Your blood glucose has to stay available between meals and overnight, even though no carbohydrate is entering the intestine. The liver makes that possible. It stores part of the incoming glucose after a meal and releases glucose later when the circulation needs it.
That makes the liver a metabolic buffer, not a passive filter. It receives nutrient-rich blood from the portal vein, senses hormonal and neural signals, chooses among storage, oxidation, conversion, and release, and communicates with muscle, fat, pancreas, kidney, gut, and brain.
The same flexibility can become dysregulated. In type 2 diabetes, the liver may continue releasing too much glucose even when insulin should suppress production. That excess output is an important contributor to fasting hyperglycemia, but it is one part of a system that also includes pancreatic beta-cell dysfunction, altered glucagon, muscle insulin resistance, adipose-tissue flux, kidney glucose handling, and other processes.
The liver sees a meal early#
Digested carbohydrate is broken into monosaccharides and absorbed from the intestine. Portal blood carries much of that nutrient load to the liver before the remainder reaches the systemic circulation; the pancreas also responds to rising glucose, incretin hormones, amino acids, neural signals, and other meal cues.
Insulin rises and the insulin-to-glucagon balance shifts. In the liver, this state favors glycogen synthesis, glycolysis, and suppression of glucose production. The liver does not remove every glucose molecule from portal blood. Skeletal muscle and other tissues also dispose of glucose. The relative contributions depend on meal composition, activity, and insulin sensitivity. They depend on prior glycogen status and methods used to estimate flux.
Insulin affects hepatic glucose handling through direct and indirect routes. Direct signaling changes enzyme activity and gene regulation in hepatocytes, and indirect effects include reducing release of free fatty acids from adipose tissue, altering substrate delivery, suppressing glucagon, and changing neural and hemodynamic signals. So a diagram showing insulin binding to a liver cell captures only part of the control system: the liver responds to the concentration of hormones reaching it, the ratio among those hormones, incoming substrates, cellular energy, prior feeding, circadian timing, and signals from other organs.
Glycogen is the short-term reserve#
The liver links glucose molecules into glycogen after feeding. Glycogen is a branched polymer that can be built and mobilized rapidly, and when food is no longer arriving, hepatic glycogen phosphorylase helps release glucose units, which enter glycolytic intermediates and can become free glucose for export.
The enzyme that makes export possible is glucose-6-phosphatase. It removes phosphate from glucose-6-phosphate, allowing glucose to leave the hepatocyte and enter the circulation. Liver expresses this system; skeletal muscle does not use its glycogen as a direct source of circulating free glucose. Muscle glycogen primarily fuels contraction within the muscle.
Calling liver glycogen a tank can be useful, but the analogy has limits. Its size changes with recent carbohydrate intake, fasting duration, and activity. It changes with hormones, illness, and liver health. Glycogen is also continuously turning over. After an ordinary overnight fast, both glycogenolysis and gluconeogenesis support glucose output rather than one pathway turning off exactly when the other begins.
As fasting continues, hepatic glycogen becomes depleted and the relative contribution of gluconeogenesis rises. The timing varies across people and circumstances. A fixed claim that glycogen always lasts a specified number of hours is too rigid.
Gluconeogenesis builds glucose from smaller molecules#
Gluconeogenesis is the synthesis of glucose from non-carbohydrate precursors, and it occurs mainly in liver and also in kidney, with the renal contribution becoming more important during prolonged fasting and certain physiologic states.
Major precursors include:
- Lactate, produced through glycolysis in red blood cells, exercising muscle, and other tissues. The liver can convert lactate back toward glucose in the Cori cycle.
- Glycerol, released when triglycerides in adipose tissue are broken down.
- Glucogenic amino-acid carbon, including alanine carried from muscle and other tissues.
- Pyruvate and related intermediates, which enter the pathway according to metabolic conditions.
Fatty acids do not generally provide net glucose carbon in humans, with limited exceptions such as the glycerol part of triglyceride and carbon from odd-chain fatty acids, and fat oxidation is still essential because it supplies energy and signals that support gluconeogenesis during fasting.
