Salivary Cortisol (4-Point): Drugs That Distort This Test

What the 4-Point Salivary Cortisol Test Measures

The 4-point salivary cortisol test captures your hypothalamic-pituitary-adrenal (HPA) axis rhythm across a full day. Cortisol peaks within 30 to 60 minutes of waking (the cortisol awakening response), declines through midday and evening, then reaches its nadir around midnight. Four timed samples map this curve.

The test measures free (unbound) cortisol, which is the biologically active fraction. Serum cortisol tests measure total cortisol, roughly 90% of which is bound to cortisol-binding globulin (CBG) and albumin [1]. Salivary cortisol correlates with free serum cortisol at r = 0.85 to 0.96 in validation studies [2]. That distinction matters for drug interactions because substances that change CBG levels (like estrogen-containing contraceptives) affect serum total cortisol dramatically but affect salivary cortisol less. "Less" does not mean "not at all." Some interference still occurs.

The 2008 Endocrine Society Clinical Practice Guideline for Cushing syndrome diagnosis recommends late-night salivary cortisol (LNSC) as one of three first-line screening tests, citing sensitivity of 92 to 100% and specificity of 93 to 100% [3]. A single abnormal LNSC result requires confirmation. Two abnormal results strongly suggest Cushing syndrome. But a drug-induced false positive sends patients down an expensive, anxiety-producing diagnostic path that includes pituitary MRI and inferior petrosal sinus sampling. Understanding which medications distort results prevents that cascade entirely.

Exogenous Glucocorticoids: The Most Common Confounder

Any form of exogenous glucocorticoid can distort 4-point salivary cortisol. This includes routes patients often forget to mention: inhaled, intranasal, topical, ophthalmic, and intra-articular.

Oral prednisone and prednisolone create the most obvious interference. Many cortisol immunoassays cross-react with prednisolone. A 2016 study in Clinical Endocrinology found that prednisolone cross-reactivity ranged from 28% to 172% depending on the assay platform [4]. A patient taking 5 mg prednisone daily may show a "cortisol" level that is actually prednisolone being misread by the antibody. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) assays can distinguish cortisol from synthetic steroids, but many commercial labs still use immunoassays.

Inhaled corticosteroids are underappreciated offenders. Fluticasone propionate at doses of 500 mcg/day or higher suppresses the HPA axis enough to lower morning salivary cortisol in 30 to 50% of users [5]. Budesonide shows less systemic absorption, but high-dose use (800+ mcg/day) still suppresses cortisol in some patients. A 2021 systematic review in the Journal of the Endocrine Society analyzed 52 studies and concluded that inhaled corticosteroids cause dose-dependent HPA suppression, with fluticasone being the most potent suppressor per microgram [6].

Topical corticosteroids applied to large body surface areas, under occlusion, or to thin-skinned regions (face, groin, axillae) achieve meaningful systemic absorption. Clobetasol propionate 0.05% applied to more than 20% of body surface area suppresses morning cortisol within one week [7].

Intra-articular and epidural steroid injections suppress salivary cortisol for 1 to 6 weeks after a single injection. A 2019 prospective study showed that a single 80 mg triamcinolone acetonide knee injection suppressed morning salivary cortisol by an average of 62% at two weeks, with some patients remaining suppressed at six weeks [8].

Clinical rule: Any salivary cortisol test performed within six weeks of a steroid injection is unreliable. Document all exogenous steroid exposure, including OTC hydrocortisone cream, before ordering the test.

Oral Contraceptives and Estrogen-Containing HRT

Estrogen increases hepatic CBG synthesis by 50 to 100%, which raises total serum cortisol [9]. The Endocrine Society guideline explicitly warns that oral estrogen-containing medications can cause false-positive results on dexamethasone suppression testing and may affect salivary cortisol [3].

The effect on salivary cortisol is debated. Because salivary cortisol reflects free cortisol, some endocrinologists assumed it would remain unaffected. That assumption proved partially wrong. A 2012 study by Meulenberg and Hofman in Clinical Chemistry found that women on combined oral contraceptives (COCs) had late-night salivary cortisol values 39% higher than controls [10]. The mechanism likely involves estrogen-driven changes in 11-beta-hydroxysteroid dehydrogenase activity and altered cortisol clearance, not just CBG changes.

