Polycystic ovary syndrome (PCOS) stands as the most prevalent endocrine disorder affecting women during their reproductive years. The complexities and diverse ways PCOS manifests often lead to diagnostic confusion. Despite its widespread impact and significant healthcare expenditure, a substantial number of women with PCOS remain undiagnosed. At the heart of this diagnostic landscape are the Rotterdam criteria, a consensus guideline that has become the cornerstone for diagnosing PCOS. This article aims to provide a comprehensive understanding of the Rotterdam Criteria For Pcos Diagnosis, exploring its evolution, components, and ongoing relevance in clinical practice.
The Evolution of PCOS Diagnostic Criteria: Setting the Stage for Rotterdam
The journey to a standardized diagnosis for PCOS has been marked by evolving understanding and debate. Initially described in 1935 by Stein and Leventhal, PCOS was recognized by a combination of symptoms including hirsutism, obesity, amenorrhea, and enlarged polycystic ovaries observed during surgery. Over time, the medical community sought to refine these observations into clear diagnostic guidelines.
The first formal attempt to define PCOS clinically came in 1990 from the National Institute of Child Health and Human Development (NIH). The 1990 NIH criteria defined PCOS by the presence of both hyperandrogenism (clinical or biochemical signs) and oligo- or chronic anovulation. While ultrasonographic evidence of polycystic ovaries was acknowledged, it was not considered a primary diagnostic feature, creating a divergence from practices in Europe where polycystic ovaries were seen as central to PCOS.
Caption: Ultrasound image illustrating polycystic ovarian morphology, a key component in Rotterdam criteria for PCOS diagnosis.
This discrepancy highlighted the need for international consensus. In 2003, a pivotal conference in Rotterdam, Netherlands, convened 27 PCOS experts from the European Society of Human Reproduction (ESHRE) and the American Society for Reproductive Medicine (ASRM). This meeting resulted in the “Rotterdam criteria,” a joint consensus statement that significantly broadened the diagnostic scope of PCOS.
The Rotterdam criteria revolutionized PCOS diagnosis by establishing a “two out of three” framework. According to these criteria, PCOS can be diagnosed if a woman presents with at least two of the following three features:
- Oligo-anovulation: Infrequent or absent ovulation, leading to irregular menstrual cycles.
- Hyperandrogenism: Clinical or biochemical signs of excessive androgens (male hormones).
- Polycystic Ovarian Morphology (PCOM): Characteristic appearance of polycystic ovaries on ultrasound.
This broadened definition had a significant impact, with some studies reporting up to a threefold increase in PCOS prevalence compared to the stricter 1990 NIH criteria. Crucially, the Rotterdam criteria allowed for the diagnosis of PCOS even in the absence of hyperandrogenism, challenging the previous view that hyperandrogenism was the primary defining characteristic.
Following the Rotterdam consensus, the Androgen Excess Society (AES) in 2006 reaffirmed the importance of ovarian morphology but emphasized hyperandrogenism as central to PCOS diagnosis. The AES guidelines required hyperandrogenism (hirsutism or biochemical hyperandrogenemia) alongside either oligo-anovulation or PCOM. This approach excluded the “mildest” PCOS phenotype – women with oligo-anovulation and PCOM but without hyperandrogenism.
Recognizing the clinical confusion arising from multiple diagnostic systems, the NIH convened another evidence-based workshop in 2012. Experts at this workshop recommended continued use of the broader 2003 Rotterdam criteria, further categorizing PCOS into four sub-phenotypes based on the presence or absence of each of the three Rotterdam criteria:
- Hyperandrogenism and ovulatory dysfunction.
- Hyperandrogenism and PCOM.
- Ovulatory dysfunction and PCOM.
- Hyperandrogenism, ovulatory dysfunction, and PCOM.
The Rotterdam criteria maintained their position as the most widely accepted diagnostic framework, receiving unanimous support in the 2018 International Evidence-Based Guideline for the Assessment and Management of PCOS. The subsequent sections will delve into each component of the 2003 Rotterdam criteria: hyperandrogenism, oligo-anovulation, and PCOM, providing a detailed understanding of their diagnostic application.
