Free Cortisol

A routine cortisol result can be hard to interpret when the proteins carrying the hormone change. Estrogen-containing birth control pills can raise total cortisol without a comparable rise in its unbound portion. Cirrhosis can lower total cortisol by reducing carrier proteins. Serum free cortisol can help clarify these situations when there's a specific question about adrenal function.

This test measures unbound cortisol in serum using LC-MS. It has published clinical validation and a role in selected adrenal evaluations, but interpretation is less standardized than for routine total cortisol testing. Its usefulness depends on the reason for testing, the collection time, and whether the sample was taken before or after a stimulation or suppression test. Routine testing in healthy people has not been shown to improve health or longevity.

What the Number Represents

Cortisol is a steroid hormone made by your adrenal glands. It helps maintain blood pressure, makes fuel available during illness and other demands, and regulates inflammation. Its release follows both a daily rhythm and shorter pulses. In people who sleep at night, levels are generally highest around waking and lowest near midnight.

Most circulating cortisol is attached to corticosteroid-binding globulin or albumin. Both are proteins made by your liver. Under usual conditions, roughly 5% to 10% circulates unbound, though that share changes with hormone concentrations, protein levels, and illness. Bound cortisol can detach and replenish the free pool.

Free cortisol reflects hormone available in circulation, but its concentration also depends on binding and clearance. Individual tissues can activate or inactivate cortisol locally. A serum result therefore doesn't measure your adrenal production rate or the exact amount of cortisol acting inside every tissue.

LC-MS separates compounds and identifies them using their mass and electrical charge. Direct free-hormone assays also require a step that separates unbound hormone from proteins. Published methods use approaches such as equilibrium dialysis or ultrafiltration; this test's catalog specifies LC-MS but does not identify the separation procedure. Results are reported in µg/dL.

An immunoassay uses antibodies to measure a substance. Cortisol immunoassays can also detect some related steroids, depending on the assay. LC-MS generally distinguishes cortisol from these compounds more specifically, though it still has its own sources of error. Calculated free cortisol is different again: it estimates the free concentration from total cortisol, binding-protein measurements, and assumptions about binding.

When Total Cortisol Misleads

Oral estrogen raises corticosteroid-binding globulin, while advanced liver disease and severe illness can lower binding proteins. Free cortisol can help explain a total result that doesn't fit the clinical picture. Pregnancy needs separate interpretation because both binding proteins and free cortisol can rise as part of normal pregnancy.

The liver study measured plasma free cortisol, which is a related blood measurement but uses a different specimen from this serum test. Its diagnostic criteria cannot be transferred directly to this assay. Even among serum LC-MS methods, differences in hormone separation and calibration matter.

Adrenal Insufficiency

Adrenal insufficiency means your body cannot supply enough cortisol. It can result from adrenal damage, inadequate signals from the brain, or suppression after glucocorticoid treatment. During an infection or injury, inadequate cortisol can lead to dangerously low blood pressure and an adrenal crisis.

Direct serum free cortisol has been studied during cosyntropin stimulation. Cosyntropin is a medicine that prompts the adrenal glands to release cortisol. In a study including healthy volunteers, people with established adrenal insufficiency, and people with mild cirrhosis, stimulated serum free cortisol measured by LC-MS/MS distinguished the adrenal-insufficiency groups from healthy volunteers. The cirrhosis group had largely preserved binding-protein levels, which limits what this study establishes for advanced liver disease.

Some frequently quoted accuracy figures answer a narrower question. A study of 338 people measured plasma free cortisol by LC-MS/MS, but its reported 98% sensitivity and 91% specificity came from the healthy subgroup and predicted a total-cortisol classification. Those figures do not describe accuracy against independently confirmed adrenal insufficiency, and they do not validate this serum assay across every binding-protein disorder.

Another study included 21 people with pituitary disease and 60 controls. After stimulation, plasma free cortisol measured by dialysis and mass spectrometry generally agreed with total-cortisol classifications and placed some results farther from the diagnostic boundary. This supports a possible role in difficult cases, with the same limitation that plasma results are not automatically interchangeable with serum results.

