Energy and the Macronutrients

The kilocalorie is a unit of heat, and food energy is still stated in it because the first measurements burned food in a calorimeter and measured the heat.⁸ One kcal (kilocalorie) is 4.184 kJ (kilojoule) (4.1868 kJ (kilojoule) in the older "International Table" definition), and a food "Calorie" with a capital C is a kilocalorie.¹⁸ The label's 4, 4 and 9 kcal (kilocalorie) per gram are "general factors" — U.S. regulation permits them alongside food-specific factors, a fiber-subtracting variant, and direct measurement.¹⁷ What a body actually gets depends on the food: in eighteen adults fed measured diets, whole almonds delivered about 4.6 kcal (kilocalorie) per gram against the 6.0–6.1 the factors predict, a 32% overestimate.²⁷

Where the energy goes. Basal metabolic rate is what a body spends at rest, in a neutral temperature, after digestion is complete — the cost of being alive.⁷ About 40% of the energy released in breaking down fuel is captured as ATP; the rest is heat, which is why the body is warm.² For a mostly sedentary adult, resting metabolism is roughly 60–75% of the day's spending, the thermic cost of processing food about 10%, and activity the remainder — the only share under direct control.²⁴ The thermic cost differs by macronutrient: about 0–3% of the energy in fat, 5–10% in carbohydrate, and 20–30% in protein.²³

Physics against physiology. Conservation of energy is exact: over any period the change in the body's stored energy equals intake minus expenditure.²⁵ What is not fixed is the expenditure side, and this is what the "3,500 kcal (kilocalorie) per pound" rule misses — it treats the daily deficit as constant, when expenditure falls as mass falls, so its predictions run on with no plateau.²⁵ The fall can outrun the loss of mass: of sixteen contestants studied after a televised weight-loss competition, fourteen were re-measured six years later — their resting metabolic rate was 704 kcal (kilocalorie) a day below where it had started, about 500 kcal (kilocalorie) a day lower than their new size alone would predict.²⁶ Intake moves too: twenty adults living on a research ward, offered ultra-processed and unprocessed menus matched for energy, macronutrients, sugar, sodium and fiber, ate 508 kcal (kilocalorie) a day more on the ultra-processed one.²⁹ For orientation, the U.S. reference estimates of daily need are 1,600–2,400 kcal (kilocalorie) for adult women and 2,000–3,000 for adult men, depending on age and activity, for reference-sized adults¹⁶ — population estimates, not anyone's prescription.

The digestive tract does six things in sequence: ingestion, propulsion (swallowing, then peristalsis — waves of muscle contraction), mechanical digestion (chewing, the stomach's churning, the small intestine's segmenting), chemical digestion, absorption, and defecation.⁹ Chemical digestion begins in the mouth with salivary amylase on starch and continues in the stomach, where acid and pepsin start on protein; the small intestine does most of the work with the pancreas's enzymes (amylase, trypsin, chymotrypsin, lipases) and its own lining's enzymes (sucrase, maltase, lactase, peptidases), and it is where the products cross.¹¹ Sucrose is split into glucose and fructose, lactose into glucose and galactose, starch down to glucose; glucose and galactose are pumped across with sodium, fructose diffuses.¹¹

Fat needs help because it does not dissolve in water: bile from the liver, stored in the gallbladder, breaks it into droplets; pancreatic lipase cuts the triglycerides; the pieces are ferried in tiny bile-salt spheres to the lining, reassembled, and packaged as chylomicrons that enter the lymph through lacteals before reaching the blood.¹¹ The fat-soluble vitamins A, D, E and K ride the same route.¹¹ Water moves in large amounts: about nine liters of fluid enter the small intestine each day and about 90% is absorbed there; the large intestine takes back most of the rest.¹¹¹⁰ Chyme takes some 3–5 hours to pass through the small intestine, and residue spends 12–24 hours in the large intestine, where bacteria ferment remaining carbohydrate and the waste is formed into stool.¹⁰

The point of the outline for the rest of the chapter: everything the body does with food it does with the molecules that crossed the wall — glucose, fatty acids and glycerol, amino acids, water, vitamins, minerals — not with the food. That is why the energy a body gets can differ from the label, and why every unit that follows starts from a molecule in the blood.

