← Surfer · Coach hub

Coach — Metabolic Hypothesis of Obesity

Obesity is usually framed as arithmetic: calories in, calories out, then a drug or a surgery that moves the scale. The working hypothesis here starts one step upstream, with stored fat as a buffer in the body's energy economy and with mitochondria that decide whether released fat is ever burned.

Safety first

The extended report opens with its cautions: "Prolonged fasting carries electrolyte imbalance, refeeding syndrome, and decompensation in frail or comorbid patients. It must be undertaken only under qualified medical supervision, with monitoring and a protocol that belongs to a clinician. This essay does not authorize a multi-day fast." On medicines it is equally direct: "Do not stop or start GLP-1 receptor agonists, other weight-loss drugs, or any metabolic medication from a webpage. Lean-mass loss, regain after cessation, and gastrointestinal effects are clinical realities to discuss with the treating team — not reasons to freestyle a protocol."

It is also fair to the drugs: "GLP-1 receptor agonists have robust trial data for weight loss and for certain cardiovascular outcomes. They are not dismissed here. They are questioned for what they leave unaddressed."

A few situations come before any of the ideas below and belong with a physician: unstable heart disease, an active eating disorder, pregnancy, serious psychiatric risk, and any medication emergency. Fasting or extreme restriction while taking an incretin drug or diabetes medicine is a medical risk, not an optimization. Nothing here promises a cure, freedom from medication, or that food, light, and training replace obesity care.

Dashboard and engine

The dominant habit in chronic disease is to treat the dashboard, the readouts and signals that can be measured and overridden, while the engine underneath goes unrepaired: the mitochondrial networks, the cell's cleanup machinery, the architecture of insulin signaling, and the protein and gene-expression state that actually generates metabolic health. In the report's reading, obesity is rarely a primary failure of willpower or simple arithmetic. It is commonly downstream of an engine that handles fuel poorly: reduced capacity to burn fuel, fragmented mitochondria, excess reactive oxygen with too little cleanup, insulin resistance as both cause and result, and storage and hunger driven by cells that sense an energy shortfall in the middle of plenty.

GLP-1 drugs such as semaglutide and tirzepatide are, in this language, a sophisticated dashboard intervention. They suppress appetite through brain signaling, slow stomach emptying, boost insulin release when glucose is high, and lower glucagon. Large trials show meaningful weight loss, often 15 percent or more of body weight, and, in selected people, fewer cardiovascular events. Those benefits are granted. The report's question is what they leave behind. Pooled analyses suggest that roughly a quarter to two-fifths of the weight lost on these drugs is lean tissue unless resistance work and protein are deliberate. Maintenance usually requires staying on the drug, and stopping often brings substantial regain. On present evidence, the report argues, the drugs lower the load on the engine without rebuilding its capacity to make new mitochondria or clear damaged ones.

Fat loss is release into oxidation

The fat-loss hypothesis in the report starts closer to the lipid itself. Stored fat is not a passive remainder and not a moral category. It is a buffer. In this model, the cell burns fatty acids when insulin is low enough to allow fat to be released, and when the energy chain can accept what fat-burning hands it. If either condition fails, the fatty acids do not vanish. They are put back into storage, rerouted to the liver, or deposited where they do not belong.

Insulin is the brake. Fat release runs through three enzymes inside the fat cell, and stress hormones press the accelerator. Insulin, even at low levels, withdraws that signal and blocks the first enzyme. The same brake is thought to work in fat under the skin and fat around the organs. There is no known private visceral pathway and no exemption for subcutaneous fat. What differs is the gain. Visceral fat cells respond less to the brake and more to the accelerator, so they release fat more readily. Thigh and hip fat is the most insulin-sensitive and the slowest to let go. Some abdominal fat drains straight to the liver through the portal vein, so its fatty acids meet liver cells before they meet muscle.

That explains a pattern the report calls spillover. Fat under the skin is not inert; it protects by expanding, recruiting new small fat cells to store surplus and keep it away from the liver, pancreas, heart, and muscle. When that expansion fails, surplus spills into visceral fat and into those organs. Visceral fat is a minor depot, but it carries a disproportionate share of risk, perhaps because it is leaky and drains to the liver. Lipodystrophy, the near-absence of fat under the skin, makes the point from the other side: people with it develop severe insulin resistance and fatty liver without being obese. The missing buffer is the disease.

