Glucose - 90 Minute Response

Plenty of people with a normal fasting blood sugar already handle a real load of sugar poorly, and a fasting draw won't show it. The 90-minute reading comes late in a sugar challenge, after the first surge, when a healthy body should be bringing the level back down.

If your number is still high at that point, or still climbing, your insulin response may be running slow. It's one point on a longer curve, and a less-studied one than the 1-hour draw beside it, so it reads best next to the readings around it.

What the 90-Minute Reading Captures

This value comes from an OGTT (oral glucose tolerance test). You arrive fasting, have blood drawn, drink a measured dose of sugar, and get drawn again at set times. The standard clinical version samples only at fasting and 2 hours. Extended versions add draws at 30, 60, and 90 minutes, mostly in research settings. This test reports the glucose in your plasma, the liquid part of your blood, from the draw taken 90 minutes after the drink. The result comes in mg/dL, the standard US unit for blood sugar.

Several organs set that number. Your gut absorbs the sugar, your pancreas releases insulin, your liver stops making its own sugar, and your muscles and fat take up what's circulating. By 90 minutes the absorption peak is usually behind you, so the reading leans heavily on how much insulin you made and how well it's working.

Where that system breaks down tends to change which part of the curve stays high. In about 700 adults newly diagnosed with type 2 diabetes, insulin resistance in the liver tracked fasting and early after-meal sugar, while failing insulin-producing cells in the pancreas tracked both fasting and later after-meal sugar. In about 200 adults studied after a high-fat meal, people whose insulin resistance sat mainly in the liver had a less favorable pattern of blood chemistry than people whose resistance sat mainly in muscle.

Shape Matters as Much as Height

Because the test samples several points, it shows the shape of your response, and the shape carries its own information. In about 125 people with mildly raised fasting sugar, those whose sugar peaked later in the test had higher cardiovascular risk scores and more diabetes. In about 570 people with type 2 diabetes, a curve that rose once and fell once was tied to more large-artery complications such as heart attack and stroke.

In more than 3,000 adults with early type 2 diabetes, a curve that kept rising through the test reflected weaker insulin production, and curve shape differed by sex, race, age, and body weight. A 90-minute value that sits above your 60-minute value is how a late or continuing rise shows up on your report.

Some findings point the other way at first glance. In about 7,500 Japanese adults, people whose 30-minute sugar dipped below their fasting level were mostly young, healthy women at low risk of diabetes. In 36 healthy older adults, a curve that fell and then rose a second time went with more insulin, more GLP-1, and lower 1-hour sugar. GLP-1 is a hormone from your gut that boosts insulin release.

So a small second bump at 90 minutes after a deep early drop is a different signal from a 90-minute value that never came down. Both can produce the same number. The difference is in the readings on either side, which is why this result should always be read as part of your whole curve.

Heart Disease and Early Death

Most outcome research on the middle of the curve has studied the 1-hour or 30-minute draw. No medical society has set a cutoff for the 90-minute value, and no outcome study has tested it on its own. Treat what follows as evidence about neighboring points on the same curve, which share the same biology.

Those neighboring points predict trouble the 2-hour value can miss. In about 980 Finnish adults followed long term, 1-hour sugar predicted heart disease and death better than fasting or 2-hour sugar and improved standard heart-risk models. In about 860 older Chinese men without diabetes followed for 20 years, 1-hour sugar beat 2-hour sugar for predicting cardiovascular events and death from any cause.

In about 5,900 Danish adults, distinct curve shapes predicted diabetes, heart disease, and death differently, and a pattern with a high 30-minute value carried extra risk beyond fasting and 2-hour levels. In about 1,600 Japanese adults whose glucose tolerance looked normal, 1-hour sugar still predicted death from heart disease and cancer.

The pattern also holds after ordinary meals. Among about 530 people with type 2 diabetes, blood sugar after eating predicted cardiovascular events better than fasting sugar, and the link was stronger in women. Two large pooled analyses found that prediabetes, including an abnormal 2-hour result on this same test, raises the risk of heart disease and death from any cause.

What this means for you: the 1-hour value is the middle point with real standing. In 2024 the International Diabetes Federation endorsed a 1-hour cutoff for diagnosing prediabetes and type 2 diabetes, citing decades of data that it predicts diabetes and its complications better than fasting sugar, the three-month average, or the 2-hour value. The 2023 European Society of Cardiology guideline was more cautious and called for further validation, and no treatment trial has yet enrolled people based on the 1-hour value. A raised 1-hour reading is still worth acting on even when your 2-hour number clears the diagnostic line. A high 90-minute value points the same way, but it hasn't been validated, so weigh it alongside your 1-hour result.

Type 2 Diabetes and Insulin Resistance

A high or slow-falling 90-minute value can point toward insulin resistance or a pancreas that can't keep up. Insulin resistance means your cells respond weakly to insulin, so it takes more of it to clear the same amount of sugar.