Cortisol Excess and Long-Term Health

Sustained cortisol excess causes Cushing's syndrome. Possible features include easy bruising, thinning skin, weakness in the thighs and shoulders, and increasing blood pressure or blood sugar. Some adrenal nodules produce a milder excess without the full set of visible features.

The strongest outcome evidence discussed here comes from total cortisol after dexamethasone suppression, not directly measured serum free cortisol. In a cohort of 3,656 people with adrenal nodules followed for a median of seven years, two categories of incomplete suppression were associated with about 52% and 77% higher mortality than nonfunctioning nodules after adjustment for several other risk factors. A meta-analysis pooling mortality data from four studies found about 54% higher mortality with mild autonomous cortisol secretion. These observational findings establish an association; they do not show that an isolated morning free-cortisol result predicts the same risk.

General-population studies have also linked higher morning plasma total cortisol with cardiovascular disease. A pooled analysis found about 18% higher odds per standard-deviation increase, but its genetic analysis was too imprecise to establish causation. Studies reporting higher cortisol in Alzheimer's disease used several specimens and methods. Neither finding validates serum free cortisol as a cardiovascular or dementia screening test, or establishes that lowering it in otherwise healthy people prevents those diseases.

Severe Illness

Severe illness can raise free cortisol while reducing its carrier proteins. In a study of critically ill adults, serum free cortisol was several times higher than in healthy controls, including among people whose low protein levels made total cortisol look less reassuring. That study separated the free fraction from bound hormone and measured it with a radioimmunoassay, so its numerical results are not directly interchangeable with this LC-MS test.

A separate study of severe sepsis measured plasma free cortisol by mass spectrometry and found that higher concentrations were associated with higher mortality. Free cortisol has not consistently predicted outcomes better than total cortisol, however. In a pneumonia cohort, it offered no clear advantage. Critical-care guidelines have therefore not endorsed replacing total cortisol with plasma free cortisol to diagnose illness-related corticosteroid insufficiency.

Liver and Kidney Disease

Cirrhosis affects both cortisol-binding proteins and hormone clearance. In a study of 137 people with stable advanced liver disease, calculated free cortisol did not differ significantly across disease stages, while lower total cortisol was associated with infections and liver-related death. This was a calculated estimate, not directly measured serum free cortisol. A hormone concentration can reflect several aspects of illness at once, so a prognostic association does not establish a treatment need.

Chronic kidney disease can reduce cortisol clearance and weaken the usual nighttime decline. Morning results vary, and altered cortisol binding is not a universal feature of kidney disease. A comparison involving people on hemodialysis found greater disagreement between automated total-cortisol immunoassays and mass spectrometry than in healthy controls. That supports attention to the method used for total cortisol; it does not establish that everyone with kidney disease needs serum free cortisol.

When Results Can Be Misleading

• Collection time and sleep schedule: cortisol usually falls through the morning. A later draw can therefore produce a lower result. Shift work and disrupted sleep can change the usual pattern.
• Illness and physical demands: infection, surgery, pain, and strenuous activity can alter cortisol without establishing chronic adrenal disease.
• Pregnancy: free cortisol itself can rise, so measuring the unbound fraction does not remove the need for pregnancy-specific interpretation.
• Food intake: a study of 22 healthy adults found higher serum free cortisol by dialysis and LC-MS after both prolonged fasting and substantial overfeeding. These responses do not by themselves diagnose cortisol excess.
• Medication exposure: hydrocortisone is cortisol, so the assay measures the medicine together with hormone your body produces. Other glucocorticoids can suppress your own production even when LC-MS distinguishes the drug from cortisol.
• The free-hormone separation step: temperature, acidity, incomplete equilibration, or problems with the separating membrane can change the measured free fraction.
• LC-MS limitations: very low concentrations, calibration differences, and substances in the sample that alter the instrument's signal can affect accuracy. Greater specificity than an immunoassay does not eliminate all interference.

What a Trend Can Establish

A change between two results may reflect timing, illness, medication exposure, or laboratory differences as well as a change in adrenal function. Serial serum free cortisol has no established role as a general stress or longevity score. In a defined adrenal workup, the response to stimulation or suppression may be more informative than a sequence of ordinary morning measurements.