Can insulin resistance be reversed?

The word reversed is used loosely in this context. Insulin sensitivity can improve substantially with weight loss, increased physical activity, and dietary changes, to the point where clinical markers of insulin resistance normalize. Whether this represents reversal at the cellular level or a managed compensation is a question of framing. The Diabetes Remission Clinical Trial (DiRECT), published in the Lancet in 2017, showed that roughly half of adults with type 2 diabetes achieved remission at one year through intensive dietary weight management, with 36 percent still in remission at two years. The DPP showed that lifestyle intervention reduced progression from prediabetes to diabetes by 58 percent over three years. The evidence for meaningful improvement is strong; whether full normalization of underlying cellular insulin sensitivity is achievable in all individuals is less clear.

What is HOMA-IR and what is a normal value?

HOMA-IR stands for homeostatic model assessment of insulin resistance. It is calculated by multiplying fasting glucose in mg/dL by fasting insulin in uIU/mL and dividing by 405. It estimates beta cell function and insulin resistance from a single fasting blood draw, making it practical for clinical and research use. The limitation is that it performs less well in people who are already using insulin or have significant beta cell dysfunction. In large population studies of adults without diabetes, median values typically fall between 1 and 2. Values above 2 to 2.5 are commonly used as a threshold for research-defined insulin resistance, though no single cutoff has been universally adopted. The reference range your clinician uses may differ depending on the population against which your results are compared.

How much does weight loss improve insulin resistance?

The relationship between weight loss and insulin sensitivity is dose-dependent and well established in clinical trial data. The DPP lifestyle intervention targeted 7 percent body weight loss and produced a 58 percent reduction in diabetes progression. Bariatric surgery trials have consistently shown that large weight losses, typically 25 to 40 percent of body weight, produce dramatic improvements in insulin sensitivity often within days to weeks, before the weight loss itself is complete, pointing to mechanisms beyond fat reduction alone. For pharmacological weight loss, the STEP 1 trial reported roughly 15 percent average weight loss with semaglutide, and the SURMOUNT-1 trial reported up to 21 percent with tirzepatide, with corresponding improvements in glycemic markers.

Is insulin resistance the same as prediabetes?

No, but they are closely related. Insulin resistance is a physiological state in which cells respond poorly to insulin. Prediabetes is a clinical classification defined by specific glucose thresholds: fasting glucose between 100 and 125 mg/dL (impaired fasting glucose), glucose between 140 and 199 mg/dL two hours after a 75-gram oral glucose load (impaired glucose tolerance), or hemoglobin A1c between 5.7 and 6.4 percent, per American Diabetes Association criteria. Most people with prediabetes have insulin resistance, but insulin resistance can exist with normal glucose levels if beta cell compensation is still adequate. Conversely, beta cell failure can drive glucose into the prediabetic or diabetic range even in the presence of relatively modest insulin resistance.

Does the triglyceride to HDL ratio reliably indicate insulin resistance?

The triglyceride to HDL cholesterol ratio correlates with insulin resistance in several population studies and is convenient because both values are part of a standard lipid panel. A ratio above 3.0 mg/dL per mg/dL has been proposed as a threshold in non-Hispanic white and Black adults in some research, though the threshold differs by ethnicity, with some studies suggesting it performs less well in Asian-American individuals. The ratio is a practical screening tool rather than a diagnostic test. It is affected by many variables beyond insulin resistance, including diet, alcohol intake, thyroid function, and genetic factors, so it should be interpreted alongside other clinical information rather than in isolation.

What role does visceral fat specifically play compared with subcutaneous fat?

Visceral adipose tissue, the fat that accumulates in the abdominal cavity around the organs, is metabolically distinct from subcutaneous fat stored beneath the skin. Visceral fat is more metabolically active, secretes higher levels of pro-inflammatory cytokines including tumor necrosis factor alpha, and releases free fatty acids more readily into the portal circulation, which drains directly to the liver. The result is higher hepatic fat exposure, greater hepatic insulin resistance, and a stronger inflammatory signal compared with the same mass of subcutaneous fat. This is why waist circumference and waist-to-hip ratio predict metabolic risk better than total body fat percentage, and why imaging studies that measure visceral fat volume show stronger correlations with insulin resistance than weight or BMI alone.