The practical implication: women on COCs or oral estrogen HRT should ideally discontinue for six weeks before 4-point salivary cortisol testing, or the clinician should interpret results knowing that mild elevations (up to 40% above reference range) may be estrogen-driven artifacts. The Endocrine Society notes, "Estrogen-containing medications should be discontinued for 6 weeks before biochemical testing" when evaluating for Cushing syndrome [3]. Transdermal estrogen bypasses first-pass hepatic metabolism and increases CBG less than oral formulations, though some effect persists.

Opioids and HPA Axis Suppression

Chronic opioid use suppresses the HPA axis at the hypothalamic and pituitary level. This produces genuinely low cortisol output, not just an assay artifact. A 2015 systematic review in the Journal of Clinical Endocrinology and Metabolism evaluated 21 studies and found that opioid-induced adrenal insufficiency (OIAI) occurred in 15 to 65% of chronic opioid users, depending on dose and duration [11].

Methadone, morphine, and hydromorphone show the strongest suppressive effects. Buprenorphine, a partial mu-agonist, appears to suppress the HPA axis less. Short-acting opioids cause acute cortisol suppression within hours of dosing, which can flatten the diurnal curve on a 4-point test.

A patient on chronic opioid therapy who shows a flat, low cortisol curve may have true opioid-induced adrenal insufficiency or may simply be showing the acute suppressive effect of a recent dose. Timing matters. If the patient took a dose of hydrocodone 2 hours before the waking sample, the morning peak will be blunted regardless of their baseline HPA function.

Dr. Lynnette Nieman, senior investigator at the National Institute of Diabetes and Digestive and Kidney Diseases, has noted that clinicians should "always ask about opioid use when evaluating abnormal cortisol results, as opioids are now one of the most common causes of low cortisol in clinical practice" [12].

Antiepileptic Drugs and CYP3A4 Inducers

Medications that induce hepatic CYP3A4 accelerate cortisol metabolism. This produces genuinely lower cortisol levels, not assay interference. The body clears cortisol faster, so all four salivary timepoints drop.

Phenytoin, carbamazepine, phenobarbital, and rifampin are the most potent CYP3A4 inducers affecting cortisol. A 2006 pharmacokinetic study showed that rifampin (600 mg/day for 7 days) increased cortisol clearance by 45% and reduced the area under the cortisol curve by approximately one-third [13]. This means a patient on rifampin for tuberculosis treatment may show cortisol values that mimic adrenal insufficiency.

Phenytoin deserves special attention because it also interferes with the dexamethasone suppression test (DST). Phenytoin accelerates dexamethasone metabolism, causing inadequate suppression and a false-positive DST. When a clinician sees a "failed" DST and then orders salivary cortisol, the salivary test may show low-normal values because cortisol itself is also being metabolized faster. The two tests send contradictory signals. Neither is reliable without adjusting for the CYP3A4 induction.

Mitotane, used in adrenocortical carcinoma, is the extreme example. It increases CBG (confounding serum cortisol) while simultaneously accelerating cortisol metabolism (lowering salivary cortisol). The Endocrine Society recommends against using standard cortisol testing in patients on mitotane [14].

Licorice, Supplements, and OTC Confounders

Glycyrrhizinic acid in natural licorice inhibits 11-beta-hydroxysteroid dehydrogenase type 2 (11-beta-HSD2), the enzyme that converts active cortisol to inactive cortisone in peripheral tissues. This raises cortisol levels at the tissue level. One study documented that 100 g/day of licorice confectionery for two weeks raised salivary cortisol by 32% and reduced cortisone-to-cortisol ratios [15].

Not all "licorice" products contain glycyrrhizin. Most American licorice candy uses anise flavoring. European and Middle Eastern licorice is more likely to contain the real compound. Dietary supplements containing licorice root extract (Glycyrrhiza glabra) are a more common source in clinical practice.

Other OTC and supplement confounders include:

• Biotin (vitamin B7): High-dose biotin (5,000 to 10 to 000 mcg/day, common in hair and nail supplements) interferes with streptavidin-biotin immunoassays, which many cortisol platforms use. Results can be falsely high or falsely low depending on the assay design [16]. The FDA issued a safety communication in 2017 warning that biotin can "significantly interfere" with lab tests including cortisol.

• Carbenoxolone: A licorice derivative sometimes used in alternative medicine, with the same 11-beta-HSD2 inhibition mechanism as glycyrrhizin.

• Grapefruit juice: Inhibits intestinal CYP3A4, which may modestly slow cortisol metabolism. The clinical significance for salivary cortisol is debated, but avoiding grapefruit on collection days is reasonable.

