Fasting insulin test · Insulin resistance

Fasting Insulin Screening: The Gate on the Brain’s Second Fuel

Fasting insulin screening finds insulin resistance years before glucose moves: in Whitehall II, insulin sensitivity was already lower across the 13 years before a diabetes diagnosis, while fasting glucose accelerated only in the final 3. An elevated result is worth acting on; a normal one is not a clearance.

The brain has no gate for ketones. The liver does, and insulin controls it. That makes fasting insulin, read as a number, one of the earliest metabolic signals a practice can capture and document.

Fasting insulin screening is still unusual in adult primary care, where the metabolic workup is organized around glucose and HbA1c. Chapter 2 of The Starved Brain, presented in the video Your Brain Was Built for Ketones, Not Sugar, argues from physiology that the human brain was built to use fat-derived fuel. For the clinician, that argument has a concrete screening consequence: whether a patient can reach that second fuel at all is decided by insulin, and insulin moves years before glucose does.

Measura [Cardiometabolic and Autonomic Health Analysis] is a measurement service. It reports to the ordering physician, who owns interpretation and management.

How does insulin control the brain’s access to ketones?

Cerebral ketone uptake is concentration-driven. The Sherbrooke group describes proportionality over at least 0.02–12 mM, while glucose entry is regulated by neuronal demand. When ketones were infused to 2.16 mM in healthy adults, cerebral ketone uptake rose from 1.11 to 5.60 µmol/100 g/min and glucose metabolism fell from 25.8 to 17.2 despite adequate glucose supply. The brain accepts whatever ketone load the circulation presents.

The limiting step is hepatic. Ketogenesis requires low insulin, and in nutritional ketosis the small amount of insulin still present is what keeps production regulated. A hyperinsulinemic patient therefore has a closed liver regardless of what the brain could use. That reframes insulin resistance from a diabetes-risk variable into a fuel-access variable, with the brain as a downstream stakeholder.

Why is glucose a late marker of insulin resistance?

Whitehall II followed 6,538 British civil servants. The 505 who progressed to diabetes had insulin sensitivity 34.2% lower and compensatory secretion 10.4% higher across the entire 13 years before diagnosis, while fasting glucose accelerated only in the final 3. A screening strategy anchored on glucose finds these patients a decade late.

Fasting insulin has its own limit. In the Kraft database, just over half of 4,185 people with normal glucose tolerance were hyperinsulinemic only by dynamic pattern, and the authors concluded fasting insulin had limited value in finding them. The error runs in one direction: a single fasting value under-detects hyperinsulinemia; it does not over-call it. An elevated result is therefore worth acting on, and a normal one is not a clearance.

What fasting insulin level is too high?

In 965 patients from one endocrine practice, mostly white, obese and female, a fasting insulin above 9.0 µIU/mL identified 80% of prediabetes. Dr. Padda’s clinic threshold is a fasting insulin above 10 µIU/mL, his own cutoff rather than a guideline, and in his patients that is where damage is already under way. LP-IR, a 0–100 score from six NMR lipoprotein measures, correlates r = 0.51 with HOMA-IR in 4,972 MESA participants; in 5,314 MESA adults followed 7.7 years, the top quartile carried a hazard ratio of 3.28 for diabetes, 1.93 after adjustment for HOMA-IR and glucose. Three of the four authors who built the score worked for its manufacturer, and no study has yet tied it to brain volume. The tier is prospective cohort data plus single-practice thresholds, which is enough to screen on and not enough to call settled.

Two biological drivers keep these patients invisible: compensatory hyperinsulinemia holding glucose in range, and hepatic fuel switching that never gets exercised. The third is behavioral and economic. Carbohydrate supplied 50.5% of energy for U.S. adults in the 2016 NHANES cycle, so the average patient never leaves the fed state long enough for anyone to observe whether the switch still works.

How should ketone results be read alongside insulin?

Patients who pursue carbohydrate restriction will bring ketone values. The frame that prevents errors is insulin presence, not ketone magnitude. Nutritional ketosis runs roughly 0.5–3 mmol/L with normal glucose and pH. The Joint British Diabetes Societies definitions put the emergency beyond 3.0 mmol/L with acidosis, and ketoacidosis can reach 20 mmol/L. The 2022 guideline added a section on SGLT2 inhibitors because they raise the risk of ketoacidosis at normal glucose. In that population a normal glucose is the reason to check ketones, and any plan to restrict carbohydrate belongs with the prescriber before it starts.

