gradient that NKCC2 and urea recycling wrote. Agre's aquaporins, Nobel 2003, are the molecules; Nielsen and Knepper's physiology is the insertion story. A natriuresis that raises plasma osmolality will recruit this peptide. A natriuresis plus a gallon of water, drunk because a comments thread treated hydration as a virtue, will recruit it too late, or will overwhelm it, and plasma sodium will fall. Hyponatraemia is how a salt-losing week plus compulsive water becomes a seizure risk. Verbalis has spent a career on that. Thirst is a sensor. A gallon is a protocol. They aren't the same object, and you can tell them apart.

In short. The water-retaining hormone inserts water channels in the kidney when blood gets concentrated or volume falls. Drinking far past thirst during a salt-losing week can drop plasma sodium.

The practical failure mode is specific and common, and you've probably seen it. Person cuts carbohydrate, diureses, reads that they must drink three to four litres, drinks it, doesn't salt, feels worse, drinks more. Free water in, sodium out, vasopressin sometimes on because volume is down, so the water is kept. Plasma sodium slides from 140 toward 130. Headache that was going to answer to broth now has a hyponatraemic component. Confusion in a worse case. The correction isn't more water. The correction is sodium, and water drunk to thirst rather than to a round number. Athletes already know exercise-associated hyponatraemia as a water-and-ADH problem; a ketogenic week is a quieter, indoor version. Diabetes insipidus, lithium, and V2 antagonists are the other aquaporin stories, and they don't belong in a diet paragraph except as a reminder that water handling is a receptor occupancy, not a wellness metric. Urine colour as a hydration assay is a crude osmolality, and in a high-ADH, low-sodium state a dark urine can coexist with too much free water on board. Measure plasma sodium if the headache isn't answering. Don't guess with another litre.

In short. A common mistake is dumping salt in urine while forcing litres of water. Thirst is the better guide. If the headache won't lift, plasma sodium is the measurement.

Magnesium hitch-hiking deserves its own mechanisms even after the TAL paragraph, because the symptoms are what people actually bring. Intracellular magnesium is a cofactor for hundreds of enzymes, including the Na+/K+-ATPase whose ATP bill we already wrote, and a large fraction of adults in high-income countries already sit at the low end of intake. Diuresis unmasks it. Loop-like wasting, DCT wasting, a diet that dropped pulses and greens, training, and a proton-pump inhibitor in the background — those stack. The cramp is the popular output. The twitch, the poor sleep, the palpitations, the extra-systolic evening, the migraine that a magnesium trial sometimes moves, are the others. Those symptoms aren't specific. Hypomagnesaemia doesn't own them. Ignoring magnesium because the serum came back 0.82 millimolar, 'normal', is how you miss a total-body deficit. Repletion is glycinate or citrate, split, with food if the gut complains, for days to weeks, not a single heroic gram that then purges itself. Oxide is cheap and poorly absorbed. Threonate has a brain story that this kidney essay won't promote. Sulphate is Epsom salt, a laxative with a magnesium label.

In short. Many people are already low in magnesium; a diuretic fortnight makes that obvious as cramp or palpitations. A well-absorbed salt for days beats a single huge dose.

Calcium is the quieter fourth cation, and it belongs in a sentence so that parathyroid hormone doesn't get forgotten. Insulin's clamp effect included a fall in urinary calcium. Volume contraction and a rising filtered load later can swing the other way. Ketogenic diets, especially in children with epilepsy, have a documented lithogenicity: uric acid stones, calcium stones, hypocitraturia, a reason paediatric ketogenic teams alkalinise and watch the urine. Adults on carnivore or keto who arrive with a stone history need more than a salt shaker; they need citrate and a urology opinion. Bone, in the long-run argument, is a different literature, mixed, confounded by weight loss and by protein. This fortnight-scale page won't settle it. Phosphate falls with insulin, rises with insulin withdrawal, and is a ward problem in refeeding, which is the opposite experiment. Let's name the four cations so that a sodium story doesn't pretend to be the whole periodic table. Then return to sodium, because sodium is the set-point the others are following.

In short. Calcium and phosphate also move, and ketogenic diets can raise stone risk in some people. Sodium remains the set-point the other minerals are following in week one.