Decoding the Rotterdam Criteria: Component 1 – Hyperandrogenism
Hyperandrogenism, a cornerstone of the Rotterdam criteria for PCOS diagnosis, signifies an excess of androgens in women. Androgens, often termed “male hormones,” play crucial roles in women’s health, including bone density, muscle mass, and sexual function. In women, androgens are produced by the ovaries (25%), adrenal glands (25%), and peripheral tissues (50%), primarily in the forms of testosterone, dihydrotestosterone, dehydroepiandrosterone sulfate (DHEA-S), dehydroepiandrosterone (DHEA), and androstenedione (ANSD). The ovaries are the primary source of testosterone, while ANSD is produced equally by ovaries and adrenal glands, and DHEAS is exclusively produced by the adrenal gland.
In the context of PCOS, hyperandrogenism is frequently observed, particularly biochemical hyperandrogenemia, which is common in women with oligo-amenorrhea and PCOS. The ovaries are the main source of this excess androgen production in PCOS. Testosterone, especially in its free form (unbound to sex-hormone-binding globulin – SHBG), is the most sensitive and frequently elevated marker in PCOS-related hyperandrogenemia. Conditions like obesity and hyperinsulinemia, often associated with PCOS, can lower SHBG levels, further increasing free testosterone. Studies indicate that elevated free testosterone is found in up to 89% of PCOS patients with hyperandrogenemia, while total testosterone elevation is seen in 49–80% of cases. ANSD and DHEAS can also be elevated in PCOS, contributing to overall androgen excess.
Biochemical Hyperandrogenism: Measurement and Interpretation
Accurate measurement of serum androgens in women presents several challenges. Standard assays can be unreliable at the lower androgen levels typical in women. Furthermore, testosterone levels fluctuate throughout the day, and other steroids can interfere with assay results. To mitigate these issues, guidelines recommend avoiding direct assays on whole serum and instead using assays that incorporate extraction and chromatography, followed by mass spectrometry or immunoassay.
Free testosterone, representing only a small fraction of total testosterone (1-3%), requires highly precise measurement techniques. Equilibrium dialysis is considered the gold standard for free testosterone measurement but is technically demanding and expensive. Therefore, calculated estimations of free testosterone, such as the free androgen index (FAI) and calculated bioavailable testosterone, are commonly used and considered reliable for PCOS diagnosis. These calculated measures, along with high-quality assays for total and free testosterone, are recommended as primary markers for biochemical hyperandrogenism in PCOS. In cases where total or free testosterone levels are not elevated, assessing ANSD and DHEAS may provide additional diagnostic information. It is crucial to interpret androgen levels in the context of the specific laboratory’s reference ranges due to assay variability.
Current Recommendations for Biochemical Hyperandrogenism (Rotterdam Criteria Context):
- Biochemical hyperandrogenism is defined by elevated total or free testosterone, measured using high-quality assays like liquid chromatography-mass spectrometry or extraction/chromatography immunoassays.
- Calculated free testosterone, free androgen index, or bioavailable testosterone can also be used.
- Consideration of ANSD and DHEAS is recommended if total or free testosterone is not elevated.
- Interpretation of androgen levels must be guided by the reference ranges of the testing laboratory.
Clinical Hyperandrogenism: Visible Signs of Androgen Excess
Clinical hyperandrogenism refers to the visible manifestations of elevated androgen levels in women. The primary clinical signs include hirsutism, acne, and female pattern hair loss. Hirsutism, characterized by excessive terminal hair growth in a male-like pattern, is a common feature of PCOS, affecting 60–70% of patients. The Modified Ferriman-Gallwey (MFG) scoring system is the gold standard for quantifying hirsutism. This system assesses terminal hair growth at nine body sites on a scale of 0 to 4, with scores summed to provide an overall hirsutism score.
Caption: Table detailing features of PCOS diagnosis according to Rotterdam criteria, including clinical hyperandrogenism assessment using Modified Ferriman-Gallwey score.
MFG score thresholds for diagnosing hyperandrogenism vary from ≥3 to ≥8 across different studies and guidelines. Despite its widespread use, the MFG score has limitations, including interobserver variability, influence of hair removal practices, and ethnic variations in hair growth patterns. Current guidelines recommend considering ethnicity when interpreting MFG scores and suggest a threshold range of ≥4 to ≥6, which may lead to higher rates of hyperandrogenism diagnosis compared to the traditional threshold of ≥8.
