In Cushing's disease, reduced insulin sensitivity with preserved β-cell compensation is detectable even in patients with normal glucose tolerance, whereas overt diabetes is characterised by impaired β-cell compensation without significantly lower insulin sensitivity than in prediabetes, with age, morning cortisol, waist-to-hip ratio, and proinsulin independently associated with HbA1c.
Key Findings
Results
Glucose metabolism disturbances affected 68% of patients with Cushing's disease, with most diabetes cases previously undiagnosed.
The study included 50 patients with Cushing's disease and 52 age- and sex-matched healthy controls.
Prediabetes was present in 42% of patients with Cushing's disease.
Diabetes was present in 26% of patients with Cushing's disease.
Most diabetes cases were previously undiagnosed at the time of the study.
Cross-sectional analysis of a prospective cohort with matched healthy controls was used.
Results
Even patients with Cushing's disease and normal glucose tolerance showed lower insulin sensitivity and enhanced β-cell secretory activity compared with healthy controls.
Metabolic phenotyping included multi-timepoint oral glucose tolerance testing with assessment of insulin sensitivity and β-cell compensatory function.
Body composition analysis was performed by dual-energy X-ray absorptiometry (DXA).
This finding indicates that metabolic abnormalities precede detectable glucose tolerance impairment in Cushing's disease.
Enhanced β-cell secretory activity in the normal glucose tolerance group reflects compensatory response to reduced insulin sensitivity.
Results
Prediabetes in Cushing's disease was characterised mainly by lower insulin sensitivity with preserved β-cell compensation.
β-cell compensatory function was assessed using multi-timepoint oral glucose tolerance testing.
The prediabetes group retained the ability to compensate for reduced insulin sensitivity through increased β-cell secretion.
This pattern contrasts with the diabetes group, where β-cell compensation was impaired.
Results
Overt diabetes in Cushing's disease was associated with a reduced insulin-to-glucose AUC ratio without significantly lower insulin sensitivity than in the prediabetes group.
The transition from prediabetes to diabetes was marked by impaired β-cell compensation rather than a further decline in insulin sensitivity.
The insulin-to-glucose area under the curve (AUC) ratio was used as a measure of β-cell compensatory function.
Insulin sensitivity did not differ significantly between the prediabetes and diabetes groups, suggesting β-cell failure as the key driver of overt diabetes.
Results
In Cushing's disease, age, morning cortisol, waist-to-hip ratio, and proinsulin were independently associated with HbA1c, explaining 59.5% of its variance.
Forward stepwise linear regression was performed separately in each group to identify variables independently associated with HbA1c.
The four-variable model (age, morning cortisol, waist-to-hip ratio, proinsulin) explained 59.5% of the variance in HbA1c in Cushing's disease patients.
In healthy controls, only age and BMI were independently associated with HbA1c.
BMI was not significantly correlated with HbA1c in Cushing's disease, whereas trunk fat percentage was positively associated with HbA1c.
Proinsulin, a marker of β-cell stress, was an independent predictor unique to the Cushing's disease group.
Results
Central fat distribution, measured by waist-to-hip ratio and trunk fat percentage, was more closely related to glycaemia than BMI in Cushing's disease.
BMI was not significantly correlated with HbA1c in Cushing's disease patients.
Trunk fat percentage was positively associated with HbA1c in Cushing's disease.
Waist-to-hip ratio was independently associated with HbA1c in the regression model for Cushing's disease patients.
In healthy controls, BMI (not central adiposity measures) was independently associated with HbA1c.
Body composition was assessed by dual-energy X-ray absorptiometry (DXA).
What This Means
This research examined how blood sugar regulation breaks down in people with Cushing's disease, a condition caused by a pituitary tumor that leads to chronically high cortisol levels. The study followed 50 Cushing's disease patients and 52 healthy matched controls through detailed testing of blood sugar, insulin function, and body composition. It found that 68% of patients had some form of blood sugar problem — 42% had prediabetes and 26% had diabetes — and most of the diabetes cases had not been previously diagnosed. Importantly, even patients whose blood sugar appeared normal by standard criteria already showed signs of reduced insulin sensitivity and increased strain on the insulin-producing cells of the pancreas, suggesting metabolic problems begin silently before any formal diagnosis of prediabetes or diabetes.
The study also traced the progression from prediabetes to diabetes in these patients. In prediabetes, the pancreas was still able to compensate for poor insulin sensitivity by producing more insulin. In overt diabetes, this compensatory ability broke down — not because insulin sensitivity got significantly worse, but because the pancreatic beta cells could no longer keep up. Four factors were independently linked to worse blood sugar control in Cushing's disease: older age, higher morning cortisol levels, a larger waist-to-hip ratio (a measure of belly fat), and higher proinsulin (a marker of beta-cell stress). Together, these four factors explained nearly 60% of the variation in patients' HbA1c levels, a standard measure of long-term blood sugar control.
One particularly notable finding was that standard body weight (BMI) was not meaningfully linked to blood sugar levels in Cushing's disease patients, while measures of central (belly) fat — such as waist-to-hip ratio and trunk fat percentage — were. This research suggests that in Cushing's disease, where fat tends to accumulate centrally due to cortisol excess, the location of fat matters more than total body weight when assessing diabetes risk. These findings may help clinicians better identify and monitor blood sugar problems in Cushing's disease patients, even before conventional thresholds for prediabetes or diabetes are reached.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Witek P, Kwiatkowski J, Witek J, Szamotulska K, Zieliński G. (2026). Diabetes in Cushing's disease: age, hypercortisolism, central adiposity, and β-cell stress as determinants of glycaemia.. Frontiers in endocrinology. https://doi.org/10.3389/fendo.2026.1941836