How Is PCOS Diagnosed? The Rotterdam Criteria Explained
How is PCOS actually diagnosed, and can normal bloodwork mean a prior diagnosis was wrong?
PCOS is diagnosed by criteria, not by a single test: an adult needs two of three findings — irregular or absent ovulation, hyperandrogenism (either visible signs or elevated androgens on labs), and polycystic ovarian morphology on ultrasound or, since 2023, an elevated AMH — after other causes have been excluded. Normal bloodwork on its own does not rule PCOS out, because clinical signs of excess androgen count as a full criterion in their own right. In fact, when irregular cycles and hyperandrogenism are both present, the international guideline says no imaging or AMH is needed to make the diagnosis at all.
What exactly are the Rotterdam criteria?
The framework started with the 2003 Rotterdam consensus and was carried forward by the 2018 and 2023 International Evidence-based Guidelines, which ASRM and ESHRE both endorse. In adults, at least two of the following three are required:
- Oligo-ovulation or anovulation — irregular or absent cycles. The guideline defines "irregular" precisely: cycles shorter than 21 or longer than 35 days, or fewer than eight cycles a year, from three years post-menarche to perimenopause.
- Clinical or biochemical hyperandrogenism — visible signs of excess androgen, or elevated androgens on bloodwork. Either route satisfies the criterion.
- Polycystic ovarian morphology — 20 or more follicles in at least one ovary, or an ovarian volume of 10 mL or more. This is a follicle-counting finding, not the "cysts" the name suggests.
Two important updates most summaries still miss. First, since 2023 a raised AMH can be used instead of ultrasound in adults. Second, the criteria work differently in adolescents: within eight years of a first period, ultrasound and AMH are not used at all because multi-follicular ovaries are normal at that stage, and both irregular cycles and hyperandrogenism are required.
Why doesn't normal bloodwork rule out the diagnosis?
Because the hyperandrogenism criterion has two equally valid routes, and labs are only one of them. Clinical hyperandrogenism — hirsutism above all, and also female-pattern hair loss and acne — meets the criterion on its own. These are alternative paths in the framework, not a primary test with a fallback.
The guideline is also candid about why labs can look unremarkable in someone with obvious symptoms. It specifically recommends LC-MS/MS assays for testosterone, because the direct immunoassays most standard panels use "have limited accuracy and demonstrate poor sensitivity and precision for diagnosing hyperandrogenism in PCOS." It adds that biochemical testing matters most in patients with minimal clinical signs — the reverse of how it is often used. And it notes that androgen levels are very difficult to interpret at all on the combined oral contraceptive pill, which raises SHBG and suppresses androgen production; a meaningful androgen panel needs about three months off the pill first.
One honest caveat in the other direction: not all clinical signs carry equal weight. Hirsutism alone is considered predictive of biochemical hyperandrogenism and PCOS in adults, while isolated acne or female-pattern hair loss are weaker, less specific predictors. So a patient with clear hirsutism and normal labs is not a gray area — she meets the criterion as written. Acne alone is a softer case that deserves a fuller look rather than a quick yes or no. Our piece on why PCOS is a metabolic condition rather than an ovary problem covers why standard testing so often misses the wider picture even in confirmed cases.
What should a thorough PCOS re-evaluation include?
Repeating labs years later is reasonable clinical practice — hormone levels and clinical pictures genuinely shift, and reassessment is not automatically dismissive. What makes a re-evaluation incomplete is disputing PCOS as the explanation for a real, ongoing finding like absent periods without offering an alternative explanation for that same finding.
Excluding other causes is not an optional extra here; it is written into the diagnostic algorithm. The guideline's exclusion workup is TSH, prolactin, and 17-hydroxyprogesterone (for non-classic congenital adrenal hyperplasia), plus FSH and LH where hypothalamic amenorrhea is possible, and further testing for Cushing's syndrome or an androgen-secreting tumour if the clinical picture warrants it. Absent or irregular cycles deserve an explanation whatever label ends up applying — "not PCOS," full stop, is not a finished workup. Our post on whether you can have PCOS while still ovulating covers the same question from the other direction, and PCOS at a normal BMI covers another presentation that routinely gets missed.
What does this mean if you're questioning an old diagnosis?
If the original diagnosis rested on a genuine clinical picture — irregular or absent periods, visible signs of excess androgen, imaging consistent with polycystic ovarian morphology — it was very likely made correctly against the actual standard, even if bloodwork at the time read normal. That does not mean re-evaluation is never warranted; presentations evolve, and a second set of eyes has real value. But a diagnosis being questioned is not by itself evidence the first one was wrong. Often it reflects how completely the criteria were applied the second time.
The practical move is to ask, plainly: which of the three criteria were assessed, which were met, and were thyroid, prolactin and 17-OHP checked to exclude the alternatives? Those are answerable questions, and the answers tell you whether you are looking at a corrected diagnosis or an incomplete one.
What's the bottom line on PCOS diagnosis?
PCOS follows a structured, internationally agreed framework rather than clinical impression, and normal bloodwork does not rule it out when clear clinical signs of hyperandrogenism sit alongside anovulation or polycystic ovarian morphology. If an old diagnosis is being disputed, the questions worth asking are whether the full criteria were applied, whether the pill or an insensitive assay was affecting the androgen results, and whether a real alternative explanation has been offered for symptoms like absent periods. Here in Huntington, this diagnostic limbo is one of the most common things patients across Long Island bring to us. Our PCOS and endometriosis care approach covers how we work through a full, criteria-based picture alongside your medical team.
This article is for general educational purposes and isn't a substitute for individualized guidance from your OB-GYN or endocrinologist. Questions about a PCOS diagnosis, past or present, should be discussed directly with your care team.
East to West Fertility is a metabolic and immune-focused fertility clinic in Huntington, Long Island, serving patients across Long Island, NYC, and beyond. Learn more about our Metabolic & Immune Fertility Evaluation or call 631-416-4940.
References:
- Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. "Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome." Fertil Steril, 2004.
- Teede HJ, et al. "Recommendations from the 2023 International Evidence-based Guideline for the Assessment and Management of Polycystic Ovary Syndrome." J Clin Endocrinol Metab, 2023. doi:10.1210/clinem/dgad463.
- International PCOS Network. 2023 International Evidence-based Guideline, Algorithm 1 (diagnostic algorithm and exclusion testing).
