AM Cortisol, Nutrition, and Fasting: What Your Morning Lab Result Actually Means

What Is AM Cortisol and Why Is It Drawn in the Morning?

Cortisol secretion follows a tight circadian pattern driven by the suprachiasmatic nucleus and the HPA axis. Levels peak within 30 to 45 minutes of waking, then fall gradually across the day, reaching a nadir around midnight. Drawing the specimen at 7 to 9 a.m. Captures that peak and gives clinicians the highest-signal window to detect both excess and deficiency 1.

The Cortisol Awakening Response

The cortisol awakening response (CAR) is a discrete, ACTH-dependent surge that adds 50 to 160% above pre-waking baseline within the first 20 to 30 minutes after eye-opening 2. A blood draw taken at exactly 8 a.m. Will capture a higher cortisol in a patient who woke at 7:30 a.m. Than in one who woke at 6:00 a.m. Standardizing wake time relative to the draw is therefore a pre-analytical variable that most outpatient labs do not control for, and its omission can mimic mild hypercortisolism or, conversely, blunt detection of insufficiency.

Serum vs. Salivary vs. Urinary Cortisol

Serum AM cortisol measures total cortisol, roughly 90% of which is protein-bound (primarily to corticosteroid-binding globulin). Salivary cortisol reflects the free, bioactive fraction and correlates more tightly with the CAR. For adrenal insufficiency screening in the outpatient setting, the Endocrine Society's 2016 Clinical Practice Guideline recommends a morning serum cortisol as the first-line test, with values below 3 mcg/dL confirming and values above 18 mcg/dL excluding adrenal insufficiency with high confidence 3.

Normal and Optimal AM Cortisol Ranges

Most certified laboratories report a reference interval of approximately 6 to 23 mcg/dL for an 8 a.m. Serum draw, but "normal" and "optimal" are not the same number 4.

Laboratory Reference Range vs. Functional Target

The 6 to 23 mcg/dL interval is derived from population statistics, the central 95% of apparently healthy adults. A result of 7 mcg/dL clears the reference range but sits close to the threshold where cosyntropin stimulation testing becomes warranted. Longevity-medicine practitioners and functional endocrinologists generally target 12 to 20 mcg/dL as the "optimal" morning serum cortisol: high enough to confirm an intact HPA axis under baseline conditions, low enough to avoid the metabolic consequences of chronic cortisol excess 5.

The 3 mcg/dL and 18 mcg/dL Clinical Decision Points

The Endocrine Society guideline language is explicit: "A morning serum cortisol concentration of less than 3 mcg/dL is highly suggestive of adrenal insufficiency, whereas a value greater than 18 mcg/dL makes the diagnosis unlikely" 3. Values between 3 and 18 mcg/dL fall in a gray zone requiring dynamic testing (typically a 250 mcg cosyntropin stimulation test) before a diagnosis is confirmed or excluded. This three-zone model is the framework most endocrinologists use when interpreting a single AM cortisol.

Assay Variability Across Platforms

Immunoassay platforms (Abbott Architect, Roche Elecsys, Siemens Centaur) show inter-platform coefficients of variation up to 20% at concentrations near 10 mcg/dL 6. Mass spectrometry-based methods (LC-MS/MS) provide superior accuracy, particularly at low concentrations where adrenal insufficiency decisions are made. Patients retesting cortisol over time should use the same laboratory platform to avoid apparent changes driven purely by assay drift.

How Fasting Duration Affects AM Cortisol

An overnight fast of at least 8 hours is the standard pre-draw requirement for AM cortisol. Eating before the draw is not merely inconvenient, it actively changes the result through at least two mechanisms 7.

The Postprandial Cortisol Suppression Effect

A mixed meal suppresses AM cortisol by approximately 15 to 25% compared to a fasted draw in the same individual. This appears to be mediated partly by the insulin response and partly by direct glucocorticoid feedback. A 2015 crossover study (N=30) published in the European Journal of Endocrinology found that subjects who ate a 600 kcal breakfast 60 minutes before an 8 a.m. Draw had mean cortisol values 18% lower than their fasted counterparts drawn on the same day of the week 8. That suppression is large enough to push a borderline-low result below the 3 mcg/dL threshold, potentially generating a false-positive adrenal insufficiency screen.

