Fasting glucose is the concentration of sugar in your blood after you have not eaten for at least eight hours, and doctors use it as a frontline screening tool for diabetes and prediabetes. A result below 100 mg/dL is generally classified as normal, 100 to 125 mg/dL falls into the prediabetes range, and 126 mg/dL or higher on two separate occasions points toward diabetes. Those cutoffs sound tidy, but real-world glucose behavior is messier than a three-tier chart suggests, and recent continuous monitoring data have made that especially clear.
What Happens to Your Blood Sugar Overnight
When you stop eating in the evening, your body does not simply let blood sugar coast downward until morning. The pancreas secretes two opposing hormones: insulin, which lowers blood sugar, and glucagon, which raises it. Together they keep glucose in a narrow band while you sleep.
Glucagon’s main job during a fast is to signal the liver to release stored glucose and manufacture new glucose from non-sugar building blocks, a process called gluconeogenesis. Research has shown that glucagon plays a primary counterregulatory role in preventing blood sugar from dropping too low overnight, and that simply reducing insulin is not enough on its own to keep glucose stable during a fast.1PubMed. Insulin, glucagon, and catecholamines in prevention of hypoglycemia during fasting The liver, in other words, is actively working all night. It is not a passive bystander. It continuously adjusts how much glucose it releases based on the hormonal signals it receives.2PubMed Central. Molecular pathophysiology of hepatic glucose production
This matters because your fasting glucose reading is not just a snapshot of “what’s left over” from last night’s dinner. It reflects an active, regulated output from your liver and the hormonal environment your body maintained through the night.
The Standard Diagnostic Ranges
Most clinical guidelines use these thresholds for fasting plasma glucose:
- Normal: below 100 mg/dL (5.6 mmol/L)
- Impaired fasting glucose (prediabetes): 100 to 125 mg/dL (5.6 to 6.9 mmol/L)
- Diabetes: 126 mg/dL (7.0 mmol/L) or higher, confirmed on a repeat test
The prediabetes threshold has shifted over the years. The original cutoff for impaired fasting glucose was set at 110 mg/dL, then lowered to 100 mg/dL in 2003, which substantially increased the number of people who qualify as prediabetic. That lower boundary was controversial from the start; early analyses acknowledged that the evidence supporting the glucose limits for impaired fasting glucose was not strong.3Diabetes Care. Impaired fasting glucose: how low should it go? In practice, the 100 mg/dL line is a useful clinical alert, but treating it as a sharp biological boundary oversimplifies what is really a continuum of metabolic risk.
Fasting glucose is also just one of several diagnostic tools. Doctors may also use an HbA1c test, which reflects average blood sugar over roughly three months, or an oral glucose tolerance test, which measures how well your body handles a sugar load. These tests do not always agree. A meta-analysis comparing the tests found that fasting plasma glucose and HbA1c have only moderate agreement in diagnosing diabetes, with an overall kappa statistic of 0.55.4PLoS ONE. Discordance in the diagnosis of diabetes: Comparison between HbA1c and fasting plasma glucose That means a meaningful number of people will get different diagnoses depending on which test is used. If you are borderline on one test, your doctor may order the other to get a fuller picture.
Why Your Number Can Be Different Every Morning
One of the most striking findings from recent research is just how much fasting glucose bounces around from day to day in the same person. A large continuous glucose monitoring study of over 8,000 non-diabetic adults found that the average day-to-day variation in fasting glucose was about 7.5 mg/dL. Among people whose first morning reading was in the normal range, 40% had at least one subsequent reading that would have landed in the prediabetes range, and about 3% had a reading in the diabetes range at some point during the monitoring period.5PubMed Central. Continuous glucose monitoring and intrapersonal variability in fasting glucose
That variability is not a sign of disease. It is a normal feature of human physiology. Your fasting number on any given morning is influenced by how well you slept, how stressed you were, what time you ate dinner, and how active you were the day before. A single fasting glucose reading is a data point, not a verdict. This is why clinical guidelines require a confirmatory test before diagnosing diabetes and why trends over time matter more than any one number.
