Why fasting causes electrolyte loss — and how thinly the prevention advice is sourced.
What the evidence says
- When insulin drops during fasting or carbohydrate restriction, the kidneys increase sodium excretion (natriuresis)[1],[2]
- The natriuresis is driven by the fall in insulin itself, not by lost calories or glucose per se — controlled clamp studies show insulin directly governs renal sodium handling independently of glucose and other hormones[2]
- Sodium loss is greatest in the first ~3 days of a fast, when the insulin drop is sharpest: in fasting studies urinary sodium exceeded intake by roughly 30–40 mmol/day over the first three days before beginning to taper[2]
- Potassium is lost alongside sodium — measured in `cna02`: under insulin, urinary potassium fell ”from 66±9 to 21±1 µeq/min” and sodium from 401±46 to 213±18 µeq/min, with blood glucose held at the fasting level by servocontrol and GFR and renal plasma flow unchanged, so the effect is insulin’s rather than glucose’s. *The mechanism previously stated here — ”via the renal sodium-potassium exchange mechanism” — is REMOVED, because this paper excludes it: ”There was no change in plasma aldosterone concentration, which was low throughout the studies”, and the sodium effect ”cannot be attributed to persistence of increased aldosterone secretion.”* Cite `cna02` for the what, never for that why[1],[3]
- DO NOT WRITE THIS CLAIM. Magnesium excretion increasing during low-insulin states rests on `cfa10` alone — a record whose directory holds no source file of any kind — and no source held here measures magnesium during fasting at all. `cna02` is not it: every one of the five ”magnesium” occurrences in its full text is a reference-list entry. A paper naming magnesium in its bibliography is not a paper that measured it. And the direction is disputed by a reading this library cannot yet check — see below[1]
- “Keto flu” and “fasting flu” symptoms (headache, fatigue, cramps, nausea, lightheadedness) are primarily electrolyte depletion, not an inherent effect of fasting — these symptoms are well-characterised clinical manifestations of hyponatremia (low sodium) — and the symptom-to-electrolyte link is itself unestablished: the same review reports the literature ”lacks studies that directly link reported symptoms during initiation with physiological measures such as electrolyte loss”. Both halves of this file’s story are under-evidenced, not only the supplementation half[1],[4],[5]
- There is a rationale for electrolyte supplementation and no clinical evidence that it works — a 2025 scoping review searching Embase, Medline and Web of Science concludes: ”There is a physiological rationale for supplementation with electrolytes and ketone esters, but a lack of clinical studies documenting their effect.” Its body is blunter: consumption ”is heavily encouraged… In the identified literature there is however little rigorous study and quantification of this requirement.” Write the rationale, never the prevention[1],[5]