• CBD (cannabidiol): Preliminary data suggest CBD inhibits cortisol release. A 1993 study in the Brazilian Journal of Medical and Biological Research showed that 300 to 600 mg of CBD decreased cortisol levels significantly compared to placebo [17]. Given the widespread use of CBD products, this is an increasingly relevant confounder.

Patients should disclose all supplements, herbal products, and OTC medications before testing. A full 72-hour washout of biotin-containing supplements is the standard recommendation.

Timing, Behavior, and Situational Confounders

Some confounders are not drugs but are equally capable of distorting results.

Smoking: Nicotine stimulates the HPA axis acutely. Salivary cortisol rises 15 to 25% within 20 minutes of smoking a cigarette [18]. A patient who smokes before providing a late-night sample can convert a normal nadir into an abnormal elevation. The Endocrine Society guideline specifies that patients should not smoke before or during sample collection [3].

Alcohol: Heavy alcohol use creates a state called pseudo-Cushing syndrome. Chronic alcohol intake activates the HPA axis, producing genuinely elevated cortisol that normalizes after 2 to 4 weeks of abstinence. The 4-point salivary cortisol test cannot distinguish pseudo-Cushing from true Cushing syndrome. In suspected cases, repeat testing after a period of abstinence is required [3].

Shift work: Night-shift workers have an inverted or flattened cortisol rhythm. A midnight sample from a night-shift worker who has been awake since 6 PM does not carry the same clinical significance as a midnight sample from a day worker. Reference ranges assume a conventional sleep-wake cycle. Testing shift workers requires adapting collection times to their sleep schedule, not the clock.

Contamination: Hydrocortisone cream on the hands can contaminate saliva samples during collection. Patients should wash their hands thoroughly and avoid applying any topical steroid products for at least 24 hours before collection.

Food, brushing teeth, and chewing gum within 30 minutes of collection can introduce blood (from gum microtrauma) or interfere with sample pH. Blood contamination adds serum cortisol to the sample, artificially raising values. Samples should be clear, not pink-tinged.

How to Ensure Accurate Results

A clinician ordering 4-point salivary cortisol should conduct a structured medication review before the test. This review covers five categories.

Category 1: Exogenous glucocorticoids. Ask about oral, inhaled, intranasal, topical, ophthalmic, and injectable steroids. Include OTC hydrocortisone. Discontinue all non-essential glucocorticoids for a minimum of 6 weeks before testing, with physician guidance on tapering [3].

Category 2: Hormonal medications. Combined oral contraceptives, oral estrogen HRT, and anti-androgen therapies should be stopped for 6 weeks when feasible. Document any hormonal medications that cannot be discontinued [3].

Category 3: CNS-active drugs. Screen for opioids (prescription and illicit), antiepileptics (phenytoin, carbamazepine, phenobarbital), and rifampin. These cannot always be stopped, but results must be interpreted in context.

Category 4: Supplements and OTC products. Ask specifically about biotin, licorice, CBD, and herbal "adrenal support" products. Stop biotin for 72 hours. Stop licorice-containing products for 2 weeks.

Category 5: Behavioral factors. Instruct the patient to avoid smoking, alcohol, vigorous exercise, and food for at least 1 hour (preferably 2) before each collection. No teeth-brushing for 30 minutes. Wash hands. Use the collection device exactly as instructed.

The Endocrine Society recommends collecting at least two late-night salivary cortisol samples on separate evenings for Cushing syndrome screening [3]. If results are discordant (one normal, one elevated), a third sample resolves the tie. Mild elevations in the setting of known confounders should prompt repeat testing after confounder removal rather than escalation to MRI.

For patients who cannot discontinue confounding medications, LC-MS/MS assay platforms reduce (but do not eliminate) interference from synthetic glucocorticoids and biotin. Request this assay method specifically when ordering through the lab.

Normal Salivary Cortisol Reference Ranges

Reference ranges vary by assay, laboratory, and the patient's age. Representative ranges for adults using immunoassay are:

• Morning (7 to 9 AM): 0.15 to 0.70 mcg/dL (4.0 to 19.3 nmol/L)
• Noon: 0.05 to 0.30 mcg/dL
• Evening (4 to 5 PM): 0.04 to 0.15 mcg/dL
• Late night (11 PM to midnight): <0.09 mcg/dL (<2.5 nmol/L)

Late-night salivary cortisol above 0.112 mcg/dL (3.1 nmol/L) on an immunoassay, or above 0.09 mcg/dL (2.5 nmol/L) on LC-MS/MS, is the threshold most commonly used to screen for Cushing syndrome [3]. These numbers mean nothing if the patient took prednisone that morning or smoked before bed.