What Is Fasting Blood Glucose and What Do Results Mean?

Fasting blood glucose is the concentration of sugar in your bloodstream after you have not eaten for at least eight hours, and it serves as one of the most common screening tools for diabetes and prediabetes. A normal result falls below 100 mg/dL (5.6 mmol/L), while readings between 100 and 125 mg/dL signal impaired fasting glucose (a form of prediabetes), and anything at or above 126 mg/dL on two separate occasions meets the diagnostic threshold for diabetes. Those numbers seem straightforward, but what determines where your reading lands on any given morning involves a surprisingly dynamic process that extends well beyond what you ate the night before.

How Your Body Keeps Blood Sugar Stable Overnight

When you stop eating, your body does not simply run down its fuel tank to zero. Your liver takes over as the primary glucose factory, releasing sugar into the bloodstream through two processes: breaking down stored glycogen and manufacturing brand-new glucose from non-sugar building blocks like amino acids and lactate.¹ After an overnight fast, roughly a third of the liver’s glucose output comes from that manufacturing pathway, with the rest coming from glycogen breakdown.² Insulin is the main signal that tells the liver to ease off glucose production. When insulin is working properly, it keeps that output in check so your blood sugar stays in a narrow range even after many hours without food.

This is why a fasting blood glucose test is so informative. It captures how well your body self-regulates when there is no food coming in. A healthy reading means insulin is successfully putting the brakes on liver glucose output. An elevated reading suggests those brakes are slipping, and the liver is releasing more glucose than it should, a hallmark of insulin resistance.³

What the Numbers Mean

The ranges used to interpret fasting blood glucose are set by organizations like the American Diabetes Association and the World Health Organization. They are based on decades of population data linking specific glucose levels to long-term health risks. Here is how the categories break down:

Normal: Below 100 mg/dL (5.6 mmol/L). Your body is regulating glucose effectively.
Impaired fasting glucose: 100 to 125 mg/dL (5.6 to 6.9 mmol/L). This falls under the umbrella of prediabetes. Your system is showing early signs of strain but has not crossed into full diabetes.
Diabetes: 126 mg/dL (7.0 mmol/L) or higher on two separate tests. At this level, the liver’s glucose regulation has deteriorated enough to warrant diagnosis.

Those thresholds have shifted over time. The impaired fasting glucose category was originally created in the late 1990s with a lower boundary of 110 mg/dL (6.1 mmol/L). In 2003, the American Diabetes Association dropped it to 100 mg/dL (5.6 mmol/L) to catch more people at risk.⁴ That change was controversial. The European Diabetes Epidemiology Group argued that the lower cutoff dramatically increased the number of people classified as prediabetic without clear evidence that those in the 100-to-109 range faced meaningfully higher risk. Different professional bodies still disagree on whether 100 or 110 is the better starting point, which is worth knowing if your result sits in that borderline zone.

Why the Same Person Can Get Different Results

If you have ever tested your fasting glucose on consecutive mornings and gotten noticeably different numbers, you are not imagining things. Research on newly diagnosed type 2 diabetic patients found that fasting glucose values varied by roughly 15 percent from day to day, with about 14 percent of that variation stemming purely from normal biological fluctuation.⁵ The higher your baseline glucose level, the bigger the absolute swings tend to be. Someone averaging 130 mg/dL might see readings ranging from about 110 to 150 on any given morning, even without any change in diet or medication.

This variability is one reason clinicians do not diagnose diabetes on a single reading. A fasting glucose of 128 mg/dL one morning could reflect a genuinely elevated baseline, or it could be a high day on the fluctuation curve for someone whose average sits in the prediabetic range. Confirmation with a second test, or with a different test altogether, is standard practice.

The Dawn Phenomenon and Morning Glucose Spikes

Your body’s internal clock has its own opinions about blood sugar. In the early morning hours, hormones like cortisol and growth hormone surge as part of the normal waking process. These hormones push the liver to ramp up glucose production, a pattern researchers call the dawn phenomenon. In people with type 1 diabetes, studies have documented a substantial early morning fasting glucose rise, averaging around 123 mg/dL with wide individual variation, that directly predicted how high blood sugar would go after breakfast.⁶

The dawn phenomenon is not limited to type 1 diabetes. In type 2 diabetes, the central circadian clock in the brain regulates daily rhythms of liver glucose production and how sensitive the liver is to insulin. Disruption of that clock is associated with an extended dawn phenomenon, meaning elevated morning glucose that is not caused by low blood sugar during the night.⁷ Even people without diabetes can experience a milder version. If your fasting glucose consistently reads higher first thing in the morning than it does after a similar fast during the afternoon, circadian biology is likely the explanation.

