PCOS Diagnosis Without an Ultrasound: When It's Still Valid

Can you be diagnosed with PCOS without an ultrasound?

Two of the three Rotterdam criteria, met without any imaging, are enough for a PCOS diagnosis in an adult. If you already have irregular cycles and clinical or biochemical androgen excess, you have satisfied two of three — and the 2023 international guideline states that ovarian imaging is not required once that bar is cleared.

This works because the Rotterdam rule was written as “any two of three,” not “these three in a fixed order.” Ovarian appearance is one of the three features, not a gatekeeper the other two have to pass through. A scan adds nothing to a diagnosis that two other features have already completed, beyond cost and a waiting-list appointment.

Note: in May 2026, PCOS was renamed polyendocrine metabolic ovarian syndrome, or PMOS, by a global consensus of more than 50 organisations. The criteria in this article are unchanged by that — only the name on the diagnosis is different. This article uses PCOS because that is still what most readers search.

Can AMH replace ultrasound for a PCOS diagnosis?

A 2024 meta-analysis of 68 adult studies found that AMH alone identifies PCOS with 79% sensitivity (95% CI 76–82%) and 87% specificity (95% CI 84–89%) — accurate enough that the 2023 guideline lists it as an alternative to ultrasound in adults who need the third feature confirmed, not an addition to it.

That matters for anyone the scan itself is a barrier for: no need for a pelvic exam, no need to be sexually active for a transvaginal probe to be appropriate, and a single blood draw instead of an ultrasound appointment that may have a multi-week wait. The guideline is explicit that ultrasound and AMH are alternatives, not a pair to run together — ordering both when one already answers the question adds cost without adding accuracy.

Why the guideline built this shortcut in at all

Ultrasound used to be treated as close to mandatory, ordered by default rather than by need. The 2023 guideline panel reviewed that pattern directly and concluded it was adding cost, appointment delay and, in some cases, unnecessary anxiety over incidental findings — without changing a single diagnosis in the group who already had two of the three features confirmed on history and bloodwork alone.

The guideline’s own diagnostic algorithm reflects that conclusion in its structure: cycle history and androgen assessment come first, and ovarian imaging is positioned as the step you take only when the first two did not already answer the question. Ordering it earlier than that is not wrong, exactly — it is simply a test run before its result could change anything.

Who still genuinely needs imaging or a blood test?

Anyone with only one of the first two features — irregular cycles without confirmed androgen excess, or androgen excess with regular cycles — needs the third feature checked one way or the other, by scan or by AMH, to reach a diagnosis at all.

The regular-cycles-plus-androgen-excess group is worth naming specifically, because it is the one most often surprised by needing a scan or AMH at all. Regular ovulation feels like reassurance that nothing else is wrong, so acne, hirsutism or hair thinning in that context often gets treated as a skin or dermatology issue rather than run through the full three-criteria check. Confirming the third feature is what turns “androgen excess with normal cycles” into either phenotype C — the Rotterdam combination requiring androgen excess plus polycystic ovaries — or a different diagnosis entirely, and that distinction changes what gets treated and how.

This is also where phenotype changes what “confirmed” looks like in practice. Lean phenotypes are the group most likely to have subtle biochemical androgen excess that a standard testosterone panel underreads, which is exactly the situation where a properly done diagnostic work-up matters more than usual — a normal-looking testosterone result in someone lean does not rule out androgen excess the way it more reliably does in someone whose baseline SHBG is lower. In this group, confirming the third feature with AMH rather than waiting on an ultrasound referral is often the faster path to a usable answer.

If you are under 20, the rule is different

Adolescents need both hyperandrogenism and ovulatory dysfunction together — not two of three — and neither ultrasound nor AMH is recommended for diagnosis within eight years of a first period, because normal puberty produces the same ovarian appearance and hormone fluctuation the criteria are trying to detect in adults. A scan or blood test done anyway in this age group is not part of the diagnostic pathway, whatever the result shows.

What skipping unnecessary imaging actually saves you

In an observational survey of 1,385 women with a reported PCOS diagnosis, 33.6% said it took more than two years and 47.1% saw three or more health professionals before the diagnosis was made. Referral waits for imaging are one of several ordinary bottlenecks inside that number — not the only one, but a removable one for anyone who already has two of the three features on paper.

Knowing you may not need a scan at all does not shorten a wait you are already in. It does mean you can ask, directly, whether your existing results already meet the threshold before agreeing to another appointment.

Where this will not apply to you

The “no ultrasound needed” shortcut only holds under specific conditions, and it is worth checking which side of them you are on before assuming it applies:

• Your androgen testing was not done reliably. Biochemical androgen excess should be measured by mass spectrometry, not a direct immunoassay alone. A borderline result on a less reliable assay may not count as confirmed, and imaging or AMH becomes the tiebreaker.
• You are within eight years of your first period. The adult shortcut does not apply, and neither test is used for diagnosis at this stage.
• You only have one of the first two features. The third feature still has to be confirmed by ultrasound or AMH — there is no way around checking it.
• You are on hormonal contraception. Both androgen testing and AMH are affected, and a washout is needed regardless of imaging.

Your next step

Pull your existing chart notes and check for two things: a documented cycle pattern (dates, not impressions) and a testosterone result with the assay method noted. If both are there and both are abnormal, ask directly whether an ultrasound referral is actually necessary before you book one.

If you are missing one of the two, that is the specific gap to close next — not a scan by default, but whichever of cycle tracking or a properly run androgen panel you do not yet have. The diagnostic criteria in full are set out here, and what the ultrasound itself measures, for when you do need it, is covered separately. Whichever way the diagnosis is reached, it is also what determines the ICD-10 code that ends up on your chart, and the complete diagnostic work-up, from cycle history through imaging or AMH, picks up from a confirmed diagnosis onward.