Among more than 5,000 Finnish adults without diabetes, the insulin-resistant ones showed the same unfavorable shifts in blood fats and other chemistry after the sugar drink as people with type 2 diabetes. In about 1,250 Asian adults without diabetes, 1-hour sugar predicted type 2 diabetes as well as the 2-hour value and was more closely tied to failing insulin production. Among older women, frailty went with exaggerated and prolonged sugar and insulin responses.

The curve picks up risk earlier in life too. In about 200 young people with obesity and normal results, a raised 1-hour value predicted progression to prediabetes. Among more than 5,000 relatives of people with type 1 diabetes who carried warning antibodies, a 1-hour test predicted type 1 diabetes as accurately as a 2-hour test.

Who Gets the Most From the Full Curve

The extra time points are most useful in groups where fasting sugar and the three-month average test tend to miss problems.

Sources: Gupta et al.; Lin et al. (Scientific Reports); Guerreiro et al.

What this means for you: if you have polycystic ovary syndrome, carry significant extra weight, are over 60, or have diabetes in your family, a normal fasting sugar isn't a clean bill of health. A full curve that includes the 1-hour point is the more complete picture, and the 90-minute point adds detail on how your sugar comes down.

When Results Can Be Misleading

Single readings of metabolic markers like blood sugar vary enough within the same healthy person that one draw can't fully characterize you. The 90-minute value also depends heavily on how you prepared, and these are the factors most likely to throw it off.

• Something before the drink: in one small study of 17 adults with obesity, the zero-calorie sweetener sucralose taken 10 minutes before a sugar load raised peak sugar and drove insulin output up by about 20%; other sweetener studies have found mixed effects. Protein shakes and ketone drinks taken before a meal lowered after-meal sugar, so arrive having had only water.
• Time of day: the same carbohydrate-rich meal produced higher after-meal sugar at dinner than at breakfast. Take the test in the morning, and at the same time each time you repeat it.
• Exercise before or during the test: a single session of resistance exercise before a meal improved after-meal sugar handling in men with prediabetes. Skip the morning workout and sit quietly between draws.
• Medication timing: in people with type 2 diabetes, metformin taken 30 to 60 minutes before sugar delivered straight into the gut lowered sugar more than the same dose taken at the time of the sugar. Keep your usual timing the same for every test so results stay comparable.

Sample handling matters too. If blood sits before the plasma is separated, the blood cells keep using up sugar and the result can read falsely low. A standardized sugar drink also isn't a meal: a mixed meal raises sugar more gradually and to a lower peak, so a home reading after dinner can't be compared with this test.

Tracking Your Curve Over Time

A single curve tells you where you stand today. A series tells you which way you're heading, and that's the more useful information. Research on the 1-hour value suggests the middle of the curve shifts before the 2-hour and fasting values do. A 90-minute reading drifting upward across tests is plausibly part of the same early shift, though that hasn't been studied directly, so watch it together with your 1-hour value.

When you repeat it, keep everything else identical: same lab, same drink, same overnight fast, same morning time, same draw schedule. Compare shapes as well as numbers. A 90-minute value that used to fall below your 60-minute value and now sits above it is a meaningful change, even if neither number crossed a flag.

Expect some built-in differences between people. In about 1,150 Japanese adults, young women handled the sugar drink better than other groups, and that advantage faded by middle age. Your own trend line is a better reference than anyone else's numbers.

What to Do With an Out-of-Pattern Result

If your 90-minute value is high or still rising while your fasting and 2-hour values look fine, start by repeating the curve under clean conditions with insulin measured at every point, and make sure the 1-hour draw is included. Seeing how much insulin it took to produce your sugar curve separates early insulin resistance from a pancreas that's starting to fall behind. Add the three-month average blood sugar test, fasting insulin, and a lipid panel with triglycerides, since high triglycerides often travel with insulin resistance.

If the 2-hour or fasting values are also raised, you're moving into the range where a diabetes diagnosis gets made, and an endocrinologist is worth involving, especially if a prescription is on the table. The same goes for a curve that worsened sharply over a few months, which can happen after starting certain psychiatric medications.

One caution about weighting a single abnormal result. In about 5,800 British adults, people diagnosed with diabetes on the sugar-drink test alone, while their three-month average stayed normal, had heart and kidney risks similar to people without diabetes. A pooled review also found that screening whole populations and treating what turns up has only a modest effect on diabetes rates.

Neither finding means you should ignore your own result. Population averages blend people who were never going to progress with people who were. A repeat curve that stays high, especially with high insulin or rising triglycerides alongside it, is a consistent pattern, and a consistent pattern is what justifies acting now instead of waiting for the fasting number to catch up.