Dry Fasting — The Record

Pathway atlas · node 28 of 41

Glycolysis

What fasting does to this node, by modality, by system and by grade. Disconfirming evidence is reported first.

Founder's key components: GLUT1 (SLC2A1), HK2, PFK, PKM2, LDHA Founder's description: "Cytoplasmic breakdown of glucose to pyruvate/lactate, supplying ATP and biosynthetic intermediates."

DRY status

DRY: none located. No peer-reviewed human or animal DRY-modality measurement of glycolysis, hepatic glycogen, glucose turnover or the glucose-to-fat switch was located. Query run: dry fasting Filonov сухое голодание метаболизм гликолиз исследование (Russian plus English, aimed at the Filonov tradition the brief names). Channel returned material only: Filonov's own books and question-and-answer PDFs, clinic pages, a Goodreads listing, blog reposts. Zero journal records, zero measured endpoints. This is the protective finding. It is not evidence that dry fasting does or does not move this node. It is the absence of any measurement to reason from. Further DRY-specific queries in Chinese, Japanese, German, French, Arabic, Persian, Turkish, Spanish and Korean could not be run: the turn's shared web search budget was exhausted. Those channels are recorded in the receipt as NOT RUN, not as empty.

Disconfirming, nulls and conflicts first

D28.1. A 72 hour fast did not lower muscle glycolytic intermediates. In lean subjects it RAISED two of them. Bak AM, Vendelbo MH, Christensen B, Viggers R, Bibby BM, Rungby J, Jorgensen JOL, Moller N, Jessen N. "Prolonged fasting-induced metabolic signatures in human skeletal muscle of lean and obese men." PLoS ONE 2018;13(9):e0200817. DOI 10.1371/journal.pone.0200817. Design: 12 h overnight fast versus 72 h fast, randomized crossover, at least 21 days apart. 18 young men, 9 lean (BMI 19 to 23), 9 obese (BMI 32 to 40). Vastus lateralis biopsies, non-targeted metabolomics. Measured: glucose-6-phosphate and fructose-6-phosphate were about 35% lower in obese than lean at 12 h, and the gap widened at 72 h because G6P and F6P ROSE during fasting in the lean participants only. Pyruvate was about 20% lower in obese at 12 h and was NOT significantly affected by fasting or BMI. Lactate was lower in obese. Maltotriose, a glycogenolysis product, was about 45% lower in obese at 12 h and fell with fasting in lean only. Muscle glucose fell about 25% during 72 h in both groups. NAD+, NADH, NADP+, ATP, ADP and AMP were unaffected. Hexokinase II protein was lower in obese than lean (P = 0.016). pAMPK Thr172/AMPK decreased slightly with fasting (fasting P = 0.048), with no group difference. Modality PF. Tier H2 (randomized crossover, node measured). Grade B+. Note on statistics: the article reports most of these as approximate percentages WITHOUT p values in the body text. Per-metabolite fold changes and p values are in S1 Table, which the served page did not include. The "about 25%" muscle glucose fall therefore carries no p value here.

D28.2. Hepatic pyruvate carboxylase flux did not rise between 12 h and 60 h of fasting. Petersen KF, Dufour S, Cline GW, Shulman GI. "Regulation of hepatic mitochondrial oxidation by glucose-alanine cycling during starvation in humans." J Clin Invest 2019;129(11):4671 to 4675. DOI 10.1172/JCI129913. Measured: hepatic pyruvate carboxylase flux (VPC) essentially UNCHANGED from 12 h to 60 h. Endogenous glucose production (EGP) DECREASED at 60 h, and the authors attribute that to lower net hepatic glycogenolysis, not to any change in gluconeogenic flux through PC. The text states gluconeogenesis stayed relatively constant. Modality PF. Tier H3 (within-subject pre/post, n = 15 by PINTA). Grade B+. This is a null against the common reading that gluconeogenic flux ramps up as a fast lengthens. What changes in these data is the SHARE of glucose production that gluconeogenesis represents, because glycogenolysis collapses.

D28.3. UNRESOLVED CONFLICT on the gluconeogenic fraction in short fasting. Both sides named. Side A: 64 +/- 5% of glucose production from gluconeogenesis at about 22 h of fasting. Rothman DL, Magnusson I, Katz LD, Shulman RG, Shulman GI. "Quantitation of hepatic glycogenolysis and gluconeogenesis in fasting humans with 13C NMR." Science 1991;254:573 to 576. Side B: less than 38% gluconeogenesis after 12 to 14 h of fasting. Nilsson LH, Hultman E. Scand J Clin Lab Invest 1973;32:325 to 330. Wahren J et al. J Clin Invest 1972;51:1870 to 1878. Consoli A et al. J Clin Invest 1987;80:1303 to 1310. Not reconciled here. The time points differ (22 h versus 12 to 14 h) and the methods differ (13C NMR versus earlier balance and tracer methods), and no source I opened ran both in the same subjects. Reported unresolved. Both of these citations reached me through the same secondary source (see retrieval debt R28.1), so the conflict is as that review presents it.

Graded rows

28-R1. Hepatic glycogen falls near-linearly over the first 22 h and is about 83% depleted by 64 h. Grade B+, tier H3, modality PF. Rothman DL, Magnusson I, Katz LD, Shulman RG, Shulman GI. Science 1991;254:573 to 576. Human 13C MRS.

RETRIEVAL ROUTE: these numbers were read out of a review chapter that quotes them and cites Rothman 1991 (Recent Progress in Hormone Research vol 56, chapter 11, "Nuclear Magnetic Resonance Studies of Hepatic Glucose Metabolism in Humans," Endocrine Society PDF). The primary Science paper itself was NOT opened. Second-hand. See R28.1.

