The Fasting Insulin Test: Why It Matters for Nerve Pain

A fasting insulin test measures the hormone your pancreas is pushing out to keep blood sugar normal. That matters because blood sugar can look fine for more than a decade while insulin climbs, and the nerves that carry pain are among the first tissues to pay.

The video for Chapter 4 of The Pained Brain, The Hormone Nobody Measured Is Starving Your Nerves, walks through the mechanism. The book is by Dr. Gurpreet Singh Padda, MD, MBA, MHP, and Dr. KrisJay Fucanan, MD. This page covers what the video could not: what the number means, where the test is weak, and what the nerve evidence shows once you look past glucose. The metabolic background is in how high insulin makes you hurt.

What a fasting insulin test shows that glucose does not

In the Whitehall II cohort of British civil servants, people who later developed diabetes were already 34.2 percent less sensitive to insulin in the years 13 to 5 before diagnosis, while their fasting glucose drifted only from 5.47 to 5.79 mmol/L. Their pancreas overproduced the whole time, and its output collapsed only near diagnosis. The gland strained for a decade while the sugar barely moved.

Pooled across nine European cohorts, 3,309 people who went on to develop diabetes were compared with 13,963 matched controls. One to six years before diagnosis, the insulin-resistant subgroup carried a fasting insulin-to-glucose ratio 88 percent above controls after age, sex and body mass were adjusted. On the reference clamp test in 80 hospital patients, normal glucose with high insulin looked the same as impaired tolerance and new diabetes, with glucose disposal of 6.23, 6.37 and 6.19 mg/kg/min.

Reading your number, and its limits

Laboratory reference ranges are not outcome thresholds. In 21,684 screened-healthy Brazilian adults, median fasting insulin was about 6.6 microU/mL and the reference interval topped out at 13.14. In 4,942 Tehran adults followed for 9.2 years, the level that best predicted diabetes was 9.16 in men and 11.13 in women. A result printed as within normal limits can still sit above the line that predicted disease.

The test has limits, and I would rather say so than oversell it. The same samples read 1.8-fold apart across eight commercial assays, and in 2025 only one of twelve immunoassays agreed fully with mass spectrometry. Among 26,976 adults without diabetes, fasting insulin alone was a borderline predictor of death at 1.13, while the highest HOMA-IR category, which pairs insulin with glucose, reached 1.34. Repeated values on the same assay are the ones worth reading.

That measurement gap is why the test is rarely ordered. A 2010 workgroup concluded there were no criteria to classify an individual as insulin resistant, and it laid out a sequence: standardize the assay, let research set cutoffs, then write guidelines. The first step never finished. Current diagnostic standards define prediabetes by glucose and HbA1c alone and name no fasting insulin recommendation. With no standard there is no guideline, and with no guideline nobody is prompted or paid to ask.

How high insulin reaches the nerve

Insulin opens blood vessels through nitric oxide and tightens them through endothelin. In insulin resistance, the opening side goes deaf while the tightening side keeps responding, so more insulin means a narrower vessel. In 40 obese adults without diabetes, the most insulin-resistant third dilated their arteries 9.2 percent against 18.0 percent in the least resistant third.

Inside the nerve, the small vessels narrow early. In a Swedish study with repeat nerve biopsies eleven years apart, men who later developed diabetes already had capillary channels of 5.5 square micrometers against 22.8 in men who stayed healthy, while their glucose was still normal or only impaired. The key comparison was tiny, a signal rather than a rate.

Insulin also acts on the pain nerve directly, though that evidence comes from laboratory and animal work, not people. In rat sensory neurons, insulin raised acid-sensing channel currents by 46.2 percent and worsened pain behavior through peripheral insulin receptors. Whether that happens at the insulin levels a hyperinsulinemic patient carries has not been tested. In human volunteers, the picture is cleaner on inflammation: raising insulin while glucose was held steady increased interleukin-6 messenger RNA in fat elevenfold at two hours.

Prediabetes neuropathy and the tests that miss it

Nerve damage does not wait for a diagnosis. In a Michigan weight-management cohort, obese participants with normal glucose and no clinical neuropathy already had lower small nerve fiber density in the leg than lean controls, 4.5 against 6.4 fibers per millimeter. Among people with severe obesity, waist separated those with neuropathy from those without, 139.3 against 129.1 centimeters, while body mass index did not.

How you test decides what you find. Across 29 studies, reported neuropathy prevalence in prediabetes ranged from 2 percent to 77 percent, with the highest estimates from small-fiber and multimodal methods. In 28 people with prediabetes at Heidelberg, a full sensory battery found nerve damage in 71 percent, and more than half showed hyperalgesia: nerves turning up, not down. A normal nerve conduction study may simply be the wrong test for small fibers, which is what nerve testing can and cannot measure.