Acne and female pattern hair loss are also common concerns among women with PCOS. While associated with biochemical hyperandrogenism, they are less reliable as diagnostic markers. Only about one-third of women with female pattern hair loss have elevated androgen levels. While scales like the Ludwig visual scale exist for assessing female pattern hair loss, there is no universally accepted visual assessment for acne severity in the context of PCOS diagnosis. Current recommendations emphasize managing these conditions in PCOS patients but do not advocate for their inclusion as primary diagnostic criteria within the Rotterdam framework due to limited reliability for PCOS diagnosis specifically.
Current Recommendations for Clinical Hyperandrogenism (Rotterdam Criteria Context):
- Clinical hyperandrogenism should be evaluated by a trained healthcare provider using the Modified Ferriman-Gallwey score.
- The threshold for “abnormal” MFG score may vary based on the patient population, ranging from ≥4 to ≥8.
- Acne and female pattern hair loss are not currently recommended as reliable diagnostic markers for PCOS within the Rotterdam criteria.
Decoding the Rotterdam Criteria: Component 2 – Oligo-anovulation
Oligo-anovulation, the second key component of the Rotterdam criteria for PCOS diagnosis, refers to infrequent or absent ovulation. Understanding normal menstrual cycle patterns is crucial for recognizing ovulatory dysfunction. The average menstrual cycle in adult women lasts 28 days, with a normal range of 21–35 days. Ovulatory cycles typically consist of a relatively constant luteal phase (around 14 days) and a more variable follicular phase. Regular monthly menses within this normal range are generally a reliable indicator of ovulatory function. However, even women with regular cycles can experience occasional anovulatory cycles.
Based on this understanding, current guidelines, within the context of Rotterdam criteria, recommend using irregular menstrual cycles as a marker for ovulatory dysfunction in PCOS. Menstrual dysfunction in PCOS typically manifests as oligo-amenorrhea, defined as cycles longer than 35 days apart or fewer than 8 cycles per year. While some guidelines include amenorrhea (absence of menstruation) as a criterion, oligo-amenorrhea is more commonly used for PCOS diagnosis.
In cases where a patient presents with strong clinical suspicion of PCOS but has regular menstrual cycles, polymenorrhea (frequent cycles), or an unclear menstrual pattern, further assessment of ovulation is recommended. Serum progesterone measurement or luteinizing hormone (LH) testing can confirm ovulatory status in these situations, as ovulatory dysfunction can occur even without classic oligo-amenorrhea. It is important to note that these recommendations primarily apply to adult women. Irregular menses are physiologically normal within the first year after menarche (first menstruation), and in the 1 to 3 years post-menarche, irregular menses are defined as cycles more than 45 days apart.
Current Recommendations for Oligo-anovulation (Rotterdam Criteria Context):
- Oligo-amenorrhea (cycles > 35 days apart or < 8 cycles per year) is used as a marker for ovulatory dysfunction.
- Ovulation confirmation with serum progesterone or LH testing is recommended for women with uncertain menstrual history or suspected ovulatory dysfunction despite seemingly regular cycles.
Decoding the Rotterdam Criteria: Component 3 – Polycystic Ovarian Morphology (PCOM)
Polycystic Ovarian Morphology (PCOM), the third component of the Rotterdam criteria, refers to the characteristic appearance of ovaries on ultrasound in women with PCOS. Historically, Stein and Leventhal initially described polycystic ovaries as bilaterally enlarged with a polycystic appearance based on surgical and pathological examinations. Since the 1980s, ultrasonography has enabled non-invasive assessment of ovarian morphology.
Early ultrasound criteria for PCOM, widely adopted in 1985 by Adams et al., defined PCOM as the presence of 10 or more follicles measuring 2–8 mm in size in one cross-section of an ovary using transabdominal ultrasound. Transabdominal ultrasound has largely been replaced by transvaginal ultrasound, offering higher resolution imaging. In 2003, Jonard et al. used receiver operator characteristic (ROC) curves to establish updated thresholds for PCOM, suggesting a follicle number per ovary threshold of 12 or more (2–9 mm diameter, mean of both ovaries) with 75% sensitivity and 99% specificity for PCOS diagnosis.
The 2003 Rotterdam criteria adopted these findings, defining PCOM as either 12 or more follicles (2–9 mm diameter) or an ovarian volume greater than 10 cm³ in either ovary. While classic descriptions of PCOM included features like peripheral follicle distribution (“string of pearls”) and increased stromal echogenicity, these stromal markers have not proven to be significantly predictive for PCOS diagnosis when used alone or in combination with follicle number and ovarian volume. Consequently, these features were excluded from the PCOM definition in 2003 and subsequent criteria.
