CAH 17-OHP Interpreter

17-hydroxyprogesterone interpretation for 21-hydroxylase deficiency, in the order that actually decides the answer: which unit, was the sample taken before 8 am, and was it baseline or cosyntropin-stimulated. A normal afternoon baseline excludes nothing — only a stimulated value excludes non-classic disease. For the units alone, see the 17-hydroxyprogesterone unit converter.

Last updated 18 Sep 2026

Reading a 17-OHP for 21-hydroxylase deficiency

Unit + timing + context → interpretation

17-hydroxyprogesterone resultEnter the number as printed on the report, then set the unit below to match it. Do not convert it yourself first.

Unit on the reportIf your report says ng/mL or µg/L, multiply by 100 and choose ng/dL: 2 ng/mL is 200 ng/dL. The conversion is 1 ng/dL = 0.0302599 nmol/L, so 1 nmol/L = 33.05 ng/dL. This is the input that most often gets set wrong and it changes the answer 33-fold.

Which sample is this?These are three different measurands, not three occasions. A dried blood spot is whole blood on filter paper and its cut-offs belong to the screening programme, not to a serum assay. Baseline serum screens. Only the stimulated value excludes non-classic disease.

When was the sample taken?17-OHP follows the ACTH rhythm and is substantially lower in the afternoon than in the morning, which is why the Endocrine Society guideline specifies "early-morning (before 8 am)". This selector bears on a baseline serum sample. It does not change a cosyntropin-stimulated result, which is driven by the injection, and it does not bear on a newborn screen.

Who is the patient?Gestational age and birthweight are on this list because they shift a newborn screening cut-off substantially and false positives in preterm infants are the dominant practical problem with the screen. Cycle phase is on it because the luteal 17-OHP reference ceiling is more than three times the follicular one.

Early-morning baseline at or above 6 nmol/L (200 ng/dL) but below 30 — indeterminate; a cosyntropin stimulation test is indicatedExample

A 24-year-old woman with hirsutism and oligomenorrhoea. Serum 17-OHP 450 ng/dL, drawn at 07:30, follicular phase.

The thresholds, in both units, and what each one can do

Baseline, early morning — below 6 nmol/L (200 ng/dL): unlikely. 6 to 30: stimulate. At or above 30 nmol/L (1,000 ng/dL): diagnostic. At or above 100 nmol/L (3,300 ng/dL): the classic range.
Cosyntropin-stimulated, 60 min after 250 µg — below 30 nmol/L: excluded. 30 to 100: non-classic. At or above 100: classic.
Newborn dried blood spot — no threshold is printed on this page. The programme’s own gestational-age-stratified cut-off governs.
Unit bridge — 1 ng/dL = 0.0302599 nmol/L; 1 nmol/L = 33.05 ng/dL; 1 ng/mL = 100 ng/dL.

6 nmol/L / 200 ng/dL: the screening threshold for a baseline, and only a screening threshold. Chesover and colleagues report a baseline at or above 6 nmol/L as 96% sensitive and 96% specific for non-classic disease; the non-classic CAH literature recommends 2 ng/mL (6 nmol/L) "if more appropriate in-house cut-off values are not available". A value below it makes the diagnosis unlikely, not excluded
30 nmol/L / 1,000 ng/dL: the definitive threshold, and it works in both directions: "definitive diagnosis requires a 17-OHP concentration ≥10 ng/ml (30 nmol/l), either basally or after cosyntropin-stimulation". As a stimulated peak, Chesover reports it at 96% sensitivity and 100% specificity
100 nmol/L / 3,300 ng/dL: the level above which classic disease becomes much the more likely reading. Honour gives basal values in classic 21-hydroxylase deficiency as "usually >100" nmol/L, with stimulated values running to 2,000, against a non-classic basal range above 6 but below 100. The bands overlap, so this separates classic from non-classic only alongside the clinical picture and the genotype
8.6 nmol/L / 285 ng/dL: the luteal-phase reference ceiling, against about 2.4 nmol/L (80 ng/dL) for the early-morning follicular phase — more than three times higher. That is why the cycle phase is asked for, and why a luteal value between 6 and 8.6 is repeated in the follicular phase rather than stimulated straight away
before 8 am: the Endocrine Society guideline’s own wording: "screening with an early-morning (before 8 am) baseline serum 17-hydroxyprogesterone measurement by liquid chromatography-tandem mass spectrometry". 17-OHP follows the ACTH rhythm and is lower in the afternoon, so the 6 nmol/L threshold does not transfer to a late sample and a normal afternoon baseline excludes nothing
why no newborn number: because there is no single right one. The Endocrine Society guideline recommends assays "standardized to a common technology with norms stratified by gestational age" and prints no threshold table. A dried blood spot is whole blood on filter paper, not serum; the cut-off depends on the assay and moves substantially with gestational age. van der Kamp’s medians — 38 nmol/L stratified by gestational age against 54 nmol/L by birthweight — are given on the page to show the size of the effect, not as cut-offs to apply