Prolonged Fasting: More Than 16 Hours

Extending the fast beyond 16 to 24 hours has the opposite effect. Cortisol rises during prolonged caloric deprivation as part of a counter-regulatory response to falling glucose and insulin. A controlled crossover trial (N=15) showed that a 24-hour fast elevated morning serum cortisol by a mean of 3.1 mcg/dL above the standard 8-hour fasted baseline 9. Patients who practice intermittent fasting with eating windows that close early in the evening (e.g., 16:8 protocols ending by 4 p.m.) may arrive to their 8 a.m. Draw 16 or more hours fasted, inadvertently inflating their result.

Practical Recommendation on Fasting Window

An 8 to 12 hour overnight fast, nothing by mouth except water after approximately 9 to 10 p.m. The prior evening, best approximates the conditions under which laboratory reference ranges were established. Communicate your actual fasting duration to the ordering clinician so it can be documented alongside the result.

Macronutrient Composition and Cortisol

The total calories consumed before a draw matter, but the macronutrient split of the prior evening's meal also modulates the next morning's HPA axis activity 10.

Carbohydrate Intake

High-carbohydrate meals (greater than 60% of calories from carbohydrates) acutely lower cortisol through insulin-mediated feedback. A randomized crossover study comparing a high-carbohydrate versus a high-fat dinner found that AM cortisol was 11% lower the morning after the high-carbohydrate condition 10. This effect is modest but directionally consistent across replication studies.

Dietary Fat

Chronic high dietary fat intake is associated with a blunted cortisol awakening response in observational data. A 12-week dietary intervention in overweight adults (N=47) published in Psychoneuroendocrinology found that subjects randomized to a high-fat diet (greater than 40% fat calories) showed a 14% reduction in CAR area under the curve compared to the low-fat group 11. The mechanism is not fully established but may involve altered glucocorticoid receptor sensitivity in the hypothalamus.

Protein and the HPA Axis

High-protein meals acutely stimulate cortisol through amino acid-driven glucagon release and subsequent glycogenolysis. A single high-protein meal (80 g protein) raised serum cortisol by approximately 20% two hours post-ingestion in a controlled feeding study (N=12) 12. The effect dissipates within 4 to 6 hours, so a high-protein dinner the night before a morning draw is unlikely to affect 8 a.m. Results under an 8-hour fast. A high-protein breakfast eaten before the draw would, however, inflate the measurement.

Caloric Restriction, Very-Low-Calorie Diets, and Cortisol

Moderate caloric restriction, the kind used in most commercial weight-loss programs, has a complex, time-dependent relationship with cortisol 13.

Short-Term Restriction (1 to 4 Weeks)

During the first weeks of a calorie-restricted diet, cortisol often rises as the HPA axis responds to energy deficit. A meta-analysis of 18 dietary intervention studies found that cortisol increased significantly during the first four weeks of caloric restriction, with a pooled mean increase of 1.6 mcg/dL in morning serum cortisol 13. Patients who recently started a weight-loss diet and then get AM cortisol tested may show results that overstate baseline HPA activity.

Very-Low-Calorie Diets

Very-low-calorie diets below 800 kcal per day produce a more pronounced response. Twenty-four-hour urinary free cortisol can increase 50 to 100% above pre-diet baseline within two weeks on a very-low-calorie diet, consistent with a meaningful counter-regulatory stress response 14. This elevation is not a marker of Cushing syndrome. It reflects physiologic adaptation to severe energy restriction and typically normalizes within 4 to 8 weeks of returning to maintenance calories.

Long-Term Caloric Restriction and Longevity Data

The CALERIE-2 trial (N=218, 24 months) tested sustained 25% caloric restriction in non-obese adults. Salivary cortisol measures at 12 and 24 months were not significantly different from the ad libitum control group, suggesting that long-term moderate restriction does not chronically raise cortisol once the body has adapted 15. Short-term cortisol spikes during caloric restriction therefore appear to be a transient adaptation, not a sustained endocrine disruption.

Caffeine and AM Cortisol

Caffeine is the variable most commonly overlooked in pre-draw instructions, yet its effect on cortisol is well-documented and clinically significant 16.

Acute Dose-Response

A single 300 mg caffeine dose (roughly two to three standard 8 oz cups of brewed coffee) raises serum cortisol approximately 30% above baseline in caffeine-naive or low-habituated adults within 60 minutes of ingestion 16. The mechanism involves adenosine receptor blockade and subsequent activation of the HPA axis at the hypothalamic level.

Habituation and Chronic Users

Regular caffeine consumers show attenuated cortisol responses over time. Daily users consuming more than 300 mg per day may show only a 10 to 15% cortisol increase after acute dosing, compared to 30% in naive individuals 17. Still, a morning coffee consumed before the 8 a.m. Draw should be considered a confounding variable regardless of habitual intake, and standard pre-draw instructions should specify no caffeine after midnight.