Acid–base is the other quiet ledger. Early ketosis is a mild, compensated metabolic acidosis: the anion gap ticks up by the ketones, bicarbonate falls a little, ventilation adjusts, the kidney makes more ammonia to excrete the acid with ammonium rather than with sodium, and over days the compensation is good enough that arterial pH in a well-formulated adult isn't a drama. Diabetic ketoacidosis is a different beast — insulin deficiency, usually type 1, much higher ketones, volume collapse, a sick patient, potassium that looks high in plasma while the body is empty of it. Conflating physiological ketosis with DKA is how a diet gets a skull sticker it hasn't earned, and how a type-1 crisis gets treated as a lifestyle. SGLT2 inhibitors, pregnancy, and starvation in a thin person sit closer to the dangerous edge. This page's audience is the otherwise healthy adult whose week-one misery is salt. If there's vomiting, Kussmaul breathing, confusion, or a history of type 1, the page you want is a hospital, not a shaker. Palmer and Clegg's starvation-ketosis paper is the nephrology version of that distinction. Hold it.

In short. Mild ketone acidosis in a well-fed low-carbohydrate week isn't diabetic ketoacidosis. Vomiting, deep breathing, confusion, or type-1 diabetes is hospital territory, not a salt problem.

Sleep, cortisol and glucose sit next door, and they belong in one paragraph here because a wretched first week is often a stacked insult. Spiegel, Leproult and Van Cauter already showed that a short night is a next-day insulin-resistance experiment. Cortisol, up at the wrong hour, holds sodium via mineralocorticoid-receptor occupancy that the 11β-HSD2 enzyme is supposed to prevent in the distal nephron; when the enzyme is outrun, cortisol is a second aldosterone. A person who cuts bread, sleeps four hours, drinks a gallon, and doesn't salt has run four experiments at once. The neighbouring sleep essay owns the clamp and the dawn phenomenon. This paragraph owns the reason a bad night plus a natriuresis feels like influenza. Slow-wave sleep is also when the largest growth-hormone pulses fire; miss it and the repair night is thinner. None of that's a reason to abandon a diet that might, later, empty a fatty liver. It's a reason to salt, to sleep, and to stop treating misery as evidence of virtue. Virtue isn't a transporter, and we shouldn't treat it as one.

In short. A short night already worsens next-day insulin handling. Stack that on salt loss and forced water and the first week feels like flu. Sleep and salt are both part of the fix.

Guyton's long-term blood-pressure control is a renal-sodium story. The kidney adjusts natriuresis until intake and output match; the matching happens at a particular filling pressure, and that pressure is the set-point. Raise the set-point — more angiotensin II, more aldosterone, more insulin, a scarred kidney — and blood pressure sits higher so that excretion can keep up with the salt you eat. Lower the set-point — less insulin, an SGLT2 inhibitor, a GLP-1 agonist's natriuretic term, a thiazide — and blood pressure sits lower, or you diurese until it does. A well-formulated low-carbohydrate diet lowers the insulin term. The set-point falls. If you keep eating a low-salt, high-water, high-virtue plate, output exceeds intake until extracellular volume is down enough to make you miserable, or until RAAS has spent a lot of potassium trying to catch up. Adding sodium raises intake to the new matching point without demanding that volume collapse. That's the whole trick. It looks undignified because public-health messaging spent fifty years telling healthy people to put the shaker down. The messaging was aimed at a different set-point.

In short. The kidney matches salt in to salt out at a certain blood volume. Low insulin lowers that target. Eating more salt meets the new target without wringing you dry.

Phinney and Volek's well-formulated ketogenic diet is the document we'll keep citing for the practical numbers, because they wrote them as physiology rather than as a sachet. Sodium in the grams, not a pinch: the usual transition range of 4 to 6 grams of sodium, roughly 10 to 15 grams of salt, is the amount a lot of people need to feel human again in the first fortnight, unless a clinician already has them restricted. Potassium in the 3 to 4 gram neighbourhood from food and, if needed, potassium salt. Magnesium in the 300 to 400 milligram elemental range, as a salt the gut will actually absorb. Those are transition amounts, not a forever prescription, and not advice for anyone on a renal or cardiac salt limit. After adaptation, many people still want more sodium than a mixed-diet guideline, because insulin is still low and the set-point is still lower. The shaker stays on the table. Broth is a delivery vehicle. Electrolyte powders are a delivery vehicle with a markup. The kidney doesn't read brands. It reads millimoles.

In short. A well-built low-carbohydrate diet adds several grams of sodium on purpose at the start, plus potassium from food and a proper magnesium salt. Those are transition amounts, not a fashion.