Not every cohort agrees. In Olmsted County cohorts, prediabetes carried no excess of neuropathic symptoms or sensory changes, and one large study found metabolic syndrome was not independently associated once prediabetes entered the model. The evidence leans one way, and it is a majority, not a consensus. Where a non-glucose marker has been measured, the body’s own insulin output stands out: among 5,249 Danes with newly diagnosed type 2 diabetes, a high C-peptide carried a prevalence ratio of 1.72 for possible neuropathy, larger than a high HbA1c at 1.42.

The loop: pain and its treatments push insulin up

Pain itself raises insulin resistance. Thirty minutes of painful skin stimulation cut glucose uptake 22 percent in ten healthy young men. Two weeks of prednisolone lowered insulin sensitivity in a dose-dependent way and cut insulin-stimulated capillary recruitment by up to 17 percent. In national survey data, adults using prescription opioids averaged a fasting insulin of 15.5, above the reference range, though that comparison was unadjusted and may reflect who takes opioids rather than the drug. When I use a steroid injection as a bridge, I want the patient’s insulin already in hand. That loop between pain and metabolism is why chronic pain costs years, not only comfort.

What lowers the insulin and regrows the nerve

Movement has the clearest nerve signal. In 100 adults with type 2 diabetes randomized to a year of supervised exercise or counseling alone, leg nerve fiber density rose 1.5 fibers per millimeter with exercise and fell 0.1 without it, while HbA1c did not change. In 14 adults with painful diabetic neuropathy, sixteen weeks of aerobic exercise cut pain interference from 4.65 to 2.97 with no change in HbA1c or body mass index. The likeliest route runs through blood flow: insulin-driven flow to muscle accounts for 25 to 40 percent of insulin’s effect on glucose uptake, and nerves depend on the same small vessels. That link is my reading of the pattern, not any single trial’s conclusion.

Weight and diet push the same way. In Look AHEAD, adults with type 2 diabetes assigned to intensive lifestyle change reported less burning pain in the feet and legs, and each kilogram lost moved the symptom score. Driving sugar down with more insulin did not do the same: in ACCORD, 10,251 patients on intensive glucose control showed no meaningful difference in the composite that included neuropathy, at a hazard ratio of 0.95. Pace matters too. In reported cases of nerve injury during rapid weight loss on newer drugs, the median loss was 3.7 kilograms a month, and any medication decision belongs with your physician.

In our practice, every new patient has a fasting insulin drawn before any needle goes in, and fewer than one in a hundred of our chronic pain patients turns out to be metabolically healthy. That is a practice-reported figure from our own patients rather than a trial outcome, and individual results vary. Ask for the number itself, ask which assay ran it, repeat it on the same one, and pair it with a fasting glucose so HOMA-IR can be calculated. The studies and their limits are in the Technical Supplement to Chapter 4, and what the degeneration on a spine report is really recording is in the post on abnormal MRI findings.

Frequently asked questions

What is a normal fasting insulin level?

It depends on the laboratory and the assay. In 21,684 screened-healthy adults, the median was about 6.6 microU/mL and the upper reference limit 13.14. In a nine-year cohort, the level that best predicted diabetes was 9.16 in men and 11.13 in women, below many printed upper limits. Read your result against your own laboratory’s range and your previous value. Why a glucose test misses insulin resistance.

Can prediabetes cause nerve pain?

It often travels with it. About a third of patients whose painful neuropathy had been labeled idiopathic had impaired glucose tolerance found only on a two-hour challenge, with normal fasting glucose and HbA1c. Prediabetes cohorts show small-fiber damage that routine testing often misses, although two studies found no prediabetic signal. Why diabetic nerve pain burns at night.

What is hyperinsulinemia?

Hyperinsulinemia means chronically high insulin, usually the pancreas compensating for cells that respond poorly to the hormone. Blood sugar can stay normal for years while insulin climbs. Among American adults without diabetes, the share with fasting insulin above 10 uU/mL rose from 28.2 percent to 41.4 percent across two decades of national surveys. Why we treat insulin before we treat the joint.

Why don’t doctors routinely test fasting insulin?

Because the assay was never standardized. When first compared, the same samples run on commercial insulin methods varied by a median of 24 percent, and in 2025 only one of twelve immunoassays agreed fully with mass spectrometry. Without a standard there are no agreed cutoffs, and current diagnostic standards name no fasting insulin recommendation. The test is still useful when repeated on one assay. What a first interventional pain appointment includes.

Can exercise reverse nerve damage from insulin resistance?