Worked example

A 24-year-old woman with hirsutism and oligomenorrhoea. Serum 17-OHP 450 ng/dL, drawn at 07:30, follicular phase.
Unit first. 450 ng/dL × 0.0302599 = 13.6 nmol/L. Had this been read as 450 nmol/L it would have looked like florid classic disease; had a 450 nmol/L result been read as ng/dL it would have looked normal. The factor is 33
Timing. 07:30 is before 8 am, so the baseline thresholds apply as published
Phase. Follicular, so the luteal reference ceiling of 8.6 nmol/L does not come into it
Against the thresholds. 13.6 nmol/L is at or above the 6 nmol/L screening threshold and below the 30 nmol/L definitive one → indeterminate
Next test. 250 µg cosyntropin with 17-OHP at 60 minutes. A peak at or above 30 nmol/L confirms 21-hydroxylase deficiency; below 30 excludes non-classic disease. A second baseline would add nothing
Change one thing. Had the same 450 ng/dL been drawn at 4 pm, the answer would have been "repeat in the early morning" rather than "stimulate" — and had an afternoon sample come back at 150 ng/dL, the answer would have been that it excludes nothing at all, which is the trap this page is built around

The same number in three units

1 nmol/L is 33.05 ng/dL, and ng/mL is ng/dL divided by 100. Reading a ng/dL report against an nmol/L threshold overstates the result 33-fold and the reverse understates it as much — the commonest single way this test is misread. The 17-hydroxyprogesterone unit converter converts in every direction.
nmol/L	ng/dL	ng/mL (µg/L)	What it means for a baseline
2.4	80	0.80	The early-morning follicular reference ceiling
6	200	2.00	Screening threshold — at or above this, stimulate
8.6	285	2.85	The luteal-phase reference ceiling
30	1,000	10.0	Definitive, basally or after stimulation
100	3,300	33.0	The classic range

What each sample can and cannot establish

Read the right-hand column. Three of the four rows cannot exclude the diagnosis, and the row most often used to do it — an untimed baseline — is the one that carries the least weight. The asymmetry between diagnosing and excluding is the whole structure of this test.
Sample	Can it diagnose?	Can it exclude?
Newborn dried blood spot	No — it screens; the programme’s gestational-age cut-off governs	No — second-tier LC-MS/MS or a cosyntropin test confirms
Baseline serum, before 8 am	Yes if at or above 30 nmol/L (1,000 ng/dL)	No — only makes non-classic disease unlikely (96% sensitive)
Baseline serum, afternoon or untimed	Yes if high — the morning value would be higher still	No, and not even weakly. Excludes nothing
Cosyntropin-stimulated, 60 min	Yes if at or above 30 nmol/L — 96% sensitive, 100% specific	Yes — this is the only sample that excludes non-classic disease

Three things decide the answer before the number does

17-hydroxyprogesterone is the substrate of the enzyme 21-hydroxylase, so when that enzyme is deficient the substrate accumulates and cortisol synthesis falls. The loss of cortisol negative feedback drives ACTH up, the adrenal cortex hypertrophies, and the precursors that pile up behind the block are shunted into androgens. That single pathway explains the whole clinical picture: glucocorticoid deficiency, mineralocorticoid deficiency in the severe forms, and androgen excess. 21-hydroxylase deficiency accounts for roughly 95% of congenital adrenal hyperplasia, and 17-OHP is the marker for it. The 17-hydroxyprogesterone unit converter is the companion page for the units and the reference intervals.

The reason this page asks five questions before it reads the number is that three of them change the interpretation more than the number itself does. The unit comes first. ng/dL and nmol/L differ by a factor of 33.05, so a North American report read against a European threshold, or the reverse, is wrong by more than an order of magnitude. This is not a hypothetical: the diagnostic threshold is 1,000 ng/dL and also 30 nmol/L, and both numbers appear in the literature without their units being emphasised. ng/mL, a third unit in circulation, is the ng/dL figure divided by 100.