The order of loss follows the same rules. Liver glycogen goes first, over hours, then liver fat, which can fall substantially within days of fasting or carbohydrate restriction. Visceral fat shrinks fastest as a percentage of itself; fat under the skin shrinks at the same time, more slowly as a percentage but in far greater total amount, because that is where most of the fat was. Early on, visceral loss outpaces subcutaneous loss; as loss deepens the two converge. No diet, drug, or fast spot-reduces. A waist that improves before the scale moves is a sign that the brake has lifted, not a sign that fat under the skin was spared.

Release is only half of loss. A freed fatty acid can be stored again, rebuilt by the liver, or burned. Burning happens in mitochondria, and the report names muscle as the sink that makes released fat safe. A muscle cell with healthy mitochondria clears fatty acids that a sedentary, insulin-resistant one refuses, and refused fat returns to the liver. That is why the lean-mass cost of rapid weight loss is not a footnote here: it is loss of the largest place the released fat can safely go. The report adds a composition point. Fat built from seed-oil polyunsaturates peroxidizes as it is mobilized, producing reactive aldehydes that damage the energy chain itself, so a loss that ignores what the fat is made of can improve the tape while worsening the cell's oxidative balance.

The drug and the fast, on the same axis

Imaging studies of the GLP-1 drugs show liver, visceral, and subcutaneous fat falling together in people who lose weight, with visceral fat shrinking faster as a percentage. A 2025 meta-analysis of thirty-six randomized trials put the pooled visceral reduction with these drugs near 14 percent, behind diet plus exercise and behind very-low-energy diets. People who do not lose weight on the drug may not lose visceral fat. The report's reading is that the drugs are not visceral-selective. They work mostly through the deficit, and stopping without a maintained deficit may return liver and visceral fat as the brake goes back on.

The report describes a medically supervised prolonged fast from the same angle. Insulin falls toward its floor, glucagon, growth hormone, and stress hormones rise, and the liver shifts to making glucose and then ketones. The brake is not adjusted; it is released. The molecular argument for fasting rests on a 2024 study in Nature Metabolism that tracked nearly three thousand plasma proteins every day in twelve healthy volunteers during a seven-day water-only fast. The first two days looked like the expected switch to fat burning. Most of the large changes came only after day three, in a second wave involving immune, stress-response, and tissue-structure proteins. Three days of refeeding largely reversed an apparent lean-mass loss while the fat loss persisted. These were healthy volunteers, not people with obesity, and the report is explicit that it does not promote the finding into a result already shown in patients.

On the fat-loss account, the advantage of the fast over the drug is depth, and the disadvantage is the same depth. Sodium, potassium, magnesium, and phosphate can leave with the water, and refeeding after depletion can trigger the syndrome named for it. A long fast in someone whose muscle is already poor, or whose medicines assume a fed insulin level, is not a cleaner version of the drug. It is a larger push on the same axis, with a narrower safety margin. Neither tool is a long-term strategy by itself. Both fail if the weeks afterward bring back high insulin, seed-oil membranes, and disuse.

Where the two hypotheses disagree

The report sets two hypotheses side by side and does not pick a winner. The repair hypothesis argues that a supervised prolonged fast with metabolic support restores engine function better than chronic GLP-1 treatment alone, because it crosses the day-three threshold, switches on cellular cleanup, and, when refeeding includes protein and resistance work, steers incoming nutrients toward lean tissue. The fat-loss hypothesis is stingier. It grants the fast a deeper signal and prices that depth as danger, and it predicts that once insulin and deficit are matched, the drug and the fast will not differ in which fat goes first. Both can be true if repair is real and depot order is still just lipolytic gain. The superiority claim fails if matched deficits produce the same fat loss, the same lean-mass fraction, and the same mitochondrial markers.

What both agree on is the practical center. The mitochondrion is upstream. Muscle matters. Fuel quality, discussed in the extended report, a coherent daily rhythm of morning light and dark nights, and progressive resistance training with steady aerobic work are how the brake lifts without a permanent clamp. Measures of capacity, such as grip, fitness, lean mass, and recovery, belong beside weight and waist; the common measures of capacity and the endocrinology measures are described elsewhere, along with the shared failure modes, such as treating a drug as the only strategy. Diabetes shares the same engine and has its own essay.

The report is clear about what it refuses: "It does not tell anyone to start or stop a GLP-1 or any other drug." GLP-1 therapy can be a bridge for urgent need. The destination, in this hypothesis, is an engine that can spend its own buffer again.

The short version is in the lessons. Diagnosis, medication decisions, and any decision about a prolonged fast stay with you and the physician who knows your case. Talk with your own physician before changing medications or supplements.

A working hypothesis, open to correction and revised as new evidence arrives. Last revised October 31, 2026.

See the notice at the top of this page. Working hypothesis, not medical advice. Individual care requires personalized evaluation by a qualified physician.