Insulin Resistance: Biology, Measurement, and What the Research Shows

Published: May 10, 2026•Updated: May 10, 2026

Research use only. This article discusses compounds that include approved medications, investigational drugs, and research peptides. Material sold for research is not cleared for human administration and is not a substitute for medical advice.

Insulin resistance is the condition in which cells throughout the body respond poorly to insulin, requiring the pancreas to secrete more of the hormone to move glucose out of the bloodstream. It sits at the center of a cluster of related conditions that includes type 2 diabetes, non-alcoholic fatty liver disease, polycystic ovary syndrome, and cardiovascular disease. Understanding what causes it at the cellular level, how it is detected, and what the research shows about reversing or managing it has become one of the more active areas of metabolic medicine over the past two decades.

What insulin resistance actually is

Insulin is a peptide hormone secreted by beta cells in the pancreatic islets of Langerhans in response to rising blood glucose, primarily after eating. Its principal role is to signal cells in muscle, liver, and adipose tissue to take up glucose from circulation. When those cells respond normally, blood glucose returns to baseline within a couple of hours. When they respond poorly, the pancreas compensates by releasing more insulin. For a time that compensation holds glucose in range, but it does so at the cost of chronically elevated insulin levels.

At the cellular level, insulin binds to a receptor on the cell surface that belongs to the receptor tyrosine kinase family. Binding activates a signaling cascade involving insulin receptor substrate proteins, phosphatidylinositol 3 kinase, and Akt, eventually resulting in translocation of GLUT4 glucose transporters to the cell membrane. In insulin-resistant cells, this cascade is disrupted at one or more points. The most studied site of disruption in skeletal muscle is inhibitory serine phosphorylation of insulin receptor substrate 1, which effectively mutes the downstream signal.

Why skeletal muscle matters most

Skeletal muscle accounts for roughly 70 to 80 percent of insulin-stimulated glucose disposal in healthy adults under euglycemic conditions. This makes it the dominant site where insulin resistance develops and the dominant target for interventions intended to improve insulin action. The liver is also critical, particularly for suppressing hepatic glucose output in the fasted state, but the sheer mass of skeletal muscle means that improving muscle insulin sensitivity produces the largest quantitative change in whole-body glucose disposal.

How insulin resistance develops

The causes of insulin resistance are multiple and interact with each other in ways that are still being worked out. The most studied contributors are excess ectopic lipid accumulation, chronic low-grade inflammation, mitochondrial dysfunction, and sedentary behavior. None of these operates in isolation, and the same person often has several operating simultaneously.

Ectopic lipid accumulation

When lipid accumulates in tissues not designed for primary fat storage, particularly skeletal muscle and the liver, metabolic byproducts including diacylglycerol and ceramides accumulate inside cells. These lipid intermediates activate protein kinases that apply the inhibitory serine phosphorylation on insulin receptor substrate proteins that blunts the insulin signal. This lipid-induced mechanism was characterized through work by Gerald Shulman's group at Yale using magnetic resonance spectroscopy to measure intramyocellular lipid directly.

Chronic low-grade inflammation

Adipose tissue, especially visceral adipose tissue, secretes pro-inflammatory cytokines including tumor necrosis factor alpha and interleukin 6. At elevated concentrations, these cytokines activate inflammatory kinases that also apply inhibitory phosphorylation on the insulin signaling pathway. The connection between visceral adiposity and systemic inflammation helps explain why central fat distribution is a stronger predictor of metabolic dysfunction than total body fat.

Mitochondrial dysfunction

Impaired mitochondrial oxidative capacity has been associated with insulin resistance in skeletal muscle in several studies. The proposed mechanism is that when mitochondria cannot oxidize fatty acids efficiently, lipid intermediates accumulate within muscle cells and interfere with insulin signaling. Whether mitochondrial dysfunction is a cause or a consequence of insulin resistance, or both in different contexts, remains an active research question.

Physical inactivity

Skeletal muscle contraction activates GLUT4 translocation through pathways that are independent of insulin, primarily via AMP-activated protein kinase. Regular physical activity improves insulin-stimulated glucose uptake by multiple mechanisms: it increases GLUT4 expression, improves mitochondrial function, reduces ectopic lipid, and maintains muscle mass. The correlation between sedentary behavior and insulin resistance is well established in epidemiological data.

