Short Synacthen Test Interpreter

The 500 nmol/L cut-off everybody quotes was set against a polyclonal immunoassay that read high because it cross-reacted with other steroids. Modern specific assays read lower, so the same patient can pass on one platform and fail on another. Pick your assay and your sampling time and the page uses the cut-off derived for them.

Last updated 16 Sep 2026

Short Synacthen test

Assay + timing → cut-off

Stimulated serum cortisolThe post-Synacthen value, not the baseline. Enter it in whichever unit your report uses and set the unit below.

Unitµg/dL × 27.589 = nmol/L. Every cut-off on this page is held in nmol/L, because that is the unit they were published in; the µg/dL figures beside them are rounded, so entering 14.6 µg/dL gives 402.8 nmol/L and falls a fraction below the 403 nmol/L threshold it was rounded from.

Cortisol assayThis is the single most important field on the page. The three specific assays have cut-offs derived against the old Elecsys I polyclonal assay by Deming regression; anything else falls back to the traditional 500 nmol/L, which for a specific assay is too high.

Sample timing250 µg of Synacthen (cosyntropin, tetracosactide) intravenously. The low-dose 1 µg test has its own separate literature and none of these cut-offs apply to it.

Adequate for this assayExample

Cortisol 420 nmol/L at 30 minutes, Roche Elecsys Cortisol II

How the test is done, and which cut-off applies

250 µg Synacthen (cosyntropin) intravenously → cortisol at 30 and/or 60 minutes
Traditional: 500 nmol/L (18 µg/dL) · UK practice: 550 nmol/L at 30 minutes
Assay-specific (Javorsky 2021), 30 min: Elecsys II 403 · Access 409 · LC-MS/MS 401 nmol/L
Assay-specific (Javorsky 2021), 60 min: Elecsys II 406 · Access 401 · LC-MS/MS 378 nmol/L

250 µg: the standard dose, given intravenously. A supraphysiological stimulus, which is why the test interrogates adrenal capacity rather than pituitary drive
30 or 60 minutes: both are used and both have published cut-offs. The 60-minute LC-MS/MS figure is the lowest of the six at 378 nmol/L, so the time point has to be matched to its own threshold rather than assumed equivalent
why the modern numbers are lower: the assay the 18 µg/dL threshold was set with was a polyclonal immunoassay that cross-reacted with other serum steroids and therefore over-read. A specific assay counts only cortisol and returns a smaller number for the same physiology
the 1 µg test: the low-dose test uses a different stimulus and has its own literature and its own cut-offs. None of the values on this page were derived for it, and this page does not cover it
what a pass does not exclude: recent-onset secondary adrenal insufficiency, because the adrenal has not yet atrophied and still responds to a pharmacological dose of ACTH. If the clinical suspicion is of pituitary disease of recent onset, a normal short Synacthen test is not reassurance

Worked example

Cortisol 420 nmol/L at 30 minutes, Roche Elecsys Cortisol II
420 nmol/L is below the traditional 500 nmol/L, and well below the UK figure of 550
The Elecsys Cortisol II 30-minute cut-off is 403 nmol/L (14.6 µg/dL), derived by Deming regression against the old Elecsys I assay at its 18 µg/dL threshold
420 is above 403, so on this platform the response is adequate
The same 420 nmol/L read against 500 would be called subnormal — and the patient would be started on lifelong glucocorticoid replacement for a response that is in fact normal
Change the assay to LC-MS/MS at 60 minutes and the cut-off falls further, to 378 nmol/L. That is the range the six published values span, and it is why the platform belongs on the request form

Assay-specific cut-offs (Javorsky 2021)

Javorsky BR et al., J Endocr Soc 2021;5(4):bvab022. 250 µg cosyntropin intravenously, 110 patients with 0 and 30-minute samples and 89 with 60-minute samples. Each value is the reading on that assay equivalent, by Deming regression, to 18 µg/dL on the Elecsys I polyclonal assay the old threshold was set with. The authors’ conclusion is that the cut-off should be 14 to 15 µg/dL rather than 18.
Assay	30 minutes	60 minutes
Roche Elecsys Cortisol II	403 nmol/L (14.6 µg/dL)	406 nmol/L (14.7 µg/dL)
Beckman Access Cortisol	409 nmol/L (14.8 µg/dL)	401 nmol/L (14.5 µg/dL)
LC-MS/MS	401 nmol/L (14.5 µg/dL)	378 nmol/L (13.7 µg/dL)
Roche Elecsys I (polyclonal, historical)	500 nmol/L (18 µg/dL)	500 nmol/L (18 µg/dL)

