Why type 2 diabetes often begins in the liver
A fasting glucose goes 5.4, then 5.8, then 6.1 across three annual checks while the after-meal readings stay respectable. The usual suspect is a tiring pancreas. But the number that drifted first was taken at the end of a long night.
A fasting glucose goes 5.4, then 5.8, then 6.1 across three annual checks. Nothing else on the panel has moved, and the readings after meals are still respectable. The usual suspect is a tiring pancreas, the organ that makes insulin. But the number that drifted first was taken at the end of a long night, before anything had been eaten.
By the end of this piece you will know why a morning glucose is often the earliest number to move. You will also see why rising blood fats keep it company, and why the organ behind both is invisible from the outside.
What is the liver doing while you sleep?
The hours between dinner and breakfast are not a pause. The brain alone gets through something like 120 grams of glucose a day and takes no break at night. The liver covers the demand. First it releases glucose from its own store, a packed form of sugar called glycogen. Then it builds new glucose from scratch. Between the evening meal and breakfast, almost all blood glucose is glucose the liver made.
Then you eat, and the tap must close. Otherwise the meal is added to a supply already running. The liver hears that instruction first. Blood from the pancreas reaches it directly through a dedicated vessel, the portal vein, at insulin levels two to three times what the rest of the body sees. Insulin’s first job, before it moves a gram into muscle, is to tell the liver to stop.
So overnight glucose is not a leftover from dinner. It is a tap with a control, and the question is what jams it.
How does fat inside a liver cell muffle the instruction?
The fat that matters sits inside the liver cells themselves, in droplets among the working machinery. It is not the fat around the organ, and not anything anyone could pinch. As the droplets build, certain fat by-products build with them. One of these by-products switches on an enzyme that muffles insulin’s message at the point where it is passed along inside the cell. The instruction still arrives, but it lands softly. Glucose output falls less after a meal and does not stay down overnight.
If the overnight brake slips first, which number drifts earlier: the fasting glucose, or the value after a meal? The fasting one, often by years. Fat inside the liver cell, not fat around it, interferes with the instruction, which is why a morning fasting glucose can be the first number on the panel to drift.
It also explains a pairing that looks like coincidence: fasting glucose and triglycerides, the blood’s transport fats, creep up together.
Why do blood fats rise alongside a rising fasting glucose?
Insulin carries at least two messages to the liver: stop releasing glucose, and turn surplus carbohydrate into fat for export. In a fat-loaded liver the two come apart. This matters, because the failure is usually taught as though the liver had gone completely deaf.
It has not. The liver stops obeying the instruction to halt glucose output while still obeying the instruction to make fat. That is why a rising fasting glucose and rising triglycerides arrive together rather than one after the other. The new fat leaves the liver in transport particles. Triglycerides rise, and HDL, often called the good cholesterol, falls in the same traffic.
Researchers still debate whether the fat-making branch stays genuinely sensitive, or is simply driven hard by all the insulin around it. The signature itself is not debated. It is why a triglyceride to HDL ratio says more than either number alone.
Which leaves the pancreas question: if the liver goes first, how does the pancreas end up in difficulty at all?
How does the pancreas get pulled in?
The best-supported account is two linked cycles, known as the twin cycle model. In the first cycle, liver fat blunts the stop signal. Insulin climbs to compensate, and that extra insulin drives more fat-making. The loop feeds itself.
The second cycle is the spillover. The fatty liver exports more fat, and some of it reaches the pancreas. Fat within the pancreas’s insulin-producing clusters is linked to a weaker early burst of insulin, the burst released within about ten minutes of eating. That burst’s job is to shut the liver’s output down promptly. In the twin cycle model, liver fat causes the liver’s own insulin resistance first, then spills over and blunts the very response meant to close the tap.
The model is taken seriously because emptying the cycles has been tested. Substantial weight loss can put type 2 diabetes into remission, and how often it does tracks how much weight comes off.
Proven: substantial weight loss can produce remission of type 2 diabetes. In a randomised trial run through ordinary GP practices, roughly half the weight-loss group were in remission at twelve months, against a few per cent under usual care. Most of those who lost fifteen kilograms or more achieved it. Remission depends on the weight staying off, and that trial recruited people diagnosed within six years. Also well supported: liver fat rises before diabetes appears and predicts it, shown in imaging and long-term population studies. Promising: the detailed account of how fat muffles insulin’s message inside the cell, built from human tissue and animal work.
All of this would be easier if liver fat announced itself. It does not.
What reads early, when there is nothing to see?
Many people assume a fatty liver is something diabetes causes. In many people the order runs the other way.
It is also invisible. Fat inside liver cells occurs at every body weight, including in slim people never told to lose anything, and causes no symptoms until late. Liver fat cannot be seen from the outside and occurs at any body weight, so the early signal has to be looked for rather than noticed.
Four readings carry most of the early information. A fasting insulin moves long before glucose does. The triglyceride to HDL ratio, read across repeated panels, tracks the export signature. ALT, a liver enzyme that leaks into the blood when liver cells are stressed, is worth following as a trend rather than a single value. One normal ALT does not rule liver fat out. And scans where warranted: ordinary ultrasound misses milder fat, while newer scan methods can measure it.
A fasting glucose that has crept from the low fives to the high fives is not a blank waiting to become abnormal. Read beside triglycerides, HDL and the ALT trend, it is often the earliest readable sign that the brake is slipping.
- Overnight, nearly all blood glucose is liver-made, and insulin’s first job is to stop it.
- Fat inside liver cells muffles that instruction, so the morning fasting glucose often drifts first.
- The resistance is selective: glucose output keeps running while fat-making continues, so triglycerides rise alongside.
- Liver fat comes first, then spills to the pancreas and blunts the early insulin burst.
- Fasting insulin, triglyceride to HDL, the ALT trend and targeted scans read liver fat early, at any body weight.
Once a morning glucose reads as a report on the liver rather than the pancreas, the numbers change meaning. The Metabolic Health and Diabetes knowledge check will show how firmly that reading has settled.
