Ketosis & Short Fasts — A Sardine Protocol Dossier

TL;DR

A 3–5 day sardine fast reliably produces nutritional ketosis (β-hydroxybutyrate ≥ 0.5 mmol/L) in metabolically healthy adults — typically by mid-day-2 if the prior week was moderate-to-low carbohydrate, by day 3 otherwise. Peak βHB on a sardine fast typically runs in the 1.5–2.5 mmol/L range during days 3–5, lower than a true water fast (often 3–5 mmol/L) because dietary protein supplies gluconeogenic substrate. The ketosis biology of short sardine fasts is built on fifty years of human starvation physiology research ^{1 2}, augmented by modern ketogenic-diet outcome literature ^{3 4} and the protein-spared-modified-fast tradition ^{5 6} from which the protocol descends.

What we mean by ketosis

Ketosis is the metabolic state in which the liver produces β-hydroxybutyrate (βHB) and acetoacetate — collectively, ketone bodies — at rates high enough that they become a meaningful systemic fuel. Nutritional ketosis is conventionally the range from about 0.5 to 3.0 mmol/L of blood βHB. Diabetic ketoacidosis is a separate, pathological condition reaching 10–25 mmol/L with severe acidosis; it is not what happens during a healthy adult's short fast, and the conflation is one of the most persistent confusions in popular nutrition writing.

The substrate-level transition is well-mapped. The liver stores roughly 80–120 g of glycogen — perhaps 18–24 hours of fuel at typical resting metabolism, less with exertion. Once liver glycogen is depleted, the liver shifts to gluconeogenesis (producing glucose from amino acids and glycerol) and ketogenesis (producing βHB and acetoacetate from fatty acids delivered by adipose lipolysis). Owen and colleagues demonstrated through cerebral arteriovenous-difference measurements that the brain progressively shifts from glucose oxidation to ketone oxidation as the fast extends, with βHB and acetoacetate eventually supplying the majority of cerebral oxidative metabolism after multi-week starvation ². Cahill's 1970 review integrated this with whole-body fuel partitioning across organ systems and remains the foundational synthesis the modern fasting literature builds on ¹.

What makes ketones biologically interesting isn't only that they are an emergency fuel. β-hydroxybutyrate also acts as a signaling molecule — it inhibits class I histone deacetylases (HDAC1/2/3), modulates NLRP3 inflammasome activation, and binds receptors regulating inflammation and oxidative stress ^{7 8}. Some of the long-term benefits attributed to ketosis (anti-inflammatory effects, oxidative stress reduction, neuroprotection in animal models) are likely mediated by βHB-as-signal rather than only βHB-as-fuel.

What the evidence says (the public preview cuts here)

The clearest human evidence for ketosis biology comes from three converging literatures: classical starvation physiology (Cahill, Owen), modern ketogenic-diet clinical trials (Volek, Hallberg, Westman), and the intermittent-fasting and fasting-mimicking-diet research lineage (Mattson, Anton, Brandhorst-Longo).

Substrate-level human ketosis: The Cahill 1970 review ¹ integrates preceding starvation work into the canonical map of substrate transitions across days and weeks of fasting. Owen 1967 ² supplies the direct measurement of cerebral ketone uptake. Klein & Wolfe 1992 ⁹ contributes the carbohydrate-restriction-versus-true-fast comparison that shows how a low-carb diet largely reproduces the substrate biology of fasting without the energy deficit.

Modern ketogenic-diet outcomes: The Virta Health 2-year cohort in Hallberg 2018 ³ reported substantial T2D remission/reversal with sustained nutritional ketosis. Westman 2008 ⁴ is one of the early RCTs in T2D. Volek 2009 ¹⁰ reports metabolic-syndrome biomarker improvements in a randomized comparison.

Athletic ketogenic adaptation: This is where the literature is honestly mixed. Phinney 1983 ¹¹ reported preserved sub-maximal endurance after ketogenic adaptation in cyclists; Volek and Phinney's 2016 FASTER study ¹² extended this with detailed substrate-utilization measurements in elite ultra-endurance runners. The Burke 2017 race-walkers study ¹³ found clear performance impairment at race-pace efforts in elite walkers undergoing ketogenic adaptation. The honest reading is: ketogenic adaptation appears to preserve sub-maximal endurance well but impairs high-intensity glycolytic-dependent performance.

Intermittent fasting and metabolic switching: Mattson and colleagues developed the "metabolic switch" framing in a series of reviews ^{14 15 16}, arguing that repeated engagement of the glucose-to-ketone switch — not absolute ketone level — drives the adaptive responses associated with intermittent fasting protocols. The Halberg 2005 intermittent-fasting-and-insulin paper ¹⁷ supplies clamp-measured insulin sensitivity improvements in healthy young men after 15 days of every-other-day 20-hour fasts. The Brandhorst-Longo 2015 fasting-mimicking-diet paper ¹⁸ tests a 5-day, very-low-calorie, low-protein cyclical pattern that engages the same biology a sardine fast targets, with healthspan-marker readouts in midlife adults.

βHB as a signaling molecule: The Veech 2004 review ⁸ is the canonical synthesis of βHB's metabolic and signaling roles. Newman & Verdin 2014 ⁷ updates this with the HDAC-inhibitor and inflammasome-modulation literature. Most of this evidence is preclinical (cell-line, animal); the human data on βHB-as-signal is much sparser, and the specific mechanism-to-outcome attribution that some popular content makes is not yet supported by direct human evidence.