A second, independent dataset pointing the same way

El-Farhan N, Pickett A, Ducroq D, et al., Society for Endocrinology BES 2011, Endocrine Abstracts 25 OC3.2. 165 healthy volunteers. These are population-derived limits per platform, not regression equivalents of an old threshold, so they are NOT the same quantity as the Javorsky figures and the two sets are not pooled here. They agree on the direction and on the abstract’s own starting point: a 30-minute cortisol below 550 nmol/L is widely accepted as diagnostic of adrenal insufficiency, and that figure is too high for most modern platforms.
Platform	30-minute limit	Note
GC-MS	420.4 nmol/L	The reference method in this dataset
Siemens Immulite	430.4 nmol/L
Beckman	458.6 nmol/L
Roche	474.4 nmol/L
Siemens Centaur	498.7 nmol/L
Abbott	573.5 nmol/L (men) / 524.4 (women)	The only platform in the set above 500
Roche, women on the combined oral contraceptive	649.4 nmol/L	Oestrogen raises cortisol-binding globulin and therefore total cortisol. A separate limit is required, not a footnote

What the short Synacthen test does not answer

The test answers one question well — can the adrenal make cortisol when maximally stimulated — and is routinely asked four others it cannot answer.
Question	Why not	What to use
Primary or secondary insufficiency?	A supraphysiological ACTH dose stimulates the adrenal directly and says nothing about pituitary drive	Paired plasma ACTH with the baseline cortisol; renin and aldosterone for mineralocorticoid deficiency
Recent-onset pituitary disease?	The adrenal has not yet atrophied and still responds, so the test can be normal	Insulin tolerance test, or repeat after several weeks, guided by the clinical picture
Is the free cortisol adequate?	The assay measures total cortisol, and oestrogen, pregnancy and low albumin all move binding proteins	Interpret against a binding-adjusted expectation; the El-Farhan data give a separate limit for women on the combined pill
Does the low-dose test say the same thing?	The 1 µg test has its own literature and its own cut-offs	A protocol and threshold derived for the 1 µg test

Why the threshold moved without the physiology moving

The short Synacthen test gives 250 µg of synthetic ACTH intravenously and measures cortisol 30 or 60 minutes later. The dose is far above anything the pituitary produces, so the test asks a narrow question: can this adrenal cortex make cortisol when it is maximally driven? A threshold of 18 µg/dL, or 500 nmol/L, became entrenched decades ago as the answer to that question, and UK practice widely used 550 nmol/L at 30 minutes. Both numbers were established against the cortisol immunoassays of the time, which were polyclonal and cross-reacted with other circulating steroids. They read high.

That is the whole of the problem. When laboratories moved to monoclonal and mass-spectrometric methods, the assays became specific: they count cortisol and largely ignore the steroids that used to be counted alongside it. The same patient, the same adrenal, the same physiology now produces a smaller number. Nobody relaxed the standard; the measurement changed underneath a threshold that stayed where it was. Javorsky and colleagues quantified the gap by running samples on three newer specific assays and regressing each against the old Elecsys I polyclonal assay at its 18 µg/dL threshold. The equivalent readings came out between 13.7 and 14.8 µg/dL — 378 to 409 nmol/L — depending on the platform and the time point, and their conclusion was that the cut-off should now be 14 to 15 µg/dL rather than 18. Keeping 18 on a modern assay, they calculated, would diagnose about twice as many patients with adrenal insufficiency as the old threshold ever did on the old assay: people committed to lifelong glucocorticoid replacement, and to the sick-day rules and the identification jewellery that go with it, because of a number that moved for analytical reasons.