It can regrow small nerve fibers. In a randomized trial of adults with type 2 diabetes, a year of supervised exercise raised leg nerve fiber density 1.5 fibers per millimeter while the counseling group lost 0.1, with no change in HbA1c. In people with prediabetes and neuropathy, a year of diet and exercise counseling regrew thigh fibers, and those who regrew most had the largest drop in pain. Why losing muscle switches off the body’s own anti-inflammatory signals.

Sources

1. 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. Lancet, 373(9682), 2215–2221. https://doi.org/10.1016/S0140-6736(09)60619-X
2. Yang, G., Li, C., Gong, Y., Fang, F., Tian, H., Li, J., & Cheng, X. (2015). Assessment of Insulin Resistance in Subjects with Normal Glucose Tolerance, Hyperinsulinemia with Normal Blood Glucose Tolerance, Impaired Glucose Tolerance, and Newly Diagnosed Type 2 Diabetes (Prediabetes Insulin Resistance Research). Journal of Diabetes Research, 2016, 9270768. https://doi.org/10.1155/2016/9270768
3. Ghasemi, A., Tohidi, M., Derakhshan, A., Hasheminia, M., Azizi, F., & Hadaegh, F. (2015). Cut-off points of homeostasis model assessment of insulin resistance, beta-cell function, and fasting serum insulin to identify future type 2 diabetes: Tehran Lipid and Glucose Study. Acta Diabetologica, 52(5), 905–915. https://doi.org/10.1007/s00592-015-0730-3
4. Schrank, Y., Fontes, R., Perozo, A. F. D. F., Araújo, P. B., Pinheiro, M. F. M. C., Gomes, D. M. V., & Vieira, L. M. F. (2024). Proposal for fasting insulin and HOMA-IR reference intervals based on an extensive Brazilian laboratory database. Archives of Endocrinology and Metabolism, 68, e230483. https://doi.org/10.20945/2359-4292-2023-0483
5. Rohlfing, C., Petroski, G., Hatten-Beck, M., Hanson, S., Hoofnagle, A. N., Little, R. R., & Kabytaev, K. (2025). The current status of serum insulin measurements and the need for standardization. Clinical Chemistry and Laboratory Medicine, 63(12), 2442–2446. https://doi.org/10.1515/cclm-2025-0552
6. Galvao, R., Plavnik, F. L., Ribeiro, F. F., Ajzen, S. A., Christofalo, D. M. de J., & Kohlmann, O. (2011). Effects of different degrees of insulin sensitivity on endothelial function in obese patients. Arquivos brasileiros de cardiologia, 98(1), 45–51. https://doi.org/10.1590/s0066-782×2011005000119
7. Mohseni, S., Badii, M., Kylhammar, A., Thomsen, N. O. B., Eriksson, K.-F., Malik, R. A., Rosen, I., & Dahlin, L. B. (2017). Longitudinal study of neuropathy, microangiopathy, and autophagy in sural nerve: Implications for diabetic neuropathy. Brain and behavior, 7(8), e00763. https://doi.org/10.1002/brb3.763
8. Callaghan, B. C., Xia, R., Reynolds, E., Banerjee, M., Rothberg, A. E., Burant, C. F., Villegas-Umana, E., Pop-Busui, R., & Feldman, E. L. (2016). Association Between Metabolic Syndrome Components and Polyneuropathy in an Obese Population. JAMA Neurology, 73(12), 1468–1476. https://doi.org/10.1001/jamaneurol.2016.3745
9. Kopf, S., Groener, J. B., Kender, Z., Fleming, T., Bischoff, S., Jende, J., Schumann, C., Ries, S., Bendszus, M., Schuh-Hofer, S., Treede, R.-D., & Nawroth, P. P. (2018). Deep phenotyping neuropathy: An underestimated complication in patients with pre-diabetes and type 2 diabetes associated with albuminuria. Diabetes Research and Clinical Practice, 146, 191–201. https://doi.org/10.1016/j.diabres.2018.10.020
10. Singleton, J. R., Marcus, R. L., Jackson, J. E., K Lessard, M., Graham, T. E., & Smith, A. G. (2014). Exercise increases cutaneous nerve density in diabetic patients without neuropathy. Annals of Clinical and Translational Neurology, 1(10), 844–849. https://doi.org/10.1002/acn3.125

Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP, board certified in anesthesiology, pain medicine, interventional pain management, addiction medicine and obesity medicine. Adapted from Chapter 4 of The Pained Brain by Dr. Padda and Dr. KrisJay Fucanan, MD. Last reviewed September 2026.

Dr. Gurpreet Singh Padda, MD, MBA, MHP