The commission requires these kept apart. They are kept apart here with nothing moved across the line to make either column look better.
G1. DOCUMENTED. A human was measured and this happened.
| Finding | Measured value | Exposure | Grade | Identifier |
|---|---|---|---|---|
| Hypernatremia to 187 mmol/L with rhabdomyolysis, oliguric AKI, respiratory failure on VV-ECMO and encephalopathy | Na 187 mmol/L, Cr 14.5 mg/dL, CK >33,000 U/L | Continuous dry fast, about 4 days | C | DOI 10.1093/ajrccm/aamag162.4785 |
| Severe hypernatremia mortality, Turkish ED cohort | 49.5% in-hospital mortality at Na >160 mmol/L, n = 256 | Hypernatremia of any cause | B+ | DOI 10.1007/s11739-015-1368-4 |
| Severe hypernatremia mortality, French review | 42 to 60% for acute onset above 160 mmol/L; overall often above 50% | Hypernatremia of any cause | B+ | Rev Med Suisse no. 3238, Berwert, Vogt, Burnier |
| New venous thromboembolism during Ramadan fasting | 15 of 311 (4.8%), Caprini 6.0 ± 2.6 vs 3.0 ± 2.2, p < 0.001 | IDF, 14.02 ± 0.08 h/day × 30 d | B+ (no control group) | DOI 10.7759/cureus.89406 |
| Euglycemic DKA on an SGLT2 inhibitor during a fast | pH 6.92, bicarbonate 4 mmol/L, anion gap 30 mEq/L, glucose only 243 mg/dL | IDF, 14 h/day × 30 d | C | DOI 10.1186/s13256-022-03347-1 |
| Hypohydration during a 19-day dry fast | 6 of 34 participants; 3 exceeded 830 mosm/kg; 4 below obligatory urine volume | IDF, 11 to 12 h/day × 19 d | B+ | Klatte 2023, Charité, handle fub188/37553 |
| Potassium fell, hematocrit fell, plasma osmolality fell during a 19-day dry fast | K p < 0.001, Hct p < 0.001, plasma osmolality p < 0.001 (lower), Na p = 0.129 (no change) | IDF, 11 to 12 h/day × 19 d | B+ | Klatte 2023, handle fub188/37553 |
| Dizziness and circulatory weakness rose during the fasting weeks | 0.31 → 0.46, p = 0.039, n = 91 | IDF, 11 to 12 h/day × 19 d | B+ | Kleimaier 2022, Charité, handle fub188/36381 |
| 24-hour urine volume collapsed in fasting stone formers | 604 ± 260 mL vs 1031 ± 542 mL, p = 0.001; citrate 1.40 vs 1.90, p = 0.034 | IDF, Ramadan | B+ (reliability caveats) | Rom J Mil Med 2021;CXXIV(4):525-532 |
| Renal colic ED visits rose in the first two weeks of Ramadan | RR 1.267 (95% CI 1.031 to 1.502) among Muslims, not among Jews | IDF, Ramadan | B+ | within DOI 10.5489/cuaj.6664 |
| Acute renal failure in CKD patients fasting Ramadan | 3 of 60 in week 1, 4 of 60 by month end; higher risk with baseline CrCl <60 | IDF, Ramadan | B+ | Saudi J Kidney Dis Transpl 2015;26(2):320-4 |
| Hypoglycaemia and DKA in adolescents with type 1 diabetes | Pooled 50.79 per 100 (30.36 to 71.10) and 11.02 per 100 (0.36 to 30.78), 458 patients | IDF, Ramadan | A | DOI 10.1186/s12902-025-01835-1 |
| Third-trimester birth weight in pregnancy | −60 g, p < 0.007; 2nd+3rd trimester −80 g, p = 0.03 | IDF, Ramadan | A | within DOI 10.1002/ijgo.16127 |
| Supratherapeutic INR after Ramadan on warfarin | RR 1.69 (95% CI 1.22 to 2.33), p = 0.001, 446 patients | IDF, Ramadan | A | DOI 10.1007/s00228-022-03281-7 |
| AKI from NSAID added to RAS inhibitor plus diuretic | 9.6 → 20.0 per 1,000 person-years, IRR 2.08 (1.58 to 2.74) | No fasting | B+ | DOI 10.1186/s40780-025-00485-8 |
| Median survival after voluntarily stopping eating and drinking | 7 days, n = 86 | Complete fluid abstention, end-of-life population | B+ | DOI 10.1370/afm.1814 |
| Fluid needed to stay euhydrated doing manual labour in WBGT >27.5°C | 5.44 L per 12-hour shift, n = 22 | No fasting | B+ | DOI 10.1186/1745-6673-3-21 |
| Hypohydration thresholds | Thermoregulation compromised >1%; heat illness risk up >3%; severe >10% | General | A (position statement) | DOI 10.4085/1062-6050-52.9.02 |
G2. THEORIZED. A mechanism is named but no human on a dry fast has been measured for it.