Timing of the Draw Relative to Caffeine

If a patient consumed 200 mg caffeine at 6 a.m. And is drawn at 8 a.m., peak cortisol effect (60 to 90 minutes post-dose) coincides directly with specimen collection. A morning cortisol of 22 mcg/dL under those conditions does not rule in hypercortisolism; it rules in coffee. Repeat the draw fasted and caffeine-free before pursuing further workup for suspected Cushing syndrome.

Alcohol, Micronutrients, and Other Nutritional Variables

Alcohol

Acute alcohol ingestion suppresses cortisol acutely but may produce a rebound elevation 8 to 12 hours later via HPA axis activation. A controlled alcohol challenge study (N=20) demonstrated mean AM cortisol values 22% above fasted baseline the morning after consumption of 0.75 g ethanol per kg body weight the prior evening 18. Patients should abstain from alcohol for at least 24 hours before AM cortisol testing.

Vitamin C and Adaptogenic Supplements

High-dose ascorbic acid (vitamin C, 1 to 3 g per day) may reduce cortisol modestly by protecting adrenal tissue from oxidative stress. A randomized controlled trial in marathon runners (N=56) found that 1.5 g/day ascorbic acid reduced post-race cortisol by 30% compared to placebo 19. The clinical relevance for resting AM cortisol in non-athletes is less clear, but high-dose supplementation taken on the morning of the draw is worth documenting.

Sodium and the Renin-Angiotensin-Aldosterone Axis

Very low sodium intake (below 50 mEq/day) stimulates the renin-angiotensin-aldosterone system, which may secondarily raise ACTH and cortisol in some individuals. The effect is modest under normal dietary conditions but becomes relevant in patients on sodium-restricted therapeutic diets or those presenting with primary aldosteronism workup 20.

Pre-Draw Protocol: Standardizing the AM Cortisol

A single inconsistently collected AM cortisol is difficult to interpret. Standardizing the pre-draw protocol eliminates most nutritional confounders and improves clinical utility 3.

Recommended Pre-Draw Instructions

The following conditions should be met for a valid AM cortisol:

• Nothing by mouth except water for 8 to 12 hours before the draw (avoid both short and prolonged fasting extremes).
• No caffeine after midnight the prior evening.
• No alcohol for at least 24 hours.
• Draw performed between 7:00 and 9:00 a.m., ideally within 60 minutes of waking.
• No acute exercise on the morning of the draw (moderate-intensity exercise raises cortisol 60 to 80% above resting baseline transiently) 21.
• Document wake time, fasting duration, and any supplements taken.

When to Repeat a Non-Diagnostic Result

Any AM cortisol between 3 and 18 mcg/dL that falls in the gray zone, or any result obtained under suboptimal conditions (short fast, caffeine consumed, very-low-calorie diet started within four weeks), warrants repeat testing under standardized conditions before proceeding to dynamic stimulation testing. Unnecessary cosyntropin stimulation tests carry low but real risks, including cost, patient anxiety, and rare hypersensitivity reactions.

Clinical Scenarios Where Nutrition Confounds the Result

The Patient on a GLP-1 Agonist

Patients using semaglutide (Ozempic, Wegovy) or tirzepatide (Mounjaro, Zepbound) experience significant reductions in caloric intake, particularly during the first 12 to 16 weeks of dose escalation. The caloric restriction effect described above applies: cortisol may be transiently elevated by 1 to 3 mcg/dL above that patient's true baseline during rapid weight loss phases. AM cortisol drawn during active dose escalation should be interpreted with this in mind, and clinicians may choose to defer adrenal screening until weight has stabilized.

The Patient Practicing Time-Restricted Eating

A 16:8 eating window ending at 4 p.m. Means the patient arrives at an 8 a.m. Draw having fasted approximately 16 hours. As noted, prolonged fasting raises cortisol by a mean of 3.1 mcg/dL in controlled settings 9. A result of 16 mcg/dL in this patient might be 13 mcg/dL under standard 8-hour fasting conditions. Document the eating window explicitly.

The Patient With Suspected Adrenal Insufficiency

In this context, nutritional confounders work against diagnosis. A pre-draw meal, a short fast, or recent caffeine consumption will all raise cortisol and push a truly insufficient patient above the 3 mcg/dL diagnostic threshold. The Endocrine Society guideline notes that the diagnostic threshold assumes a standardized morning draw; values obtained under fed conditions cannot be directly compared to the 3 mcg/dL cutoff 3.