Pregnancy needs its own caution. The cohort behind the guidance to avoid ketones, 223 women of whom eighty-four percent had diabetes, found weak inverse correlations with offspring scores, and a 2021 review called the evidence conflicting and inconsistent. Harm is unproven and so is safety; the physiology described in the book is not a pregnancy recommendation.

Dr. Padda applied the same discipline to his own material, correcting an early claim that newborns run substantially on ketones once the human tracer data showed a smaller share than the rat studies suggested. Screening deserves that standard too.

Who should get a fasting insulin test, and what does a result change?

• Adults with central adiposity or other features of insulin resistance whose glucose and HbA1c are still normal.
• Patients planning significant carbohydrate restriction, especially those on insulin or SGLT2 inhibitors.
• Patients with cognitive complaints, where brain fuel supply is part of the differential.

A documented elevated insulin changes follow-up: it justifies a shorter repeat interval, prompts bioimpedance body composition to separate fat gain from muscle loss, supports indirect calorimetry to replace an estimated resting metabolic rate with a measured one, and, where memory or attention is a concern, a scored cognitive assessment baseline. Selection triggers are laid out in selection criteria, and the rationale in clinical rationale.

Workflow and documentation

Standing orders let staff route qualifying patients into laboratory panels without a separate decision at every visit. Results need to land where they will be seen again; getting diagnostic results into the chart covers the mechanics. Serial values support chronic care follow-up and HEDIS and value-based care documentation, described in HEDIS and value-based care. The full study set, with every limit named, is in the Chapter 2 companion to The Starved Brain by Dr. Gurpreet Singh Padda, MD, MBA, MHP; the prior installment covers pairing cognitive screening with metabolic data.

Frequently asked questions

Why screen fasting insulin when HbA1c is normal?

Because insulin compensates for years before glucose fails. In Whitehall II, people who later developed diabetes had lower insulin sensitivity and higher secretion across all 13 years before diagnosis, while glucose accelerated only in the last 3. HbA1c records the failure of compensation; insulin records the compensation itself. The glucose-late timeline and what it means for dementia screening is in insulin resistance and dementia screening.

What fasting insulin value should prompt follow-up?

Published data put a useful line near 9.0 µIU/mL, which identified 80% of prediabetes in one practice of 965 patients. Dr. Padda’s clinic uses above 10 µIU/mL as its own cutoff. Because a single fasting value under-detects hyperinsulinemia, an elevated result deserves action and a normal one is not reassurance. See interpreting the report.

How should ketone values be read in patients on SGLT2 inhibitors?

Always with glucose, and never with a normal glucose taken as reassurance, since these agents carry a risk of ketoacidosis at normal blood sugar. Emergency definitions begin above 3.0 mmol/L with acidosis. Any carbohydrate restriction in these patients should be discussed with the prescriber before it begins. The ketone ranges and the SGLT2 exception are laid out for patients in ketosis vs ketoacidosis.

Which measurements pair best with an elevated insulin result?

Body composition shows whether the metabolic strain is accompanied by fat gain, sarcopenic wasting or both. Measured resting metabolic rate replaces a formula estimate. A cognitive baseline adds the brain side where symptoms warrant it. Together they turn one laboratory value into a trackable profile. Why a measured rate replaces the formula is covered in measured versus estimated metabolic rate.

How do results get back to the ordering physician?

Results are reported to the ordering physician, who retains interpretation and management. The practical work is making sure values are filed as discrete, searchable data so serial insulin results can be trended rather than rediscovered at each visit. See getting results into the record for the options.

What does a fasting insulin level tell you?

It shows how hard the pancreas is working to hold blood sugar in range, a strain that glucose and HbA1c do not reveal until late. High insulin also keeps the liver from making ketones, so the number decides whether the brain can reach its second fuel. A single fasting value can miss high insulin but does not falsely flag it, so an elevated result is worth acting on.

Why don’t doctors routinely check fasting insulin?