Early weight loss is mostly not fat, and pretending it's has sold a lot of programmes. Glycogen and its water, the sodium and its water, a bit of gut content if fibre collapsed: that's the two to five kilograms of week one. Fat comes off slower and quieter, as an energy deficit, as a liver emptying, as a waist tape. Trust the tape more than the scale in week one. The scale is putting on a show. A well-formulated pattern that salts will often show a smaller week-one drop because the person kept a litre of extracellular water they actually needed. That's a feature. A comments thread will call it a stall. A comments thread isn't a nephron. DiRECT's remission tracked weight loss at a different timescale, around 15 kilograms, with hepatic and pancreatic fat as the interesting mass. Retatrutide's Phase 2 curve, Jastreboff, New England Journal of Medicine 2023, emptied the same depots from the receptor side over months, not over a salty weekend. Different instruments. Week-one natriuresis isn't the twin cycle. It's the entrance fee some people pay, unnecessarily heavily, because they wouldn't pick up the shaker you left on the table.

In short. The first week's weight drop is mostly water and glycogen. Fat loss is slower. Salting can make the scale move less, which is often a good sign, not a stall.

Blood pressure is the other readout, and it's why this page keeps saying clinician. A hypertensive adult on two agents who cuts carbohydrate hard can, within days, be too dry: rising creatinine, postural dizziness, a potassium that goes the wrong way depending on whether the second agent is a thiazide or an ACE inhibitor. Deprescribing of antihypertensives under supervision is a known feature of low-carbohydrate clinics and of bariatric wards. An unsupervised internet cut is how that feature becomes an acute kidney injury. Conversely, a young adult with a baseline systolic of 110, a high training load, and no salt can faint in a warm room. The correction is still sodium, not a personality seminar about grit. Home blood-pressure cuffs are cheap. Standing and lying pressures are a volume exam. A sitting pressure of 95/60 plus a headache is data. So is a pressure that was 150/95 and is now 118/76 on the same two pills. Bring both to a person who can change the pills. Don't crowd-source the titration; that's a clinician's job. We can name that.

In short. Blood pressure often falls when insulin falls. That can be welcome, or it can overshoot in people already on water pills. Measure it. Don't adjust prescription drugs from a comments thread.

Salt hunger is a real behaviour, and it's allowed to be one. Hyponatraemic animals seek sodium; so do humans, with a lag and with culture sitting on the signal. A week-one craving for cheese, broth, olives, or actual salt is often the set-point speaking. Suppressing it with a virtue story is how you get the cramp. The opposite error is using the set-point as a licence to eat processed food by the pallet because 'keto needs salt'. The millimoles can come from a shaker on food you already chose. They don't require a new tribe of bacon-branded electrolytes. Chloride comes with the sodium in table salt; some of the volume story is chloride, and some hypochloraemic alkalosis on a hard cut is the kidney making bicarbonate while it dumps acid as ammonium. A little unglamorous chemistry. The sentence remains: the kidney changed its set-point. The shaker is the correction. After that, eat the diet you can actually defend to a lipid clinic, a liver scan, and a night's sleep. Salt doesn't excuse a surplus, and it doesn't empty a fatty liver by itself.

In short. Wanting salt in week one is often the body asking for the new target. Add it to food you already chose. It isn't a liver treatment.

GLP-1 is a natriuretic peptide, which is easy to forget when the headlines are weight and glucose. Muskiet, Tonneijck, Smits and colleagues have reviewed the renal physiology: occupancy at GLP-1R inhibits NHE3 in the proximal tubule, there's an afferent arteriolar term that can raise glomerular filtration rate in healthy kidneys, and urinary sodium rises. The licensed GLP-1 receptor agonists carry that pharmacology whether or not the indication was salt. GIP's renal file is thinner. Glucagon has a proximal natriuretic history of its own, older than the triple-agonist decade, and a uraemic-solute story that a liver-fat essay won't settle. Retatrutide, LY3437943, occupies GLP-1R, GIPR and GCGR as a unimolecular fatty-acylated chain. Coskun, Cell Metabolism 2018, is the engineering; Jastreboff, NEJM 2023, is the Phase 2 weight curve. Organism-level insulin will fall as energy intake falls and as glucose falls; the kidney will notice that insulin term on top of any direct GLP-1 natriuresis. A research synthesis of the published backbone is a ligand for occupancy work, not a salt tablet. Neighbourhood, not identity — sharing a shelf isn't sharing a mechanism. We can name that.