The Dawn Phenomenon and Morning Spikes
If you have ever tested your blood sugar first thing in the morning and found it higher than expected, the dawn phenomenon is a likely explanation. In the early morning hours, the body releases a surge of hormones, including cortisol and growth hormone, that naturally increase insulin resistance. Even in people without diabetes, the body needs more insulin in the predawn hours to keep glucose stable.6PubMed. Fasting early morning rise in peripheral insulin: evidence of the dawn phenomenon in nondiabetes If you have any degree of insulin resistance, that extra demand can push your morning reading up.
For people with diabetes, the dawn phenomenon can cause genuine morning hyperglycemia. One study of insulin-dependent diabetes patients found that a dawn rise in blood glucose greater than 50 mg/dL could meaningfully worsen daytime blood sugar control.7Diabetes Care. Dawn Phenomenon and Somogyi Effect in IDDM There is also a distinct but sometimes confused phenomenon called the Somogyi effect, where overnight hypoglycemia triggers a rebound spike in morning glucose. However, the same study found that nocturnal hypoglycemia actually led to lower morning glucose, not higher, calling into question how often the Somogyi effect truly occurs in practice.
What Drives Fasting Glucose Higher Over Time
When fasting glucose creeps upward from the 80s into the 90s or past 100, the main culprit is usually the liver making too much glucose overnight. In people with impaired fasting glucose, the rate of gluconeogenesis (the liver’s production of new glucose) is measurably higher than in people with normal readings.8PubMed Central. Pathogenesis of prediabetes: role of the liver in isolated fasting hyperglycemia and combined fasting and postprandial hyperglycemia This overproduction is driven by insulin resistance in the liver. Normally, insulin tells the liver to slow down glucose output. When the liver becomes less responsive to that signal, it keeps pumping out glucose even when blood levels are already adequate.
This same mechanism is amplified in full-blown type 2 diabetes. Abnormally increased liver glucose production is a significant contributor to fasting hyperglycemia in type 2 diabetes, driven by insulin resistance that fails to properly suppress the liver’s output.9PubMed. Increased hepatic gluconeogenesis and type 2 diabetes mellitus This is why metformin, the most commonly prescribed diabetes drug, works primarily by reducing the liver’s glucose production rather than by stimulating the pancreas to make more insulin.
Things That Can Throw Off Your Reading
Beyond the dawn phenomenon and day-to-day variability, several factors can nudge a fasting glucose result higher or lower than your true metabolic baseline.
Stress is one of the most common confounders. The body responds to physical or emotional stress by releasing hormones like cortisol and adrenaline, which raise blood sugar as part of the fight-or-flight response.10PubMed Central. Stress-Induced Diabetes: A Review If you are anxious about the blood draw itself, or had a terrible night, your reading may be higher than it would be on a calm Tuesday morning.
Sleep deprivation also has a direct effect. Even a few nights of poor sleep can impair glucose metabolism and raise cortisol levels.11PubMed Central. Metabolic, endocrine, and immune consequences of sleep deprivation If your fasting glucose comes back higher than expected and you have been sleeping badly, a retest after a stretch of normal sleep is reasonable before drawing conclusions.
Meal timing the night before matters more than most people realize. Eating dinner late, particularly within a few hours of bedtime, can raise overnight blood glucose. A randomized crossover trial found that eating dinner at 10 p.m. instead of 6 p.m. resulted in a postdinner glucose peak about 18% higher, and while fasting glucose the next morning was similar between the two conditions, glucose after the following day’s breakfast was elevated in the late-dinner group.12The Journal of Clinical Endocrinology & Metabolism. Metabolic Effects of Late Dinner in Healthy Volunteers—A Randomized Crossover Clinical Trial Another study confirmed that eating dinner early improved average blood glucose levels across the nighttime period.13PubMed Central. Eating Dinner Early Improves 24-h Blood Glucose Levels and Boosts Lipid Metabolism after Breakfast the Next Day: A Randomized Cross-Over Trial
Certain medications can also push fasting glucose up. Corticosteroids are well known for this effect, and several classes of commonly prescribed drugs, including some blood pressure medications and statins, can raise glucose levels to varying degrees.14PubMed Central. Medication-Induced Hyperglycemia and Diabetes Mellitus: A Review of Current Literature and Practical Management Strategies If you have started a new medication and your fasting glucose has risen, it is worth raising the question with your doctor rather than assuming you have suddenly become prediabetic.