28-R2. The fed and early postabsorptive anchor for the same measurement. Grade B, tier H3, modality: none of the nine, this is a fed-state measurement.

Both second-hand through the same review chapter. See R28.1.

28-R3. At 60 h of fasting, glucose and insulin are about 25% lower and glucose-alanine cycling has fallen about 30%. Grade B+, tier H3, modality PF. Petersen KF, Dufour S, Cline GW, Shulman GI. J Clin Invest 2019;129(11):4671 to 4675. DOI 10.1172/JCI129913. 20 healthy, lean, nonsmoking men recruited. PINTA in 15 subjects after 12 h and 60 h fasts. Nine repeated the 60 h fast with alanine infusion. A separate group of 5 had alanine turnover measured at 12 h and 60 h.

Cited inside the paper, not measured in it: forearm net alanine release about 75% reduction during prolonged starvation; beta-hydroxybutyrate turnover about 12-fold increase during starvation. Those two are second-hand. Rodent comparison cited inside the paper: rats fasted 48 h, glucose-alanine cycling fell about 40%, plasma glucose about 6 mM to about 5 mM. That row is tier A1, cited not measured here.

28-R4. Whole-body fuel selection shifts to fat: measured two independent ways, in two different protocols. Grade B+, tier H2 and H3, modality PF and FMD.

28-R5. Brain fuel switches from glucose to ketone bodies in prolonged starvation. Grade B+, tier H3, modality PF. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, Cahill GF Jr. "Brain Metabolism during Fasting." J Clin Invest 1967;46(10):1589 to 1595. DOI 10.1172/JCI105650. PMID 6061736. Three obese patients, 5 to 6 weeks of starvation, cerebral vessel catheterization. Beta-hydroxybutyrate and acetoacetate became the main fuels for brain metabolism, replacing glucose. The authors describe the respiratory quotient as strikingly low and give NO numerical value in the abstract served. See retrieval debt R28.2.

28-R6. In 5 to 6 week starvation, kidney supplies about 45% of total glucose production. Grade B+, tier H3, modality PF. Owen OE, Felig P, Morgan AP, Wahren J, Cahill GF Jr. "Liver and kidney metabolism during prolonged starvation." J Clin Invest 1969;48(3):574 to 583. DOI 10.1172/JCI106016. 11 obese subjects. Catheterization at 35 to 41 days in 5 of them.

DERIVED, NOT STATED: 45% of 86 g/24 h is about 39 g/day of renal glucose output. Hops shown: (1) paper states total 86 g/24 h; (2) paper states kidney share about 45%; (3) 0.45 x 86 = 38.7. The paper does not print 39 g/day.

28-R7. The Randle cycle is a mechanism row for this node, not a human fasting result. Grade C, tier M, modality: none, no fasting intervention. Original: Randle PJ, Garland PB, Hales CN, Newsholme EA. "The glucose fatty-acid cycle. Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus." Lancet 1963;1:785 to 789. DOI 10.1016/S0140-6736(63)91500-9. Read through: Challa AA. "The enduring relevance of the Randle cycle." Nat Rev Endocrinol 2025;21:269. DOI 10.1038/s41574-025-01092-1. What that source states: in isolated heart and diaphragm preparations, and when circulating non-esterified fatty acids rise after adipose lipolysis, fatty acid oxidation increases and glucose utilization falls, independently of hormonal control. Acetyl-CoA from fatty acid oxidation inhibits pyruvate dehydrogenase, named as "the primary mechanism by which fatty acids inhibit glucose utilization." Reverse arm: when glucose is abundant, glucose use in adipose tissue suppresses lipolysis and NEFA release. WHAT THIS SOURCE DOES NOT SAY, so this file does not say it either: phosphofructokinase, hexokinase, citrate and glucose-6-phosphate are not mentioned in the retrieved content. The only inhibitory step named is PDH by acetyl-CoA. The textbook citrate-inhibits-PFK and G6P-inhibits-hexokinase steps are commonly attached to the Randle cycle and were NOT retrieved from an opened source. They are omitted rather than asserted. The 1963 experiments were isolated rat heart and diaphragm. There is no fasting intervention anywhere in this row. The link from the Randle cycle to human fasting physiology is M: it is inferred from the shared upstream node (NEFA rise, which Hoeks 2010 measured at about 9-fold in a 60 h human fast) and not from any experiment that tested the cycle under fasting in humans.

28-R8. Semi-starvation lowers basal metabolic rate by about 40%. CR, not fasting, and a weak retrieval. Grade C, tier H3, modality CR. Minnesota Starvation Experiment, begun 1944. Conscientious objector college students, number not stated in the source opened. Semi-starvation phase 6 months at about 1570 Calories (6570 kJ)/day. Weight loss about 25% of body weight. Basal metabolic rate reduction 40%. Energy saving about 600 Calories/day. Strength decrease 30%. About half of lost fat tissue regained within 3 months of re-feeding. RETRIEVAL QUALITY, stated plainly: this reached me through a teaching handout (ceed.org.au, "Starvation Syndrome: The Biology of Starvation, Ancel Keys' Minnesota study") whose only listed source is a secondary chapter: Garner DM (1997), "Starvation symptoms. The effects of starvation on behaviour: Implications for eating disorders," in Garner D and Garfinkel P (eds), Handbook for Treatment of Eating Disorders, Guilford Press, New York, pp. 145 to 177. The primary Keys et al. 1950 two-volume work was NOT opened and its citation was NOT retrieved. Graded C for that reason, not for the quality of the original study. See retrieval debt R28.3. This is CR by the brief's labels. It is not a fasting modality and must not be read as one.

Open questions, node 28


← 27. Telomere maintenance / telomerase All 41 nodes 29. Glutaminolysis →


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.