In short. The gut hormone GLP-1 also makes the kidney dump some salt. A three-receptor chain that includes GLP-1 will move insulin and salt handling as weight and glucose move. That isn't a salt pill.

SGLT2 inhibitors are the pharmacological cartoon of this entire page, and they're a licensed medicine, which this catalogue isn't. Block the proximal sodium–glucose cotransporter, glycosuria follows, natriuresis follows, volume falls, blood pressure falls a little, and EMPA-REG OUTCOME, Zinman, NEJM 2015, then a stack of heart-failure and chronic-kidney-disease trials, showed outcome benefits that are still being unpacked as haemodynamic, as ketogenic, as tubuloglomerular. The early eGFR dip is the afferent arteriole noticing distal sodium delivery. The euglycaemic DKA risk is the ketone-and-insulin-and-volume edge in people who already have an insulin problem. A well-formulated ketogenic diet isn't an SGLT2 inhibitor. It shares a natriuretic first week and a rise in ketones and a fall in insulin, and stacking the diet on the drug is how a few patients have arrived acidotic. That sentence is for the clinic, loudly. For a paper, the gliflozins are the existence proof that proximal sodium-and-glucose handling is a blood-pressure and volume lever you can pull on purpose. The diet pulls a different lever, insulin, and arrives at a similar volume term. Two keys. One extracellular space.

In short. Drugs that block a kidney glucose-and-salt transporter also cause a salt-and-water loss. They are medicines. Stacking them on a hard carbohydrate cut is a clinic problem, not a comments-thread trick.

ACE inhibitors, angiotensin-receptor blockers, mineralocorticoid-receptor antagonists and potassium-sparing diuretics are the other stop. They already write on the axis this diet just recruited. An ARB plus a salt-free ketogenic week can raise potassium while volume falls. Spironolactone plus potassium salt from a sachet is a textbook hyperkalaemia. A thiazide plus the diet is a stacked DCT natriuresis: magnesium and potassium both leave. Loop diuretics plus the diet is a stacked TAL insult. SGLT2 plus the diet is the paragraph above. None of these is a reason to stamp a diet 'dangerous' and walk away. All of them are a reason titration belongs to a person who can order a metabolic panel. A journal page isn't that person. Creatinine, sodium, potassium, magnesium if you can get it, lying and standing blood pressure, and a medication list: that's the week-one panel for anyone already on a nephron drug. For everyone else, broth and a cuff will do more than a shopping basket of capsules.

In short. Blood-pressure and water pills already act on the same kidney hormones this diet moves. Mixing them without a clinician can push potassium or kidney numbers the wrong way.

Atrial natriuretic peptide and brain natriuretic peptide are the peptides the heart writes when filling pressures are high, and they're the physiological opposite of the keto-flu problem. Occupancy at GC-A raises cGMP, the inner-medullary collecting duct dumps sodium, the vasculature opens, and volume falls. A heart-failure patient lives on that axis and on the drugs that mimic or antagonise it. Carbohydrate restriction in that person isn't this page. It's a specialist conversation about fluid, about SGLT2, about whether an energy deficit will help the myocardium or wreck the preload. ANP belongs on the peptide map so that insulin isn't the only sodium-retaining peptide a reader can name, and so that 'the body wants to dump salt' is recognised as a receptor occupancy the heart can write as well as a receptor occupancy the β-cell can withdraw. Different ligands. Same extracellular volume. A bench assay of ANP is a radioimmunoassay or an ELISA on the right sample; a bench assay of insulin at ENaC is a Ussing chamber or a split-open tubule. Those machines still exist. Hashtags don't replace them, and we shouldn't let them.

In short. The heart also writes a salt-dumping peptide when it's overfilled. That's a different hormone from insulin. Heart-failure diets are specialist ground, not week-one keto advice.