How insulin resistance is measured

No single test definitively quantifies insulin resistance across all clinical and research contexts. The gold standard for research is the hyperinsulinemic-euglycemic clamp, developed by Ralph DeFronzo and colleagues in 1979. In this procedure, insulin is infused at a fixed rate while glucose is infused at a variable rate adjusted to maintain blood glucose at a target, typically around 90 mg/dL. The amount of glucose required to maintain euglycemia under those conditions, the glucose infusion rate, directly measures how much glucose the body is disposing per unit of insulin. A lower glucose infusion rate indicates more severe insulin resistance.

The clamp is not practical for routine clinical use because it requires several hours, continuous blood sampling, and close clinical supervision. Clinical practice relies on surrogate markers that correlate with clamp-measured insulin resistance but are far simpler to obtain.

•Fasting insulin: elevated fasting insulin in the context of normal glucose suggests compensatory hyperinsulinemia and is an early marker of resistance
•HOMA-IR: calculated as fasting glucose times fasting insulin divided by 405 when using mg/dL and uIU/mL units; values above 2 to 2.5 are often used as a practical threshold in research contexts
•Fasting glucose and hemoglobin A1c: when beta cell compensation fails, blood glucose rises; an A1c between 5.7 and 6.4 percent meets the prediabetes definition per ADA criteria
•Triglyceride to HDL ratio: high triglycerides combined with low HDL is associated with insulin resistance and visceral adiposity; a ratio above 3.0 in non-Hispanic white and Black adults has been proposed as a threshold in some studies
•Oral glucose tolerance test: measures glucose at 30-minute intervals after a 75-gram glucose load; a 2-hour value between 140 and 199 mg/dL meets the impaired glucose tolerance definition

The natural history: from insulin resistance to type 2 diabetes

Insulin resistance does not inevitably lead to type 2 diabetes. Whether it does depends largely on whether beta cells can maintain adequate compensatory insulin secretion over time. Individuals with sufficient beta cell reserve can remain at the prediabetes stage for years or decades, though this chronic compensation carries its own metabolic costs. When beta cell function declines enough that secretion can no longer offset resistance, blood glucose rises into the diabetic range.

The Diabetes Prevention Program (DPP), a large randomized trial published in the New England Journal of Medicine in 2002, followed adults with impaired glucose tolerance, which is essentially documented insulin resistance with elevated post-load glucose. The trial found that an intensive lifestyle intervention targeting at least 7 percent weight loss and 150 minutes of moderate activity per week reduced progression to type 2 diabetes by 58 percent over three years compared with placebo. Metformin reduced progression by 31 percent. These are among the most rigorously established intervention effects in metabolic medicine.

The metabolic syndrome definition

Metabolic syndrome is a clustering of insulin resistance-related findings that together predict cardiovascular risk better than any individual component. The 2009 harmonized definition from the International Diabetes Federation, American Heart Association, and other organizations requires three of five criteria: elevated waist circumference by population-specific thresholds, triglycerides of 150 mg/dL or above, HDL below 40 mg/dL in men or 50 mg/dL in women, blood pressure of 130/85 mmHg or above, and fasting glucose of 100 mg/dL or above. The National Health and Nutrition Examination Survey has estimated that roughly one third of American adults meet these criteria, making metabolic syndrome one of the most prevalent conditions in modern medicine.

What the exercise research shows

The evidence that physical activity improves insulin sensitivity is among the strongest in metabolic medicine, with the effect being both robust and mechanistically well characterized. A single bout of aerobic exercise has been shown to improve insulin-stimulated glucose uptake in skeletal muscle for up to 48 hours afterward, an effect mediated partly through AMP-activated protein kinase and partly through increased GLUT4 content at the cell membrane.

Resistance training has attracted growing research interest for metabolic applications because it increases muscle mass, and larger muscles create a bigger glucose sink. A meta-analysis published in the British Journal of Sports Medicine in 2019 found that resistance training significantly reduced fasting glucose and HOMA-IR in adults with or at risk for type 2 diabetes. The effect sizes were comparable to some pharmacological interventions in that population. Combined aerobic and resistance training programs have shown additive or synergistic effects in several trials.