Over the past decade, debate has centered on refining PCOM thresholds. Studies revealed that using the 2003 Rotterdam criteria, 30–50% of normo-androgenic, ovulatory women would meet PCOM criteria, suggesting potential overdiagnosis. This led to investigations into revised thresholds. Advances in ultrasound technology, particularly higher transducer frequencies (≥8 MHz), improved antral follicle detection, necessitating higher follicle number thresholds. Furthermore, variations in study control groups (some excluding healthy women with PCOM) contributed to differing proposed cut-offs.
In 2014, an Androgen Excess and PCOS (AEPCOS) society task force report recommended raising the PCOM threshold to ≥25 follicles per ovary and/or an ovarian volume ≥10 cm³, based on transvaginal ultrasound with ≥8 MHz transducer frequency. However, this higher threshold has been challenged for potentially excluding a significant group of oligo-anovulatory women and limiting diagnostic sensitivity without substantial improvement in predicting metabolic risk. More recently, the 2018 International Evidence Based Guidelines for PCOS assessment and management proposed a slightly reduced follicle number threshold of ≥20 follicles per ovary and/or ovarian volume ≥10 cm³.
The clinical relevance of varying follicle number thresholds beyond providing a diagnostic label is debatable. The degree of hyperandrogenemia and oligo-anovulation are stronger predictors of metabolic risk and clinical relevance in PCOS than ovarian morphology alone. Ultrasound criteria for PCOM are likely to continue evolving with technological advancements.
Given the heterogeneity in ultrasound assessment of PCOM, anti-Müllerian hormone (AMH) has been explored as a potential surrogate marker. AMH, secreted by granulosa cells of small ovarian follicles, is significantly higher in women with PCOS. While promising, standardization of AMH measurement and consistent diagnostic thresholds are still needed. Current guidelines do not recommend AMH as an alternative to ultrasound for PCOM detection or as a single test for PCOS diagnosis within the Rotterdam framework. Follicle number and AMH levels naturally decline with age in women, with and without PCOS. Age-specific thresholds for AMH and PCOM are being investigated to potentially improve diagnostic accuracy across different age groups. While not yet widely adopted, age-stratified criteria may represent a future refinement in PCOS diagnosis, particularly for older women.
Current Recommendations for PCOM (Rotterdam Criteria Context):
- PCOM is defined as either ≥20 follicles per ovary and/or an ovarian volume ≥10 cm³ in either ovary, using transvaginal ultrasound with a transducer frequency of 8 MHz or higher.
- AMH is not currently recommended as an alternative marker for PCOM or as a single diagnostic test for PCOS.
Conclusion: The Enduring Relevance of the Rotterdam Criteria for PCOS Diagnosis
The Rotterdam criteria remain the cornerstone for PCOS diagnosis, requiring the presence of at least two out of three criteria: oligo-anovulation, hyperandrogenism, and/or polycystic ovarian morphology (PCOM). While refinements and ongoing discussions continue, particularly regarding PCOM criteria and testosterone assay standardization, the Rotterdam framework has significantly advanced PCOS diagnosis.
The adoption of the Rotterdam criteria broadened the diagnostic net for PCOS, increasing prevalence estimates. However, this broader definition also encompasses diverse PCOS phenotypes, potentially complicating research into underlying pathophysiology, risk stratification, and targeted treatment strategies. Future research should focus on characterizing specific PCOS phenotypes within the Rotterdam framework rather than treating “PCOS as defined by Rotterdam” as a homogenous entity.
PCOS diagnosis should not be made lightly. Receiving a PCOS diagnosis can be psychologically distressing, impacting well-being, increasing depression risk, and raising concerns about future health and fertility. Many women report feeling inadequately informed about their diagnosis. Furthermore, diagnostic delays are common. While the burden associated with PCOS diagnosis may stem from the condition itself or the diagnostic process, timely and appropriate diagnosis is crucial. Given ongoing debates about diagnostic criteria and the need for improved clinical care, revisiting and refining PCOS diagnostic approaches within the Rotterdam framework remains a critical endeavor to optimize patient outcomes and advance our understanding of this complex syndrome.
