The study team, led by researchers from Aerospace Center Hospital and the Chinese PLA General Hospital in Beijing, enrolled 30,343 adults who received routine health examinations between January 2017 and November 2021. For participants who visited more than once, the earliest examination served as the baseline record. This large, unselected health-examination population offers something clinical trials often lack: a snapshot of ordinary people across the full spectrum of metabolic health, from lean individuals with normal insulin to those with obesity, elevated blood pressure, and overt insulin resistance.

The researchers focused on two distinct targets of metabolic injury. The first was arterial stiffness, measured using brachial-ankle pulse wave velocity, or baPWV, a noninvasive technique that tracks how quickly a pressure wave travels along the arteries between the arm and the ankle. Stiffer arteries transmit the pulse faster, so a higher velocity indicates harder vessel walls. The team defined arterial stiffness as a mean baPWV of 1400 centimeters per second or higher. The second target was subclinical myocardial injury, assessed through cardiac troponin I or T, proteins released by heart muscle cells when they are damaged. Detectable troponin at low levels in apparently healthy people can flag early heart muscle stress long before symptoms appear.

Fasting insulin was examined in three complementary ways: as quartile categories comparing the highest insulin group with the lowest, as a log-transformed continuous variable to capture dose-response patterns, and as a binary hyperinsulinemia indicator defined by the cohort-specific 75th percentile. The investigators also calculated HOMA-IR, the homeostatic model assessment of insulin resistance, which combines fasting insulin and glucose into a single index of how hard the body must work to keep blood sugar under control. Logistic regression models quantified the associations, and restricted cubic spline modeling explored whether the relationships were linear or followed more complex curves. All statistical analysis was performed using Python version 3.13.0.

The results painted a consistent picture. People in higher insulin categories carried more adiposity, higher blood pressure, higher glucose, greater insulin resistance as measured by HOMA-IR, higher triglycerides, and, critically, higher baPWV values and more arterial stiffness. In the fully adjusted model, hyperinsulinemia was associated with a 31 percent increase in the odds of arterial stiffness, with an odds ratio of 1.31 and a 95 percent confidence interval of 1.22 to 1.40, a highly statistically significant result. When the researchers compared the highest insulin quartile with the lowest, the contrast was even sharper: the top quartile faced 56 percent greater odds of arterial stiffness, with an odds ratio of 1.56 and a confidence interval of 1.41 to 1.74.

The spline analysis added an important technical nuance. Both fasting insulin and HOMA-IR showed positive but non-linear relationships with mean baPWV, meaning the association was not a simple straight line. This kind of dose-response detail matters because it suggests the vascular consequences of elevated insulin may not accumulate uniformly across the metabolic spectrum; the shape of the curve hints at where along the insulin range risk begins to accelerate, information that could eventually help clinicians decide when an insulin measurement becomes clinically meaningful rather than merely a laboratory value.

Not every finding favored insulin as a cardiovascular culprit. When the researchers turned to the composite myocardial injury endpoint, built from troponin I or T measurements, the association with hyperinsulinemia did not reach statistical significance. In other words, in this cohort, elevated fasting insulin tracked strongly with stiff arteries but only weakly, if at all, with early heart muscle damage detectable by troponin. The authors interpret this asymmetry carefully: fasting insulin may serve as a potential early metabolic indicator of vascular stiffening, while its relationship with myocardial injury remains unresolved and requires further longitudinal investigation to clarify.

The study’s design deserves scrutiny when weighing these conclusions. As a retrospective, cross-sectional analysis of health examination records, it captures associations at a single point in time and cannot prove that hyperinsulinemia causes arteries to stiffen. It is possible that insulin is a marker riding along with other processes, such as visceral fat accumulation, sympathetic nervous system activation, or chronic low-grade inflammation, that independently damage vessel walls. The researchers addressed this concern with subgroup analyses and sensitivity analyses, the latter reported in supplementary tables, to test whether the association held across different population strata and modeling choices. The consistency of the signal across unadjusted and adjusted models strengthens the case, but only longitudinal follow-up can establish temporal sequence.

Why would excess insulin stiffen arteries in the first place? Biologically, insulin is more than a glucose-regulating hormone. At normal levels it promotes vasodilation through nitric oxide pathways in the endothelium, the inner lining of blood vessels. But in states of insulin resistance, this vascular-protective signaling blunts while other insulin-driven pathways, including those promoting smooth muscle cell growth, sodium retention, and sympathetic activation, may persist or even intensify. The result is a milieu that favors vascular remodeling, reduced arterial elasticity, and rising blood pressure. Elevated insulin also correlates with dyslipidemia and adiposity, both of which feed the same pathological loop. The strong correlation between insulin quartiles and nearly every cardiometabolic measure in this dataset illustrates how tightly these mechanisms intertwine.

The clinical implications are potentially significant. Fasting insulin is inexpensive, widely available, and already measured in many health examinations, yet it rarely receives the attention given to glucose, cholesterol, or blood pressure. If future prospective studies confirm that insulin identifies people at risk of vascular stiffening before hypertension or diabetes emerges, a simple blood test could help flag patients who would benefit from earlier lifestyle intervention, whether weight management, exercise, or dietary change, at a stage when arteries may still recover elasticity. For now, the study’s authors stop short of such recommendations, noting that the link between hyperinsulinemia and subclinical myocardial injury was weaker and that longitudinal research is needed. But with more than 30,000 participants and a robust dose-response signal for arterial stiffness, the findings add weight to a growing argument that insulin itself, not merely the blood glucose it governs, deserves a place in the early warning system for cardiovascular disease.

Subject of Research: The association of hyperinsulinemia with arterial stiffness and myocardial injury in a large real-world health examination cohort

Article Title: Hyperinsulinemia and its associations with arterial stiffness and myocardial injury: a real-world cohort study

Article References: Liu, H., Liu, J., Li, N., Hua, Y., Wang, H., Deng, X., Wang, L., Dong, S., Fu, X., & Yan, S. (2026). Hyperinsulinemia and its associations with arterial stiffness and myocardial injury: a real-world cohort study. BMC Endocrine Disorders. https://doi.org/10.1186/s12902-026-02521-6

Image Credits: AI Generated

DOI: 10.1186/s12902-026-02521-6

Keywords: hyperinsulinemia, arterial stiffness, baPWV, insulin resistance, HOMA-IR, myocardial injury, troponin, fasting insulin, cardiovascular risk, cohort study, metabolic syndrome, restricted cubic spline