Adult primary care builds the metabolic workup around glucose and HbA1c, so fasting insulin is still an unusual order. A single fasting value also has a known limit: in one large database, just over half of people with normal glucose tolerance showed high insulin only in a dynamic pattern. That limit runs one way, though. It misses some cases; it does not over-call them.

Is a fasting insulin test the same as an A1c test?

No. HbA1c records the failure of compensation: it moves late, once the pancreas can no longer hold glucose in range. Fasting insulin records the compensation itself, meaning how hard the pancreas is working while glucose and HbA1c still look normal. In Whitehall II, insulin sensitivity was already lower across the 13 years before a diabetes diagnosis, while fasting glucose accelerated only in the final 3.

References

• Tabák, A. G., Jokela, M., Akbaraly, T. N., Brunner, E. J., Kivimäki, M., & Witte, D. R. (2009). Trajectories of glycaemia, insulin sensitivity, and insulin secretion before diagnosis of type 2 diabetes: An analysis from the Whitehall II study. The Lancet, 373(9682), 2215-2221. https://doi.org/10.1016/S0140-6736(09)60619-X
• Crofts, C., Schofield, G., Zinn, C., Wheldon, M., & Kraft, J. (2016). Identifying hyperinsulinaemia in the absence of impaired glucose tolerance: An examination of the Kraft database. Diabetes Research and Clinical Practice, 118, 50-57. https://doi.org/10.1016/j.diabres.2016.06.007
• Johnson, J. L., Duick, D. S., Chui, M. A., & Aldasouqi, S. A. (2010). Identifying prediabetes using fasting insulin levels. Endocrine Practice, 16(1), 47-52. https://doi.org/10.4158/EP09031.OR
• Mackey, R. H., Mora, S., Bertoni, A. G., Wassel, C. L., Carnethon, M. R., Sibley, C. T., & Goff, D. C. (2015). Lipoprotein particles and incident type 2 diabetes in the Multi-Ethnic Study of Atherosclerosis. Diabetes Care, 38(4), 628-636. https://doi.org/10.2337/dc14-0645
• Shalaurova, I., Connelly, M. A., Garvey, W. T., & Otvos, J. D. (2014). Lipoprotein insulin resistance index: A lipoprotein particle-derived measure of insulin resistance. Metabolic Syndrome and Related Disorders, 12(8), 422-429. https://doi.org/10.1089/met.2014.0050
• Cunnane, S. C., Courchesne-Loyer, A., Vandenberghe, C., St-Pierre, V., Fortier, M., Hennebelle, M., Croteau, E., Bocti, C., Fulop, T., & Castellano, C.-A. (2016). Can ketones help rescue brain fuel supply in later life? Implications for cognitive health during aging and the treatment of Alzheimer’s disease. Frontiers in Molecular Neuroscience, 9, Article 53. https://doi.org/10.3389/fnmol.2016.00053
• Hasselbalch, S. G., Madsen, P. L., Hageman, L. P., Olsen, K. S., Justesen, N., Holm, S., & Paulson, O. B. (1996). Changes in cerebral blood flow and carbohydrate metabolism during acute hyperketonemia. American Journal of Physiology, 270(5 Pt 1), E746–E751. https://doi.org/10.1152/ajpendo.1996.270.5.E746
• Rizzo, T., Metzger, B. E., Burns, W. J., & Burns, K. (1991). Correlations between antepartum maternal metabolism and intelligence of offspring. New England Journal of Medicine, 325(13), 911–916. https://doi.org/10.1056/NEJM199109263251303
• Tanner, H. L., Dekker Nitert, M., Callaway, L. K., & Barrett, H. L. (2021). Ketones in pregnancy: Why is it considered necessary to avoid them and what is the evidence behind their perceived risk? Diabetes Care, 44(1), 280–289. https://doi.org/10.2337/dc20-2008
• Dhatariya, K. K., & Joint British Diabetes Societies for Inpatient Care (2022). The management of diabetic ketoacidosis in adults—An updated guideline from the Joint British Diabetes Society for Inpatient Care. Diabetic Medicine, 39(6), Article e14788. https://doi.org/10.1111/dme.14788

Related reading

• Cognitive Screening in Primary Care: Add the Metabolic Picture
• Insulin Resistance and Dementia: Screening Past the Glucose
• Laboratory Panels

Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP, medical director of Measura. Last reviewed September 12, 2026.