Every item in this column is labeled INFERRED or mechanism. None of them has a measured human dry-fasting number behind it, and none of them is written as if it did.
| Theorized hazard | The named mechanism | Why it stays theorized |
|---|---|---|
| Hemoconcentration driving thrombosis during a dry fast | Fluid loss raises hematocrit and blood viscosity, slowing venous return | The Cureus VTE study (DOI 10.7759/cureus.89406) names this mechanism and did not measure hematocrit, viscosity, coagulation markers or osmolality. The one dry-fast study that did measure hematocrit (Klatte 2023) found it significantly LOWER, p < 0.001. The mechanism is currently contradicted by the only direct measurement located. |
| Lithium toxicity from volume depletion during a dry fast | Volume depletion raises proximal tubular sodium and lithium reabsorption | The only human study (DOI 10.1186/s40345-025-00378-7, n = 250) found lithium unchanged or slightly lower, p = 0.052. No multi-day dry-fast lithium study exists in any language. |
| Triple-whammy AKI during a dry fast | NSAID blocks afferent vasodilation, RAS inhibitor blocks efferent constriction, no fluid intake removes volume reserve | The quantified cohort (IRR 2.08) studied euvolemic outpatients with no fasting. Nobody has measured the combination during a fast. |
| Diuretic-plus-dry-fast volume depletion | Drug removes output while intake is zero | Zero studies located in any language. |
| Rhabdomyolysis from a dry fast | Reduced muscle perfusion plus raised core temperature | One case report (CK >33,000 U/L). The professional-body link is NATA Recommendation 27, which NATA itself grades SOR C. |
| Syncope from a dry fast | Reduced plasma volume, orthostatic intolerance | Zero studies measuring syncope as an endpoint. The nearest measured proxy is a questionnaire item for dizziness, p = 0.039. |
| Death from a dry fast | Progressive free-water deficit to fatal hypernatremia or circulatory collapse | Zero peer-reviewed reports. One news report of a 58-year-old man after 19 days, with no autopsy and no cause of death established. |
| Kidney stone formation from a dry fast | Low urine volume and low citrate raise supersaturation | The mechanism is measured (604 mL vs 1031 mL urine volume, p = 0.001; citrate p = 0.034). The outcome is not: 7 of 9 studies found no change in renal colic, and the systematic review's verdict is that increased stone risk is unlikely. Mechanism documented, outcome not demonstrated. |
| Harm to a healthy child from a dry fast | Higher body-water turnover, lower reserve | Zero studies in healthy children located. All pediatric evidence located is in children with type 1 diabetes or adrenal insufficiency. |
| Eating-disorder relapse from a dry fast | Fasting as a restriction behavior | The evidence located concerns intermittent fasting generally (DOI 10.1016/j.eatbeh.2022.101681), direction only, no effect estimates obtained. Zero dry-fasting studies. |
G3. The findings that cut AGAINST the hazards, carried at full weight
This register publishes what cuts against it, which in this strand means publishing what cuts against the harms as well. These are not footnotes.
1. AKI was 34% lower in Ramadan fasters. 998 propensity-matched patients: 13.4% versus 20.4%, adjusted OR 0.65 (0.44 to 0.98). DOI 10.1186/s12882-022-02674-1. MEASURED. 2. Emergency department attendances fell during Ramadan. Median 137 per day versus 190 before, p < 0.001, across 44,146 attendances. DOI 10.4274/eajem.galenos.2019.89410. MEASURED. 3. Plasma osmolality went down, not up, in a measured 19-day dry fast, p < 0.001, and serum sodium did not change (p = 0.129). Klatte 2023. MEASURED. This directly contradicts the dehydration-by-concentration story. 4. Lithium did not rise. p = 0.052 in the direction of lower, n = 250. DOI 10.1186/s40345-025-00378-7. MEASURED. 5. Renal colic was unchanged in 7 of 9 studies, 9,906 fasting participants, and the systematic review's conclusion is that increased stone risk is unlikely. DOI 10.5489/cuaj.6664. MEASURED and READ. 6. Pregnancy umbrella review verdict: "There is no sufficient evidence" and "little evidence" of negative impact, across 13 reviews. DOI 10.1002/ijgo.16127. READ. 7. Breastfed infant growth did not differ, 55 infants, all between-group p values NS. J Fasting Health 2013;1(2):66-69. MEASURED. 8. Warfarin INR and time in therapeutic range were unaffected during the fast, 446 patients. DOI 10.1007/s00228-022-03281-7. MEASURED. 9. Two German dissertations measuring a 19-day dry fast both concluded it was safe. Klatte 2023: "Insgesamt wird der zeitlich begrenzte Verzicht auf Nahrung und Flüssigkeit als sicher durchführbar eingestuft." Kleimaier 2022: positive effects on physical and psychological well-being. READ. 10. The daytime dry fast in 11 healthy men was called physiologically harmless. DOI 10.3390/nu14010148, Nutrients 2022;14(1):148, Mähler A, Jahn C, Klug L, Klatte C, Michalsen A, Koppold-Liebscher D, Boschmann M. MEASURED: weight −1.7 ± 1.0 kg (p = 0.0002), BMI −0.5 ± 0.3 kg/m² (p = 0.0003), body fat −1.1 ± 1.6% (p = 0.04), 19 days, no dropouts. READ, verbatim conclusion: "BF, as a form of daytime dry fasting, can be considered physiologically harmless over a period of two weeks". PMID not confirmed: a PubMed link on that page includes the number 35011024, but I could not verify it as the PMID, so it is not stated as one. 11. UAE construction workers in WBGT above 27.5°C showed no adverse physiological effects when given fluid. DOI 10.1186/1745-6673-3-21. MEASURED and READ.
The single sentence that reconciles G1 and G3, and it is a statement about exposure, not a softening of either column: every finding in G3 and almost every finding in G1 was measured on an intermittent dry fast of 11 to 18 hours with overnight rehydration. The only continuous multi-day dry fast that was measured in a hospital produced sodium 187 mmol/L, CK above 33,000 U/L and multiorgan failure in a single patient. The evidence base is reassuring about the exposure it studied and silent about the exposure it did not.
dryfasting.com has been published since 2014. This evidence record was compiled on 22 November 2026. Every row names the fasting modality the study actually ran, the system it was measured in and the grade. Where no source carries a DOI or PMID the row says so rather than supplying one. Nothing is sold on this site.