•A single aerobic exercise session can acutely improve insulin-stimulated glucose disposal for 24 to 48 hours
•Chronic aerobic training increases GLUT4 expression in skeletal muscle, improving the structural capacity for insulin-stimulated glucose uptake
•Resistance training increases muscle cross-sectional area, expanding the volume of tissue capable of disposing glucose
•High-intensity interval training has shown insulin-sensitizing effects in studies of short duration, though optimal protocols for clinical populations are still being characterized
•Breaking up prolonged sitting with short activity bouts has been shown to blunt post-meal glucose excursions compared with uninterrupted sitting, even without formal exercise

Pharmacological interventions: the GLP-1 connection

GLP-1 receptor agonists have become the most widely discussed pharmacological approach to metabolic disease over the past decade. Their primary mechanism in glucose control is glucose-dependent insulin stimulation combined with glucagon suppression and slowed gastric emptying. The weight loss produced by these agents also removes a major driver of insulin resistance: excess adiposity.

The STEP 1 trial, published in the New England Journal of Medicine in 2021, found that semaglutide 2.4 mg weekly produced average body weight reductions of about 14.9 percent at 68 weeks. Weight loss of that magnitude substantially reduces visceral adiposity and the ectopic lipid burden that drives insulin resistance in muscle and liver. In the SURMOUNT-1 trial, published in NEJM in 2022, tirzepatide produced body weight reductions of 15 to 21 percent across dose groups at 72 weeks. The dual GLP-1 and GIP mechanism of tirzepatide also appears to act more directly on glucose disposal in some studies.

Metformin and the insulin resistance evidence base

Metformin remains one of the most widely used medications for insulin resistance, particularly in type 2 diabetes. Its primary mechanism is suppression of hepatic glucose output through activation of AMP-activated protein kinase in liver cells, reducing the fasted-state glucose production that contributes to elevated fasting glucose. The DPP trial established a 31 percent reduction in diabetes progression compared with placebo. Metformin does not produce meaningful weight loss on its own and does not carry the cardiovascular outcome data that semaglutide has accumulated through the SELECT trial.

The SELECT trial, published in 2023, demonstrated that semaglutide 2.4 mg weekly reduced major adverse cardiovascular events by about 20 percent compared with placebo in adults with overweight or obesity and established cardiovascular disease but without diabetes. The cardiovascular benefit was observed across subgroups and represents the strongest pharmacological evidence to date that treating excess adiposity, and by extension a major driver of insulin resistance, produces cardiovascular outcomes that matter.

Dietary approaches and the evidence

No single dietary pattern has been shown to be uniquely superior for improving insulin sensitivity across all populations, but several approaches have accumulating evidence. The common thread among effective dietary strategies is caloric deficit or macronutrient composition that reduces postprandial glucose and insulin excursions.

Low-carbohydrate diets

Reducing dietary carbohydrate directly reduces the glucose load that requires insulin to clear. Multiple short-term randomized trials have shown that low-carbohydrate diets reduce fasting glucose, fasting insulin, and HOMA-IR more rapidly than isocaloric low-fat diets in people with insulin resistance or type 2 diabetes. The longer-term advantage over other calorie-matched approaches is less clear and adherence over years is variable.

Mediterranean-style diet

The Mediterranean dietary pattern, characterized by high vegetable and legume intake, olive oil as the primary fat source, moderate fish consumption, and low red and processed meat intake, has been associated with lower insulin resistance in observational studies and with improved glycemic markers in randomized trials. The PREDIMED trial demonstrated reduced cardiovascular events in high-risk Spanish adults following a Mediterranean diet with added olive oil or nuts compared with a low-fat control diet.

Time-restricted eating

Time-restricted eating, sometimes called intermittent fasting in popular media, limits caloric intake to a defined window each day, often 8 hours. Some trials have shown improvements in fasting insulin and HOMA-IR with time-restricted eating, though separating the effects of the eating window from total caloric reduction is methodologically difficult. A randomized trial published in the New England Journal of Medicine in 2022 found that time-restricted eating did not produce superior weight loss or metabolic improvement compared with caloric restriction without a time window when overall calories were matched.

Ultra-processed food reduction

A randomized controlled trial published in Cell Metabolism in 2019 found that an unprocessed whole-food diet, matched for calories, sugar, fat, and fiber to an ultra-processed diet, still produced greater weight loss and lower fasting insulin compared with the ultra-processed condition. The mechanistic reasons for this difference are not fully established but may include differences in satiety signaling, eating rate, and food matrix effects on digestion.