Gluconeogenesis is not an emergency-only pathway and is not synonymous with starvation. It contributes during an overnight fast and helps maintain glucose for tissues with substantial glucose needs. Its rate reflects substrate delivery, enzyme regulation, redox state, energy, and hormones.
Insulin and glucagon set opposing instructions#
Insulin is the principal signal that feeding is available. In the liver it suppresses glycogen breakdown and gluconeogenic programs while promoting glycogen synthesis and other fed-state pathways; it also suppresses adipose lipolysis, reducing glycerol and fatty-acid delivery that would otherwise support hepatic glucose production.
Glucagon is secreted by pancreatic alpha cells. During falling glucose and fasting, it acts strongly on liver. Acute glucagon signaling promotes glycogenolysis; with longer stimulation it supports the enzymatic and transcriptional conditions for gluconeogenesis. Its effect depends on available glycogen, substrate supply, insulin, and other hormones.
Glucagon is not simply “the opposite of insulin” in every tissue. The liver is its major glucose-regulating target. Skeletal muscle has different receptor biology and controls its own glycogen for local use.
Epinephrine can rapidly increase hepatic glucose output during exercise, stress, or hypoglycemia. Cortisol and growth hormone have slower permissive and counterregulatory effects. The autonomic nervous system and central sensing contribute as well.
Counterregulation is protective, but it can complicate diabetes management. After hypoglycemia, hormonal responses can raise glucose, while overtreatment of the rebound can produce a further rise, and your own pattern needs data, not a universal story assigned from one morning value.
Why glucose rises before breakfast#
Some people experience a pre-breakfast rise called the dawn phenomenon: circadian increases in growth hormone, cortisol, catecholamines, and hepatic glucose output occur while insulin action or secretion may be insufficient to offset them.
Not every high morning reading is the dawn phenomenon. Food timing, overnight medication action, and sleep can each produce a different overnight trajectory. So can illness, stress, alcohol, and late exercise. So can insulin delivery problems and nocturnal hypoglycemia.
One fasting result cannot tell those apart. Continuous glucose monitoring or structured overnight measurements can. Take the pattern to your clinician before changing anything, especially if you use insulin or a medicine that can cause hypoglycemia.
What hepatic insulin resistance means#
In a healthy fed state, insulin suppresses hepatic glucose production. In hepatic insulin resistance, a given insulin signal suppresses that output less effectively. The liver may continue releasing glucose despite hyperglycemia and elevated insulin.
This is especially relevant to fasting glucose. Overnight hepatic output supplies much of the glucose entering the circulation. If production remains high and pancreatic insulin secretion cannot compensate, your fasting glucose rises.
The mechanism is not confined to a defective insulin receptor in isolation. Excess fatty-acid delivery, ectopic lipid intermediates, and inflammation can contribute. So can mitochondrial flux, altered glucagon, and substrate supply. So can neural signals and gene regulation. Which mechanisms dominate varies with disease stage and experimental model.
Type 2 diabetes also includes beta-cell dysfunction. Early in disease, beta cells may produce more insulin to compensate for resistance. As compensation becomes inadequate, both fasting and post-meal glucose can rise. Muscle insulin resistance can reduce post-meal disposal, while adipose insulin resistance increases fatty-acid and glycerol flow to liver; that whole network is why one fasting glucose result cannot tell you which organ is at fault.
The selective insulin-resistance puzzle#
The liver in type 2 diabetes can continue producing glucose while also making lipid. At first glance, that seems contradictory: if insulin signaling is resisted, why does insulin-associated lipogenesis continue?
The phrase selective hepatic insulin resistance describes this apparent split. Some insulin-controlled pathways that suppress glucose production appear impaired, while lipogenic pathways remain responsive or are driven by other signals, including nutrient substrate and hyperinsulinemia. Glucose-responsive transcription and fatty-acid delivery are among them too.
The concept is useful but incomplete. Hepatic pathways are connected, experimental models differ, and whole-body flux cannot always be assigned to one intracellular branch. Increased liver fat may reflect dietary fat, de novo lipogenesis, adipose release, reduced oxidation, and impaired export in different proportions, and the safe conclusion is that high glucose production and liver fat accumulation can coexist. It is not necessary to claim that every insulin action is equally resistant.