Then the timing. 17-OHP follows the ACTH circadian rhythm and is substantially lower in the afternoon than in the early morning, which is why the Endocrine Society guideline specifies screening with an early-morning sample taken before 8 am, measured by liquid chromatography-tandem mass spectrometry. The consequence is asymmetric and worth stating carefully. A high afternoon value is still a high value, and arguably more concerning, because the morning figure would have been higher still. A normal afternoon value, however, excludes nothing at all: the threshold it is being compared against was calibrated on morning samples, and a patient with non-classic disease can have an afternoon baseline squarely in the reference range. An afternoon 17-OHP used to rule out 21-hydroxylase deficiency is the most common way this test produces a false reassurance.

Then baseline against stimulated, which is the distinction that decides what the result is capable of. A baseline 17-OHP screens. At or above 6 nmol/L, about 200 ng/dL, it triggers a stimulation test; Chesover and colleagues report that threshold as 96% sensitive and 96% specific for non-classic disease. At or above 30 nmol/L, about 1,000 ng/dL, it is definitive on its own. But below 6 nmol/L it only makes the diagnosis unlikely, because at 96% sensitivity roughly one non-classic patient in twenty-five sits beneath it. What formally excludes non-classic disease is a cosyntropin-stimulated value: 250 µg intravenously with 17-OHP measured at 60 minutes, where a peak below 30 nmol/L rules it out and a peak above confirms it at 96% sensitivity and 100% specificity. So a normal baseline drawn in the afternoon excludes nothing, and even a normal baseline drawn properly at 7 am does not formally exclude it — only the stimulated value does.

Classic and non-classic disease are two ends of one spectrum rather than two diseases. Classic 21-hydroxylase deficiency presents in the newborn period, with ambiguous genitalia in a 46,XX infant and with salt-wasting crisis in the first two weeks in the severe form, and Honour gives its basal 17-OHP as usually above 100 nmol/L with stimulated values reaching 2,000. Non-classic disease reflects partial enzyme deficiency: it presents later with premature pubarche, or in a woman with hirsutism, acne, oligomenorrhoea and subfertility, and often never presents at all in a man. Its basal values sit above 6 but below 100 nmol/L. It does not cause salt wasting and does not usually need daily glucocorticoid. The two bands overlap, so a single basal number should not be used to sort classic from non-classic — that rests on the clinical picture, the mineralocorticoid axis and CYP21A2 genotyping.

Newborn screening deserves its own note, because it is where this test is used most and performs least well. This page prints no newborn cut-off, and that is deliberate. The Endocrine Society guideline recommends that first-tier screens use 17-OHP assays standardised to a common technology with norms stratified by gestational age, and it publishes no numeric threshold — because the right value depends on the assay, on the fact that a dried blood spot is whole blood rather than serum, and above all on gestational age. Premature, sick or stressed infants have genuinely higher 17-OHP than term infants, and the guideline states that this generates many false positives. The size of the problem is stark: in the Swedish programme the positive predictive value was 25% in full-term infants and 1.4% in preterm ones. van der Kamp and colleagues showed gestational age is a better predictor of 17-OHP than birthweight, with the median cut-off falling to 38 nmol/L when stratified by gestational age against 54 nmol/L by birthweight, and adopting gestational-age criteria raised the Dutch programme’s positive predictive value from 4.5% to 16%. Those figures are on the page to show how far the threshold moves, not as thresholds to apply — the cut-off that governs a real result is the one published by the programme that issued it. The guideline’s recommended second tier is liquid chromatography-tandem mass spectrometry in preference to all other methods, with a cosyntropin stimulation test where that is unavailable, performed before corticosteroid treatment is started.

Related pages: the 17-hydroxyprogesterone unit converter for the units and the reference intervals by sex, cycle phase and age; the short Synacthen test interpreter for the same cosyntropin test read for adrenal insufficiency rather than for CAH; the androstenedione converter and the DHEA sulfate converter for the androgens that quantify the excess; and the aldosterone-to-renin ratio calculator, which is the other adrenal test whose interpretation turns on getting the sampling conditions right.

Frequently asked questions

What 17-OHP level is diagnostic of congenital adrenal hyperplasia?

At or above 30 nmol/L, about 1,000 ng/dL, either on an early-morning baseline or 60 minutes after cosyntropin: the non-classic CAH literature states that definitive diagnosis requires a 17-OHP concentration of at least 10 ng/mL (30 nmol/L) basally or after stimulation, and Chesover and colleagues report a stimulated peak above 30 nmol/L as 96% sensitive and 100% specific. Above about 100 nmol/L (3,300 ng/dL) classic disease becomes much the more likely reading, with Honour giving basal values in classic disease as usually above 100. Check the unit before applying any of these: 1 nmol/L is 33.05 ng/dL.

Does a normal 17-OHP exclude congenital adrenal hyperplasia?