The sugars. Glucose, fructose and galactose are single sugars; sucrose (table sugar) is glucose joined to fructose and lactose (milk sugar) glucose joined to galactose; starch and glycogen are long chains of glucose, and cellulose — the main fiber of plants — is a glucose chain in a linkage no human enzyme can cut.¹¹³ So digestible carbohydrate arrives in the blood mostly as glucose, and it is glucose the loop regulates.³ The U.S. reference sets a Recommended Dietary Allowance for carbohydrate of 130 g a day for adults and a fiber goal of 14 g per 1,000 kcal (kilocalorie) — reference figures for a population, not a rule for a reader.¹⁶

The loop, in the source's own terms. In the absorptive state, the hours after a meal, rising blood glucose makes the pancreas's beta cells release insulin; insulin has liver, muscle and fat cells take glucose up, has liver and muscle store it as glycogen, and promotes fat and protein synthesis.⁶¹² In the postabsorptive state, once the meal is stored and glucose drifts down, the alpha cells release glucagon, which has the liver break glycogen back to glucose (glycogenolysis) and make new glucose from lactate, glycerol and amino acids (gluconeogenesis).⁶³ The source gives the band the loop defends as roughly 70–110 mg/dL, about 4–6 mmol/L.¹² If a fast runs to days, the body spares glucose for the brain, muscle turns to fatty acids, and after several days ketone bodies made from fat become a major fuel for the heart and other organs.⁶

Where surplus goes. Glycogen is a bounded store — in three men fed a large carbohydrate excess after depletion, capacity was about 15 g per kilogram of body weight and could take roughly 500 g before net fat synthesis began; once it was full, the surplus was disposed of by burning more carbohydrate and by making fat, around 150 g a day from some 475 g of carbohydrate.²⁸ The order matters for the trap: carbohydrate becomes body fat at sustained excess, not on the evening it is eaten.

Fuel. In the postabsorptive state glucagon stimulates lipolysis — triglyceride split into fatty acids and glycerol — and the fatty acids are broken down in mitochondria, two carbons at a time, to feed the same cycle that burns glucose; the glycerol goes to the liver, where it can become glucose.¹²⁴ Fatty acids cannot be turned back into glucose, which is why the brain, in a long fast, is supplied by gluconeogenesis and then increasingly by ketone bodies made from fat.⁶⁴ Per gram, fat carries more than twice the energy of carbohydrate or protein — the 9 against 4 of the label — because its carbons are more reduced; the same reason makes it the compact way to store energy.⁴¹⁷

Store. Dietary fat, digested and reassembled, reaches fat tissue as triglyceride; surplus glucose is converted to fatty acids in the liver (lipogenesis) once glycogen is full, and stored the same way.⁴²⁸ The store is not inert: it is drawn on every day between meals and refilled after them, and its size is the running total of energy balance.²

Nutrient. Linoleic acid (an omega-6 fatty acid) and alpha-linolenic acid (an omega-3) are essential — the body cannot make them, so they must come from food; the body converts alpha-linolenic acid to the longer omega-3s EPA and DHA only to a small extent, under 15% by reported rates.²⁰ The U.S. reference sets adequate intakes of 12 g of linoleic and 1.1 g of alpha-linolenic acid a day for women aged 19–30, and 17 g and 1.6 g for men — population reference values.¹⁶ The fat-soluble vitamins are absorbed inside the same bile-salt spheres as dietary fat, so a meal's fat carries them in.¹¹

The nine essential amino acids are histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan and valine; the other eleven the body makes, some only when it is well.³² Digestion frees them — pepsin in the stomach, trypsin and chymotrypsin from the pancreas, peptidases in the intestinal lining — and they cross into blood as single amino acids for the liver and every other tissue.⁵

Turnover and disposal. Body proteins are made and unmade continually, so the pool is always in use; what is drawn from it beyond building is deaminated — the amino group removed as ammonia, which the liver combines with carbon dioxide into urea for the kidneys to excrete — and the carbon skeleton enters the same pathways as any fuel, as pyruvate, acetyl-CoA or a cycle intermediate.⁵ Because there is no store, the source states the rule plainly: amino acids in excess are converted to glucose or ketones, or decomposed.⁵ Processing protein is itself costly — 20–30% of its energy is spent in the handling, against 5–10% for carbohydrate and 0–3% for fat.²³

Reference values, and what they are. The U.S. Recommended Dietary Allowance for protein is 0.8 g per kilogram of body weight a day for adults — 46 g for the reference woman and 56 g for the reference man — an allowance set to cover the needs of about 97% of healthy people, not a ceiling on what the body can use and not a target for any one reader.³¹¹⁶ The trials that measured what more protein does found the extra gain in lean mass, in adults doing resistance training, level off at intakes above about 1.6 g per kilogram a day.³⁰ What the number is not: a rule for the learner, whose needs are their own and their clinician's question.