Fatty liver and diabetes form a two-way relationship#
Metabolic dysfunction-associated steatotic liver disease, or MASLD, is common in people with type 2 diabetes and obesity. Insulin resistance, adipose dysfunction, and hepatic lipid flux connect the conditions. So do genetics, diet, and activity. So do sleep and other factors.
Most people with simple steatosis do not rapidly develop cirrhosis, but risk rises with steatohepatitis and fibrosis; fibrosis stage is a stronger predictor of liver-related outcomes than the amount of fat alone.
The relationship is bidirectional. Type 2 diabetes increases risk of progressive liver disease, and liver metabolic dysfunction is associated with incident diabetes and more difficult metabolic control, though that does not mean liver fat is the only cause of diabetes or that every person with diabetes has clinically significant fibrosis.
Current American Diabetes Association Standards recommend screening adults with type 2 diabetes or prediabetes, particularly those with obesity, other cardiometabolic risk, or established cardiovascular disease, for risk of cirrhosis related to metabolic steatohepatitis using the fibrosis-4 index, or FIB-4, even when liver enzymes are normal. Persistently elevated aminotransferases for more than six months with a low FIB-4 should prompt evaluation for other liver causes.
FIB-4 is a first-line risk tool based on age, AST, ALT, and platelet count. It is not a diagnosis. Age affects interpretation, acute illness can distort inputs, and an indeterminate or high result usually leads to a second noninvasive assessment or specialist pathway. Guidance and thresholds should be applied by clinicians in context.
Liver enzymes are not liver function and not glucose tests#
ALT and AST can rise with hepatocellular injury, but normal values do not exclude steatohepatitis or advanced fibrosis. These enzymes are also not direct measures of liver glucose output.
Albumin, bilirubin, international normalized ratio, and platelet count answer different questions. So do imaging, elastography, clinical signs, and other tests. Those questions are about synthesis, excretion, and portal hypertension, as well as structure and fibrosis risk. No single panel gives a complete metabolic map.
Diabetes diagnosis uses validated glucose criteria or hemoglobin A1C in appropriate circumstances, with confirmation when required. A liver panel neither diagnoses nor rules out diabetes. Keeping them separate prevents two errors: reassuring you about metabolic risk because ALT is normal, and reading a mild enzyme elevation as proof that the liver is causing a glucose result.
Advanced liver disease can lower glucose too#
Early metabolic liver disease often coexists with insulin resistance and hyperglycemia. Advanced liver failure can create a different problem. Reduced glycogen reserves, impaired gluconeogenesis, and malnutrition can raise hypoglycemia risk. So can infection, altered hormone clearance, kidney dysfunction, and reduced medication clearance.
Your hemoglobin A1C can also become less reliable when red-cell survival changes because of anemia, bleeding, or transfusion. The same holds for hemolysis, hypersplenism, or kidney disease. Glucose monitoring and alternative measures may be needed under clinical supervision.
Medication choice and dose require special care in cirrhosis. The direction and degree of risk vary by medicine, liver severity, kidney function, nutrition, and acute decompensation. Broad claims that one diabetes medicine is always safe or always prohibited in all liver disease are rarely accurate.
Alcohol changes the fasting equation#
The liver prioritizes ethanol metabolism. The resulting change in the hepatic redox state can inhibit key gluconeogenic reactions. During fasting, after strenuous activity, with low glycogen, or if you use insulin or an insulin secretagogue, this can contribute to delayed hypoglycemia.
Alcoholic drinks can also contain carbohydrate, initially raising glucose. The same occasion can therefore produce an early rise and a later fall. Dose, food, liver health, medicines, and individual metabolism change the pattern.
This is more precise than saying the liver is busy “detoxing.” Ethanol metabolism changes biochemical pathways that help maintain glucose. If you are at risk for hypoglycemia, you need individualized guidance about alcohol, food, monitoring, and medication.
Exercise recruits liver glucose output#
Contracting muscle uses more glucose and its own glycogen. To prevent circulating glucose from falling too far, hepatic glycogenolysis and gluconeogenesis increase. Glucagon and catecholamines rise relative to insulin, with intensity and duration shaping the response.