It depends entirely on which sample it was. A normal cosyntropin-stimulated value — a peak below 30 nmol/L at 60 minutes — does exclude both classic and non-classic 21-hydroxylase deficiency, and it is the only measurement that does. A normal early-morning baseline below 6 nmol/L makes it unlikely but does not formally exclude it: the threshold is about 96% sensitive, so roughly one non-classic patient in twenty-five falls beneath it. And a normal baseline taken in the afternoon or at an unrecorded time excludes nothing at all, because 17-OHP is substantially lower later in the day and the threshold was calibrated on morning samples.

Why does the sample have to be taken before 8 am?

Because 17-OHP follows the ACTH circadian rhythm and is substantially lower in the afternoon than in the early morning, so the published baseline thresholds only apply to a morning sample. The Endocrine Society guideline specifies screening with an early-morning baseline measurement taken before 8 am by liquid chromatography-tandem mass spectrometry. The error runs one way: a high afternoon value is still high, and would have been higher at 8 am, but a normal afternoon value is uninterpretable for exclusion.

How do I convert 17-OHP between ng/dL and nmol/L?

Multiply ng/dL by 0.0302599 to get nmol/L, or divide nmol/L by that, which is the same as multiplying by 33.05. So 200 ng/dL is 6.05 nmol/L, 1,000 ng/dL is 30.3 nmol/L and 3,300 ng/dL is 99.9 nmol/L. A report in ng/mL or µg/L is the ng/dL figure divided by 100, so 2 ng/mL is 200 ng/dL. A conversion error here is the single most common way this test is misread, because the factor of 33 turns a normal result into a florid one or the reverse. The 17-hydroxyprogesterone unit converter does it in every direction.

Why do preterm babies fail the newborn screen so often?

Because they genuinely have higher 17-OHP than term infants, for reasons unrelated to 21-hydroxylase deficiency — an immature adrenal cortex, immaturity of the 11-beta-hydroxylase step, and the ACTH response to illness itself. The Endocrine Society guideline states this generates many false positives, and the numbers bear it out: in the Swedish programme the positive predictive value was 25% for full-term infants and 1.4% for preterm ones. The recommended remedy is to stratify cut-offs by gestational age rather than birthweight; van der Kamp and colleagues found gestational age the better predictor, and in the Dutch programme adopting gestational-age criteria raised the positive predictive value from 4.5% to 16%.

What is the difference between classic and non-classic CAH?

The degree of residual 21-hydroxylase activity, which puts them at opposite ends of one spectrum. Classic disease presents in the newborn period — ambiguous genitalia in a 46,XX infant, and in the salt-wasting form a crisis in the first two weeks of life — and Honour gives its basal 17-OHP as usually above 100 nmol/L. Non-classic disease reflects partial deficiency and presents later or not at all: premature pubarche in a child, or hirsutism, acne, oligomenorrhoea and subfertility in a woman, with basal values above 6 but below 100 nmol/L. Non-classic disease does not cause salt wasting and does not usually require daily glucocorticoid. Because the bands overlap, the distinction is made on the clinical picture, the mineralocorticoid axis and CYP21A2 genotyping rather than on a single 17-OHP.

Does the menstrual cycle phase matter?

Yes, by more than threefold. The corpus luteum produces 17-OHP, so the luteal reference ceiling is around 2.85 ng/mL (285 ng/dL, 8.6 nmol/L) against about 0.8 ng/mL (80 ng/dL, 2.4 nmol/L) for the early-morning follicular phase. A luteal-phase baseline between 6 and 8.6 nmol/L is inside the luteal reference interval and may be entirely physiological, so the sensible next step is a follicular-phase repeat rather than a stimulation test. Where the cycle is too irregular to phase — which is common in exactly the patients being investigated — the cosyntropin test is the cleaner route.

Does a stimulated 17-OHP identify carriers?

Not reliably, and it should not be used for that. Heterozygote values overlap the unaffected range: Honour gives a basal carrier range of 6 to 44 nmol/L against 2 to 9 nmol/L for unaffected men, so a raised value in that region does not distinguish a carrier from an unaffected person or from a non-classic patient. CYP21A2 genotyping is the tool for carrier determination, and it is what matters for reproductive counselling — including partner testing, since a person with non-classic disease who carries one severe allele can have a child with classic CAH.