Capillary Versus Venous Blood and Home Meters
The number you get from a finger-stick glucose meter at home is not identical to the number from a lab blood draw, and the reasons are worth understanding if you track your readings regularly. Lab tests measure glucose in venous plasma (from a vein in your arm), while home meters measure glucose in capillary whole blood (from a fingertip). Most modern home meters apply a correction factor to estimate a plasma-equivalent value, but some discrepancy remains.
Research comparing the two has found that venous plasma readings tend to be slightly higher than capillary readings for fasting samples, while the relationship can reverse after eating.15PubMed. Comparability of venous and capillary glucose measurements in blood One study of healthy individuals found a mean difference of about 0.3 mmol/L (roughly 5 mg/dL) between capillary and venous samples, which was statistically real but not clinically meaningful for most purposes.16PubMed. A Comparison of Venous versus Capillary Blood Samples when Measuring Blood Glucose Using a Point-of-Care, Capillary-Based Glucometer However, the accuracy of home meters also depends on factors like the specific meter’s technology and the oxygen level in the blood sample, which can introduce additional bias.17PubMed Central. Capillary and Venous Blood Glucose Accuracy in Blood Glucose Meters Versus Reference Standards: The Impact of Study Design on Accuracy Evaluations
The practical takeaway: if your home meter says 103 and your lab result says 98, or vice versa, the discrepancy alone is not a reason to panic. Trends from the same device, tested under the same conditions, are more informative than any single reading from either source.
Cardiovascular Risk Within the “Normal” Range
The prediabetes cutoff of 100 mg/dL is not a magic number below which all metabolic risk disappears. Several large studies have found that cardiovascular risk begins rising while fasting glucose is still technically normal. In one study of over 13,000 people, those with fasting glucose in the 95 to 99 mg/dL range had a 53% higher risk of cardiovascular events compared to those below 80 mg/dL, even after adjusting for age, blood pressure, cholesterol, smoking, and body weight.18PubMed Central. Fasting glucose levels within the high normal range predict cardiovascular outcome
Another large study found that once fasting glucose rose above 110 mg/dL, the risk of heart attack and stroke increased by 10 to 20%, and those in the highest glucose categories saw roughly double the risk.19Diabetes Care. Fasting Glucose Level and the Risk of Incident Atherosclerotic Cardiovascular Diseases There is a nuance here, though. At least one study found that the association between high-normal fasting glucose and cardiovascular risk lost its statistical significance after accounting for other risk factors like obesity, high blood pressure, and abnormal cholesterol. The elevated glucose may be a marker traveling alongside those other risks rather than an independent driver on its own.20PubMed. Cardiovascular risk in patients with fasting blood glucose levels within normal range
What this means for you: a fasting glucose of 96 mg/dL is not the same as a fasting glucose of 78 mg/dL, even though both are technically “normal.” If yours is consistently in the upper 90s, it is worth looking at the broader metabolic picture, including waist circumference, blood pressure, triglycerides, and family history, rather than resting easy because the number is under 100.
When Fasting Glucose Is Too Low
Most conversations about fasting glucose focus on numbers that are too high, but low readings deserve attention too. A fasting glucose below about 70 mg/dL (3.9 mmol/L) is generally considered hypoglycemic, and symptoms can include shakiness, sweating, confusion, and irritability. In people without diabetes, true fasting hypoglycemia is uncommon and usually signals an underlying cause that needs investigation, ranging from certain medications and alcohol use to rare conditions like insulin-producing tumors or the aftermath of bariatric surgery.21PubMed Central. Non-Diabetic Hypoglycemia: Evaluation and Management in Adults
For people on insulin or certain oral diabetes medications, low fasting glucose is a more practical concern. Overnight hypoglycemia can occur if the medication dose is too high or if dinner was smaller than usual. This is one reason continuous glucose monitors have become valuable for people on intensive insulin therapy: they can alert you to drops while you sleep rather than waiting for a morning finger stick to discover the problem after the fact.