Moderate aerobic activity often lowers glucose during and after activity. Very intense exercise can briefly raise glucose through catecholamines and rapid hepatic output. Improved insulin sensitivity afterward can increase delayed hypoglycemia risk in people using insulin or certain secretagogues. Training status, recent food, starting glucose, insulin on board, activity type, and time of day all matter, which is why no single glucose direction describes one exercise session for everyone.
Reading a liver-glucose claim#
Ask what was actually measured. Fasting glucose, tracer-estimated endogenous glucose production, and liver fat by imaging are not synonyms. Neither are ALT, insulin concentration, glycogen by spectroscopy, and a clamp-derived measure.
Check whether the study distinguishes total endogenous glucose production from liver production. The kidney also makes glucose, especially during longer fasting. Some methods estimate combined endogenous output unless organ-specific techniques are used.
Separate association from intervention. People with more liver fat may have higher diabetes risk, but that relationship includes shared causes. An intervention that reduces liver fat may improve glucose, yet the degree of change and causal pathway depend on the intervention.
Look at feeding state, time, and medication washout. Hepatic flux changes rapidly after food and across the day. Results obtained during a clamp or prolonged fast answer a controlled physiology question, not every routine-life scenario.
Finally, distinguish mechanism from personal diagnosis. A plausible hepatic mechanism cannot tell you why your glucose changed without measurements and clinical context.
The glucose-buffer conclusion#
The liver protects both sides of glucose balance. It helps prevent an excessive rise after food by storing and processing incoming fuel, and it helps prevent a dangerous fall between meals by breaking down glycogen and making new glucose.
Insulin, glucagon, substrates, nerves, circadian signals, muscle activity, fat tissue, pancreas, and kidney continually adjust that work. Type 2 diabetes disrupts the coordination, so glucose production can remain high when it should be suppressed, and advanced liver disease can disrupt it in the opposite direction and reduce the capacity to maintain glucose.
Understanding the liver therefore replaces a simple food-in, glucose-up story with a regulated-flow model. Blood glucose reflects entry, production, uptake, storage, and clearance across several organs. The liver is central, but it never acts alone.
References#
- DeFronzo RA, et al. Pathogenesis of type 2 diabetes and regulation of hepatic glucose production. Diabetes Care. 2015.
- Rui L. Energy metabolism in the liver. Comprehensive Physiology. 2014.
- Petersen MC, Shulman GI. Mechanisms of insulin action and insulin resistance. Physiological Reviews. 2018.
- Feingold KR, et al. Glucose metabolism during fasting. Endotext.
- American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes 2026. Diabetes Care. 2026.
- Rinella ME, et al. AASLD practice guidance on clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023.
- European Association for the Study of the Liver, European Association for the Study of Diabetes, and European Association for the Study of Obesity. Clinical practice guidelines on MASLD. 2024.
For your own health, talk with your clinician.*
Questions and answers
Does the liver make sugar even if a person eats no carbohydrate?
Yes. It releases glucose from glycogen and makes glucose from lactate, glycerol, and amino-acid carbon. The kidney also contributes, especially during prolonged fasting.
Why can fasting glucose be high when nothing was eaten overnight?
The liver continues supplying glucose during sleep. Hepatic insulin resistance, insufficient insulin secretion, glucagon, circadian hormones, illness, and medication timing can leave that output higher than needed.
Can normal liver enzymes rule out fatty liver or fibrosis?
No. ALT and AST can be normal despite clinically important disease. Risk assessment uses clinical context and validated noninvasive pathways rather than enzymes alone.
Does muscle glycogen raise blood glucose directly?
Not directly. Muscle lacks the glucose-6-phosphatase system needed to export substantial free glucose. Its glycogen fuels muscle, while metabolites such as lactate can travel to liver for recycling.
Can liver disease cause low blood glucose?
Yes, particularly in advanced disease or acute failure, when glycogen storage, gluconeogenesis, nutrition, and medication clearance may be impaired. Hypoglycemia requires prompt clinical assessment.