How much water, and where. Water is about 75% of body mass in infants, 50–60% in adult men and women, and as little as 45% in old age, the proportions of muscle, fat and bone shifting the total; most of it is inside cells.¹³ A typical adult takes in about 2,500 mL a day, of which some 230 mL is made in cells as fuel is burned and the rest arrives in food and drink; about the same leaves, most as urine — around 1.5 L — and the remainder through the skin, from the lungs (together the "insensible" loss), and in stool.¹⁴ The National Academies' 2004 reference report put average total water intake for adequately hydrated adults at about 2.7 L for women and 3.7 L for men, about 80% of it from water and other beverages, caffeinated ones included, and about 20% from food.³³

The regulators. Osmoreceptors in the hypothalamus watch the concentration of the blood; when water is lost in breath, sweat or urine and the blood grows more concentrated, they trigger thirst and the release of ADH (antidiuretic hormone), which has the kidney's collecting ducts insert water channels and reabsorb water, so urine falls and concentrates; when the blood is dilute, ADH (antidiuretic hormone) falls, the channels are withdrawn and more water is passed.¹⁴ Dehydration in the source's sense — a net loss of water leaving too little in blood and tissues — is what the loop exists to prevent, and it can be outrun.¹⁴

Two things this unit deliberately does not say. It does not say how much any reader should drink: the review that traced the "eight glasses" rule found no scientific study behind it and, in the surveys it examined, healthy adults in temperate, mostly sedentary lives met their needs at lower totals — while saying plainly that vigorous work, heat and some illnesses call for much more.³⁴ And it does not turn the reference intakes into targets: the report that set them said it offered no rule of thumb in glasses, and the figures are population averages.³³

The designs, in the standard vocabulary. In an experimental study the investigator decides who is exposed — in a randomized trial, by lot; in an observational study the investigator only records exposure and outcome.²¹ Observational designs differ in direction: a cohort study records exposure and follows people forward; a case-control study starts from people with the disease and looks back at exposure; a cross-sectional study measures both at once.²¹ The CDC's own text says the limit plainly: epidemiology by itself can never prove that a particular exposure caused a particular outcome.²¹ Two failures to name in any observational result — confounding, where the exposed differ in something else that carries the outcome, and reverse causation, where the outcome shaped the exposure — are why the placebo adherers of the Coronary Drug Project matter to nutrition: adherence itself marked a lower-risk person, 15.1% against 28.3% five-year mortality with no active drug in either group.³⁸

What a trial can say, and what tight control buys. Randomization spreads the unmeasured differences across the arms, so a difference in outcome can be laid to the intervention; the price is scope. The Women's Health Initiative randomized 48,835 postmenopausal women in 1993–98 and followed them 8.1 years; the intervention group cut fat by about 8 percentage points of energy, not to the 20% of energy planned, and neither coronary disease nor invasive breast cancer fell significantly.³⁶³⁷ That is a result about that intervention as delivered, in those women, over that time. At the other extreme of control, twenty adults lived on a research ward for four weeks, two on each of two menus matched for energy, macronutrients, sugar, sodium and fiber and eaten as they pleased; on the ultra-processed menu they ate 508 kcal (kilocalorie) a day more and gained 0.9 kg, on the unprocessed one they lost 0.9 kg.²⁹ Intake was measured, not asked — which is what makes the difference believable — and the study says nothing about years or about life outside a ward.

Measurement, and multiplicity. In 484 adults whose true energy expenditure was measured with doubly labeled water, food-frequency questionnaires understated energy intake by 31–36% in men and 34–38% in women; even 24-hour recalls understated it by 12–20%.³⁹ And single studies are many: of 50 ingredients drawn from random cookbook recipes, 40 had published studies of their cancer risk; 72% of 264 single-study assessments claimed an increased or a decreased risk, most on weak or no statistical support, and the pooled estimates from meta-analyses were on average null.⁴⁰ The chapter's three questions — what was measured, who was compared, what could not be shown — are what to ask of the next headline.