What Continuous Glucose Monitors Reveal in Healthy People
Continuous glucose monitors were originally designed for diabetes management, but their use in non-diabetic populations has produced some eye-opening data. One study that placed monitors on healthy volunteers with low fasting glucose and normal HbA1c found that 93% of them reached glucose levels above the threshold usually associated with impaired glucose tolerance at some point during the monitoring period, spending a median of about 26 minutes per day above that level. About 9% of these healthy volunteers hit glucose levels above the diabetes diagnostic threshold at some point during monitoring.22PubMed Central. Real-life glycaemic profiles in non-diabetic individuals with low fasting glucose and normal HbA1c: the A1C-Derived Average Glucose (ADAG) study
These transient spikes do not mean those people have hidden diabetes. They illustrate that blood sugar in healthy humans is not a flat line; it oscillates throughout the day in response to meals, activity, and hormonal cycles. What matters more than any single spike is the overall pattern: how much time you spend in a stable range and how quickly glucose returns to baseline after eating. Research on over 4,000 non-diabetic adults found that lower glucose variability and more time spent in a tighter range were associated with better metabolic markers like lower insulin resistance and lower estimated cardiovascular risk.23PubMed Central. Glycaemic variability, assessed with continuous glucose monitors, is associated with diet, lifestyle and health in people without diabetes The same study found that lower carbohydrate intake, less ultra-processed food, and a longer overnight fast were all associated with steadier glucose patterns.
Exercise and Fasting Glucose
If your fasting glucose is in the prediabetes range, exercise is one of the most effective ways to bring it down. A network meta-analysis comparing different types of exercise in people with prediabetes found that resistance training, aerobic exercise, and combinations of the two all significantly reduced fasting blood glucose. Resistance training at a low-to-moderate load showed the largest reduction, averaging roughly 8.6 mg/dL, while moderate-intensity aerobic exercise reduced fasting glucose by about 6.8 mg/dL.24PubMed Central. Exercise training modalities in prediabetes: a systematic review and network meta-analysis Combining aerobic and resistance training was also effective and may offer the broadest metabolic benefits, including improvements in BMI and insulin levels.25PubMed Central. Comparisons of different exercise interventions on glycemic control and insulin resistance in prediabetes: a network meta-analysis
The size of these effects is modest in isolation, but they compound with dietary changes and weight loss. For someone with a fasting glucose of 110 whose goal is to get below 100 without medication, a consistent exercise routine combined with earlier dinners and reduced ultra-processed food intake represents a realistic path. None of these changes need to be extreme; they need to be sustained.
Fasting Glucose During Pregnancy
Pregnancy changes the picture in ways that sometimes confuse both patients and clinicians. In early pregnancy, fasting glucose typically drops below pre-pregnancy levels. This dip is thought to result from increased insulin sensitivity and the dilutional effect of expanded blood volume. A longitudinal study of women with gestational diabetes risk factors found that fasting glucose was substantially lower in early pregnancy compared to later trimesters, and suggested that diagnostic thresholds for early-pregnancy screening might need to be set about 0.7 mmol/L (roughly 13 mg/dL) lower than the standard gestational diabetes cutoffs used later in pregnancy.26PubMed Central. Longitudinal changes in glucose during pregnancy in women with gestational diabetes risk factors
Later in pregnancy, rising levels of placental hormones increase insulin resistance, which is why gestational diabetes screening is typically performed between 24 and 28 weeks. A fasting glucose reading of 92 mg/dL that would be unremarkable outside of pregnancy is actually one of the diagnostic criteria for gestational diabetes when measured during that screening window. The numbers mean something different depending on when in pregnancy they are measured, which is a subtlety that standard glucose charts designed for non-pregnant adults do not capture.