A separate British dataset reaches the same place by a different route. El-Farhan and colleagues measured 165 healthy volunteers across seven platforms and derived a normal limit for each: GC-MS 420.4 nmol/L, Immulite 430.4, Beckman 458.6, Roche 474.4, Centaur 498.7, and Abbott alone above 500 at 573.5 in men and 524.4 in women. Five of seven platforms fall below the traditional threshold and all seven fall below 550. Those are population-derived limits rather than regression equivalents of an old cut-off, so they are not the same quantity as Javorsky’s and this page does not pool them — but two independent groups asking different questions both find that the modern specific assays need lower numbers. The same study produced the other finding worth carrying into practice: women taking the combined oral contraceptive needed a limit of 649.4 nmol/L on the Roche platform, because oestrogen raises cortisol-binding globulin and therefore total cortisol without raising the free, active fraction.

Two limits on the test itself are worth stating whatever cut-off is used. It does not distinguish primary from secondary adrenal insufficiency, because a pharmacological ACTH dose bypasses the pituitary entirely; that separation needs a paired ACTH, and renin and aldosterone for the mineralocorticoid axis. And it can be normal in recent-onset secondary insufficiency, because an adrenal that has not yet atrophied still responds to a large dose of ACTH even though its own trophic drive has failed. Finally, the low-dose 1 µg test is a different test with its own literature and its own thresholds; none of the values on this page were derived for it.

Frequently asked questions

What cortisol level is normal after a short Synacthen test?

It depends on the assay. Javorsky and colleagues derived 403 nmol/L at 30 minutes for the Roche Elecsys Cortisol II, 409 for the Beckman Access and 401 for LC-MS/MS, with 406, 401 and 378 respectively at 60 minutes. The traditional figures are 500 nmol/L (18 µg/dL) internationally and 550 nmol/L in UK practice, and both were set against an older, less specific assay.

Why is the cut-off lower than the 500 nmol/L I was taught?

Because the assay changed, not the standard. The 18 µg/dL threshold was established against a polyclonal immunoassay that cross-reacted with other serum steroids and therefore over-read. Modern specific assays count only cortisol, so the same physiology gives a smaller number, and keeping the old threshold would roughly double the number of adrenal insufficiency diagnoses.

Should I sample at 30 or 60 minutes?

Both are used and both have published cut-offs, so the important thing is to compare the sample against the threshold for its own time point. The six Javorsky values span 378 to 409 nmol/L, with the lowest being LC-MS/MS at 60 minutes — reading a 60-minute LC-MS/MS result against a 30-minute Roche cut-off is exactly the error this page exists to prevent.

Does a normal short Synacthen test exclude adrenal insufficiency?

Not entirely. In recent-onset secondary adrenal insufficiency the adrenal has not yet atrophied and still responds to a pharmacological dose of ACTH, so the test can be normal while the pituitary drive has already failed. If pituitary disease of recent onset is suspected, a normal result is not reassurance.

Does the contraceptive pill affect the result?

Yes, and substantially. Oestrogen raises cortisol-binding globulin, so total cortisol rises without the free, active fraction rising. El-Farhan and colleagues found women on the combined oral contraceptive needed a 30-minute limit of 649.4 nmol/L on the Roche platform against 474.4 for everyone else — a separate limit, not a footnote.

Do these cut-offs apply to the low-dose 1 µg test?

No. The low-dose test uses a different stimulus and has its own literature and thresholds. Every value on this page was derived for the standard 250 µg intravenous test, and none of them should be applied to a 1 µg protocol.

Related calculators

References

Javorsky BR, Raff H, Carroll TB, Algeciras-Schimnich A, Singh RJ, Colón-Franco JM, Findling JW. New Cutoffs for the Biochemical Diagnosis of Adrenal Insufficiency after ACTH Stimulation using Specific Cortisol Assays. J Endocr Soc. 2021;5(4):bvab022. doi:10.1210/jendso/bvab022
El-Farhan N, Pickett A, Ducroq D, Bailey C, Parham K, Morgan L, Armston A, Evans C, Rees DA. Determination of method-specific normal cortisol responses to the short Synacthen test. Society for Endocrinology BES 2011. Endocrine Abstracts. 2011;25:OC3.2.
Bornstein SR, Allolio B, Arlt W, et al. Diagnosis and Treatment of Primary Adrenal Insufficiency: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2016;101(2):364–389. doi:10.1210/jc.2015-1710

Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.