Factors That Can Throw Off Your Reading

Beyond normal biological variability and circadian rhythms, several practical factors can push a fasting glucose result higher or lower than your “true” baseline.

Sleep is one of the more underappreciated influences. A study of shift workers found that men with insufficient sleep who also worked irregular hours had roughly three times the odds of impaired fasting glucose compared to those getting adequate sleep.⁸ Sleep deprivation raises cortisol and reduces insulin sensitivity, both of which nudge morning glucose upward. A single terrible night’s sleep probably will not push a normal person into the prediabetic range on a one-off test, but chronic short sleep can shift your average reading meaningfully.

Medications are another common culprit. Multiple drug classes can raise blood glucose through various pathways. Corticosteroids like prednisone are well known for this, but the list also includes certain antipsychotics, some blood-pressure medications (particularly thiazide diuretics and beta-blockers), immunosuppressants, and protease inhibitors used in HIV treatment.⁹ If you are starting a new medication and notice your fasting glucose creeping up, the drug itself is worth investigating before assuming your metabolism has changed.

Stress, illness, and even intense exercise the day before can also skew results. Acute physical or emotional stress triggers a hormonal cascade that temporarily raises blood sugar. For the most representative reading, the standard preparation is simple: nothing but water for eight to twelve hours beforehand, no unusually late or heavy meals the prior evening, and ideally a normal night of sleep.

Home Meters Versus Lab Tests

If you own a glucose meter, you may have noticed that its readings do not always match what the lab reports. That discrepancy is real and well documented. A large review spanning over 25 years of data found that home meter readings differed from laboratory venous blood glucose by an average of about 11 mg/dL, with a mean relative difference of roughly 8 percent.¹⁰ The variation between any individual paired reading can be much larger than that average.

Part of the gap comes from the type of blood being measured. Home meters use capillary blood from a fingerstick, while labs typically measure venous plasma. Those are genuinely different samples. Oxygen levels in venous blood can also interfere with some meter designs, introducing additional bias depending on the specific device.¹¹ Newer meters designed to be insensitive to oxygen variation perform better on this front, but no consumer meter matches laboratory precision.

The practical takeaway: home meters are useful for tracking trends over time, especially if you are managing diabetes or monitoring your response to dietary changes. They are less reliable for making diagnostic decisions on a single reading. If a home reading puts you on the border between normal and prediabetic, a lab-drawn venous sample is the definitive test.

How Fasting Glucose Compares to Other Diagnostic Tests

Fasting blood glucose is not the only way to screen for diabetes. The two main alternatives are the hemoglobin A1c test (HbA1c), which estimates your average blood sugar over the previous two to three months, and the oral glucose tolerance test (OGTT), which measures how efficiently your body clears a large glucose load. Each captures a somewhat different aspect of glucose metabolism, and they do not always agree.

A meta-analysis comparing the screening performance of HbA1c and fasting glucose found no meaningful difference in their overall ability to detect type 2 diabetes.¹² But “overall performance” hides a real-world problem: the two tests sometimes identify different people. One study found that fasting glucose at the diabetes cutoff had a sensitivity of about 60 percent compared with HbA1c, meaning it missed a substantial fraction of people that HbA1c flagged, and vice versa.¹³

The oral glucose tolerance test tends to catch cases that both fasting glucose and HbA1c miss, especially in certain clinical populations. Among patients hospitalized with acute coronary syndromes, the majority of those found to have type 2 diabetes through an OGTT would not have been detected by fasting glucose or HbA1c alone.¹⁴ Similarly, in a study of young people with clinical insulin resistance, a notable number with prediabetes or diabetes were identified only by the OGTT, not by fasting glucose criteria.¹⁵

The OGTT is more cumbersome (it requires drinking a glucose solution and waiting two hours in a clinic), which is why fasting glucose and HbA1c are used far more frequently in routine screening. But if your fasting glucose keeps coming back borderline normal while you have risk factors for diabetes, asking about an OGTT is reasonable. It tests your body’s ability to handle glucose under load, something a fasting test cannot evaluate.