What a research vial is doing in a diet page is a fair question, and the answer is neighbourhood, not a stack. We synthesise the published retatrutide backbone in the United States and put HPLC-MS on the certificate because the triple-agonist occupancy is a metabolic tool a paper can name. We stock lyophilised β-NAD+ because proximal-tubule mitochondria spend that cofactor on the pump this page keeps invoking. We don't sell a salt. We don't sell a potassium tablet. We don't write a ketogenic protocol as a reconstitution. The incretin essay, the NAD+ essay, the liver-fat essay and the blood-sugar essay are the other rooms. A reading list can sit them together because fuel routing is the subject. A shopping basket that mixes a triple agonist, a cofactor, and a sachet of citrate as 'electrolytes plus peptides' hasn't named a receptor, a transporter, or a millimole. The shaker remains the correction for the insulin term. The ligands remain ligands. Confusing those floors is how a journal becomes a stack, and how a research reagent becomes a medical claim it isn't allowed to be.

In short. A research peptide that moves insulin will also move the kidney's salt handling. That doesn't make it a salt replacement. The shaker and the ligand are different jobs.

Order of operations, written as physiology rather than as a shopping list. First, sodium, in grams, in food, unless a clinician has already closed that door: the transition range above, broth if cooking is hard, a cuff so you can see the pressure. Second, water to thirst, not to a gallon; plasma sodium if the headache won't lift. Third, potassium from meat and from whatever plant the pattern allows, potassium salt only if the medication list is clean. Fourth, magnesium as a well-absorbed salt for days, not as oxide. Fifth, patience for MCT and BDH1 in brain, which is days, and for aldosterone to stand down once volume is no longer collapsing. Sixth, the diet you actually meant to run: ketosis as a fuel and a signal, carnivore as an elimination, a DiRECT-scale deficit as a liver-emptying, an incretin occupancy as a receptor-side tool. Those extra tools won't feel like themselves if the first four are still wrong. Judging a macronutrient ratio from an unsalted Tuesday is how good physiology gets a bad review. The neighbouring ketones essay can wait a week. The nephron won't.

In short. Salt first, water to thirst, then potassium, then magnesium, then wait for the brain's ketone machinery. Only after that's it fair to judge the diet you meant to run.

Assays that would actually decide, if you want the numbers. Serum sodium, potassium, chloride, bicarbonate, urea, creatinine, and magnesium, lying and standing blood pressure, weight, and a note of what was eaten and drunk for three days: that's a week-one panel, not a personality. Urine sodium on a spot sample is crude but will show whether the kidney is still dumping; a fractional excretion needs a paired creatinine. Plasma osmolality, or at least a calculated one from sodium, glucose and urea, will tell you whether the water story matches the salt story. Ketones, BHB on a meter, will tell you whether the anion term is on. An ECG if there are palpitations, because U waves and extra systoles are cheaper to catch than to miss. Renin and aldosterone, paired, if someone is going to claim the axis: they should be up in a volume-contracted unsalted week, and if they aren't you have a different diagnosis. None of this is required to pick up a shaker. All of it's how you stop a diet essay becoming a mood when the misery isn't answering.

In short. If you want measurements: blood salts, kidney numbers, lying and standing blood pressure, and what was eaten and drunk. Palpitations earn a heart tracing. Most people still just need the shaker.

Who this page isn't for, restated without drama. Reduced GFR. Heart failure. Cirrhosis with an aldosterone story of its own. Adrenal disease. Pregnancy. Type 1 diabetes, and type 2 on SGLT2 inhibitors or on insulin that's about to be stacked against a sudden carbohydrate collapse. ACE inhibitors, ARBs, spironolactone, eplerenone, amiloride, triamterene, loop and thiazide diuretics, unless a clinician is in the loop. Eating-disorder history, because a 'well-formulated' cut is a restriction that can be borrowed. Lithium, because water and sodium handling are already a clinic. These aren't fine print. They are other physiologies occupying the same transporters. A journal essay that names ENaC has a duty to name the people in whom ENaC is already a drug target. For everyone else, the duty is smaller and ruder: salt the food, stop drinking past thirst, replete the cations the axis is spending, and then decide, on the evidence of a waist and a night's sleep and a glucose curve, whether the diet is the tool you wanted. Toxins leaving the body won't be in the discussion section.

In short. Kidney disease, heart failure, pregnancy, type-1 diabetes, and several common pills take this page off the table. For others, salt the food, drink to thirst, and then judge the diet.