Related calculators

References

Speiser PW, Arlt W, Auchus RJ, Baskin LS, Conway GS, Merke DP, Meyer-Bahlburg HFL, Miller WL, Murad MH, Oberfield SE, White PC. Congenital Adrenal Hyperplasia Due to Steroid 21-Hydroxylase Deficiency: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018;103(11):4043–4088. doi:10.1210/jc.2018-01865. Recommendation 1.2, verbatim: "We recommend that first-tier screens use 17OHP assays standardized to a common technology with norms stratified by gestational age." Recommendation 1.3: "We recommend that screening laboratories employ a second-tier screen by liquid chromatography-tandem mass spectrometry in preference to all other methods", with the technical remark that "if laboratory resources do not include liquid chromatography-tandem mass spectrometry, a cosyntropin stimulation test should be performed to confirm diagnosis prior to initiation of corticosteroid treatment". Recommendation 3.2: "In symptomatic individuals past infancy, we recommend screening with an early-morning (before 8 am) baseline serum 17-hydroxyprogesterone measurement by liquid chromatography-tandem mass spectrometry." On prematurity: "premature, sick, or stressed infants have higher levels of 17OHP than do term infants, generating many false positives", citing the Swedish programme in which "the positive predictive value was 25% for full-term infants but only 1.4% for preterm infants", and Dutch data in which "adopting gestational age criteria improved the positive predictive value of screening from 4.5% to 16%". The guideline publishes no numeric 17-OHP cut-off table, which is why this page prints none for the newborn screen.
Chesover AD, Millar H, Sepiashvili L, Adeli K, Palmert MR, Hamilton J. Screening for Nonclassic Congenital Adrenal Hyperplasia in the Era of Liquid Chromatography-Tandem Mass Spectrometry. J Endocr Soc. 2020;4(2):bvz030. doi:10.1210/jendso/bvz030. A baseline 17-OHP of 6 nmol/L or more gave sensitivity and specificity both 96%; a peak 17-OHP above 30 nmol/L at 60 minutes after 250 µg of ACTH confirmed nonclassic CAH with 96% sensitivity and 100% specificity. The authors note baseline samples were "spread throughout the day" against a guideline recommendation of 8 am collection.
Bachelot A, Chakhtoura Z, Plu-Bureau G, et al. Non-classic congenital adrenal hyperplasia due to 21-hydroxylase deficiency revisited: an update with a special focus on adolescent and adult women. Hum Reprod Update. 2017;23(5):580–599. doi:10.1093/humupd/dmx014. Verbatim: "A basal 17-OHP concentration ≥2 ng/ml (6 nmol/l) should be used for screening if more appropriate in-house cut-off values are not available"; "Definitive diagnosis requires a 17-OHP concentration ≥10 ng/ml (30 nmol/l), either basally or after cosyntropin-stimulation."
Honour JW. 17-Hydroxyprogesterone in children, adolescents and adults. Ann Clin Biochem. 2014;51(4):424–440. doi:10.1177/0004563214529748. Classic CAH: basal usually above 100 nmol/L, ACTH-stimulated up to 2,000 nmol/L. Non-classic CAH: basal above 6 nmol/L (2 ng/mL) but below 100, stimulated above 30 nmol/L (10 ng/mL). Heterozygote carriers: basal 6–44 nmol/L, against 2–9 nmol/L for unaffected males. "17-OHP concentrations are lower in the afternoon than the morning."
van der Kamp HJ, Oudshoorn CGM, Elvers BH, van Baarle M, Otten BJ, Wit JM, Verkerk PH. Cutoff Levels of 17-α-Hydroxyprogesterone in Neonatal Screening for Congenital Adrenal Hyperplasia Should Be Based on Gestational Age Rather Than on Birth Weight. J Clin Endocrinol Metab. 2005;90(7):3904–3907. doi:10.1210/jc.2004-2136. "The calculated median 17OHP cutoff level was lower for GA [12.6 µg/liter (38 nmol/liter)]" than "for BW [17.6 µg/liter (54 nmol/liter)]"; "GA is a better predictor of 17OHP than BW (R2 was 50.6 vs. 35.8%, respectively)".
Held PK, Bird IM, Heather NL. Newborn Screening for Congenital Adrenal Hyperplasia: Review of Factors Affecting Screening Accuracy. Int J Neonatal Screen. 2020;6(3):67. doi:10.3390/ijns6030067
Mayo Clinic Laboratories. Test ID: OHPG — 17-Hydroxyprogesterone, Serum, by liquid chromatography-tandem mass spectrometry. Reference values: adult males below 220 ng/dL; females follicular below 80 ng/dL, luteal below 285 ng/dL, postmenopausal below 51 ng/dL; term infants 0–28 days below 630 ng/dL, with preterm infants able to exceed that. Accessed 2026.
White PC, Speiser PW. Congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Endocr Rev. 2000;21(3):245–291. doi:10.1210/edrv.21.3.0398

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.