Why Prediabetic Numbers Matter More Than People Think

A fasting glucose of, say, 108 mg/dL often gets dismissed as “just a little high.” But the prediabetic range carries its own health consequences well before diabetes develops. Cardiovascular risk begins rising at glucose levels below the diabetes cutoff. Data from large cohort studies have shown that the prediabetes state is associated with a nearly threefold higher prevalence of unrecognized heart attack compared with normal glucose tolerance, and that even modest increases in average blood sugar within the normal range are linked to higher cardiovascular mortality over the following decade.¹⁶

The good news is that prediabetes is also the stage most responsive to intervention. Lifestyle changes, including modest weight loss, increased physical activity, and dietary adjustments, consistently bring fasting glucose back into the normal range. One trial found that both structured meal replacement and standard dietary counseling lowered fasting glucose by about 11 to 12 mg/dL over six weeks in people with impaired fasting glucose, returning their average levels to normal.¹⁷ Diet composition itself can make a difference. An eight-week study comparing dietary patterns in people with impaired fasting glucose found that a higher-carbohydrate, lower-fat diet actually reduced fasting glucose and improved insulin secretion more than a lower-carbohydrate, higher-fat alternative.¹⁸ That may seem counterintuitive given the popularity of low-carb diets for blood sugar management, but the relationship between carbohydrate intake and fasting glucose is more nuanced than the popular narrative suggests. Carbohydrate quality, fiber content, and overall caloric balance all factor in.

When Fasting Blood Sugar Is Too Low

Most conversations about fasting glucose focus on values that are too high, but abnormally low readings deserve attention too. Hypoglycemia is classically defined by three criteria occurring together: a low plasma glucose concentration, symptoms like shakiness, sweating, confusion, or lightheadedness, and resolution of those symptoms once glucose returns to normal.¹⁹ In people who do not have diabetes, fasting hypoglycemia (as opposed to the reactive kind that happens a few hours after eating) can point toward a range of causes including certain medications, excessive alcohol intake, adrenal insufficiency, liver disease, or rarely insulin-producing tumors. A single low reading in an otherwise healthy person who feels fine is usually not concerning, but recurrent fasting glucose values below 70 mg/dL paired with symptoms warrant investigation.

Fasting Glucose in Pregnancy and Childhood

The standard adult reference ranges do not apply universally. During pregnancy, normal fasting glucose runs lower than in the non-pregnant state, driven by continuous glucose transfer to the fetus and hormonal changes that shift insulin dynamics across trimesters. Gestational diabetes screening uses its own set of thresholds, which are substantially lower than the general diabetes cutoff. If you are pregnant and your fasting glucose is 95 mg/dL, that may flag an issue on a gestational diabetes screen even though it would be perfectly normal outside of pregnancy.

Children present a different challenge. Fasting glucose, insulin levels, and measures of insulin resistance all shift during normal childhood development, which means adult cutoffs cannot simply be applied to kids.²⁰ A fasting glucose of 98 mg/dL in an eight-year-old carries different implications than the same number in a forty-year-old. Pediatric reference values that account for age and developmental stage are essential for accurately identifying glucose problems in younger populations, and unfortunately, many general screening programs still rely on adult ranges by default.

What to Do With a Borderline Result

If your fasting glucose comes back in the 100 to 125 range, resist the urge to dismiss it as meaningless or to panic. A single borderline result needs context. Was the test drawn under ideal conditions (adequate fasting, reasonable sleep, no acute illness)? Was it a lab draw or a home meter? If you used a home meter, remember that an 8 to 11 percent margin of error could easily place a true value of 94 into the impaired range on paper, or a true value of 105 into the normal range.

If the reading is confirmed on a second lab test, the evidence favors acting on it. Given the day-to-day variability discussed earlier, your doctor may also order an HbA1c to get a longer-term view, or an OGTT if there is reason to suspect your body handles glucose loads poorly despite acceptable fasting numbers. The combination of tests paints a fuller picture than any single number.

For people confirmed in the prediabetic range, the evidence supports starting with lifestyle changes: regular physical activity, moderate weight loss if overweight, and attention to diet quality. These interventions have been shown repeatedly to pull fasting glucose back into the normal range and, more importantly, to reduce the risk of progressing to type 2 diabetes. The window between first noticing a borderline fasting glucose and developing full diabetes is typically years long, which is both a comfort and an opportunity. It means there is time to act, but also time to procrastinate until the opportunity shrinks.