Let's retrace the argument without the catchphrase. Insulin is a sodium-retaining peptide at NHE3, the Na+/K+-ATPase and ENaC. Carbohydrate restriction drops that peptide. Natriuresis follows, water follows, and the first kilograms are glycogen and salt, not a miracle. Ketone anions pair a cation on the way out. Aldosterone then tries to save sodium and spends potassium doing it. Vasopressin tries to save water and will save too much of it if you drink past thirst. Magnesium hitch-hikes through the TAL and the DCT. The kidney's long-term set-point has moved, in the Guyton sense, and the shaker is how intake meets the new match without wringing the extracellular space dry. GLP-1 occupancy and SGLT2 blockade are pharmacological versions of a volume term; retatrutide sits upstream of insulin and appetite as a triple class-B occupancy; NAD+ is the hydride coin the pump spends. Neighbouring essays take ketones as signal, carnivore as elimination, blood sugar as a loop, liver fat as the type-2 variable, sleep as a glucose drug. This page was the nephron. Week one isn't a test of character. It's a fractional excretion.

In short. Insulin holds salt; dropping carbohydrate drops that hold. Other hormones then fight over potassium, magnesium and water. Adding salt meets the new target. That's the whole map.

The public papers are short enough to actually read. Atchley 1933, insulin withdrawal as a metabolic-ward natriuresis. DeFronzo 1975, the clamp that held glucose and still moved sodium. Tiwari, Riazi, Ecelbarger, the tubular map. Palmer and Clegg, starvation ketosis and the kidney, the electrolyte tables in a nephrology journal. Guyton 1991, the set-point. Rossier on ENaC, Lifton on the monogenic lesions that prove which transporters set blood pressure. Agre and Nielsen on aquaporin-2. Verney on the antidiuretic hormone. Muskiet on GLP-1 and the kidney. Zinman, EMPA-REG, on what happens when you pull a proximal sodium-and-glucose lever as a drug. Phinney and Volek on the well-formulated pattern that salts on purpose. Cahill on starvation fuel. Jastreboff 2023 if your question has crossed into triple occupancy. That's a fortnight of evenings, not a guru. The sachet aisle will still be there when you come back, and it will look smaller. A write-up that can't name ENaC, SGK1, ROMK, TRPM6 and V2 hasn't yet started. A social post that treats misery as toxins hasn't read Atchley.

In short. A short stack of named papers covers the insulin clamp, the kidney's salt target, the water channel, the gut-hormone effect and the practical salted diet. Read those before a sachet.

What you should leave with is a map, not a shopping list. Insulin holds sodium. Carbohydrate restriction is a natriuretic experiment. Aldosterone and vasopressin then renegotiate potassium, magnesium and water. Symptoms attributed to carbohydrate withdrawal often track that triad plus a lag in cerebral ketone uptake. A well-formulated low-carbohydrate diet adds sodium on purpose because the set-point moved. The shaker is the correction. Incretin occupancy and SGLT2 blockade move the same extracellular volume from other doors. The catalogue objects in this neighbourhood — a published triple-agonist backbone, a dinucleotide the pump's mitochondria spend — are ligands and cofactors for named assays. They aren't broth. People with a nephron already written on by drugs or disease have a clinician. Everyone else is allowed to know why day three felt dreadful, and to fix the millimoles before they abandon a diet that might, later, empty a liver. Order of operations is most of the game. Toxins leaving the body aren't in the assay list.

In short. Leave with the map: insulin holds salt, cutting carbohydrate dumps it, other hormones then move potassium, magnesium and water, and the salt shaker meets the new target.

Research-use-only. Not for human consumption / not a medicine. The physiology in the paragraphs above is public, cited, and older than the vial: Atchley, DeFronzo, Guyton, Palmer, Muskiet, the aquaporin papers, the well-formulated-diet numbers. Where a catalogue object sits next door — lyophilised β-NAD+ as the cofactor a proximal-tubule mitochondrion spends, a United-States-made published LY3437943 backbone as a triple class-B ligand — those solids are HPLC-characterised laboratory reagents, labelled for in-vitro work and for the animal protocols a laboratory already knows how to write. They aren't salt, not a ketogenic protocol, not a clinic infusion, and not a substitute for a metabolic panel in anyone already on a nephron drug. Use the physiology to design the experiment you have the controls for, with the transporter named, the millimoles written down, and the diet not confused with the occupancy. Read DeFronzo, read Palmer, read Guyton, then pick up the shaker or the pipette according to the question you actually have. We'll sell you the named ligands. We won't tell you they're broth, and we won't tell you that week-one misery was a toxin leaving the body. It was sodium. The kidney already knew.

In short. The biology is public. Neighbouring research chemicals aren't salt and not a diet. Week-one misery was sodium leaving, and the kidney already knew.