Are expanding disease definitions unnecessarily labeling women with PCOS?
Click here to read the Cover Story, "PCOS, hypothalamic amenorrhea present confusing picture for diagnosis, treatment."
Patient anxiety is a concern, and there can be anxiety from both underdiagnosing and overdiagnosing PCOS.
When talking about expanding disease definitions it’s important to understand how we got to the current criteria for diagnosing polycystic ovary syndrome. In the 1990s, we had the NIH definition that included irregular menses and high androgen levels or excessive hair growth/hirsutism, so there were two criteria. Then in 2004, the Rotterdam criteria were published, and we added certain ovarian features seen on ultrasound imaging. That’s what expanding the definition means, and now we have three criteria.
You would think that with three criteria, fewer women will be diagnosed with PCOS. However, only two of the three criteria need to be met, so this greatly broadened the population of women who meet the diagnosis. Some women can have irregular menses and the ovarian features on ultrasound, but not high androgen levels, and there is a group of physicians and researchers who do not consider them to have PCOS, because high androgen levels are central to the pathophysiology of PCOS.
Overdiagnosis has been discussed often within our PCOS community. At a 2012 consensus meeting at the NIH, we understood the importance of focusing on hyperandrogenism in PCOS, but there was a concern that the group of women with the milder phenotype of irregular menses and ultrasound features only would be left out. If we do not include this phenotype as having PCOS, who will take care of their reproductive issues? So, it was decided to keep that phenotype in the definition of the syndrome, that is, to use the Rotterdam criteria.
Now, our concern is that some women with a milder phenotype may be reading everything related to PCOS on the internet and become concerned unnecessarily. Women without high androgen levels should be less concerned about certain issues associated with PCOS. So, that’s where issues of labeling — causing anxiety about the syndrome — come in.
Are we at a point where we need to readdress definitions and criteria for PCOS diagnosis, or do we just need to make health care providers aware of the issue so they counsel patients better? I believe health care providers need to tell a patient if her phenotype is a mild or a full form. If it’s mild (normal androgens), she is likely not going to have increased risks for diabetes or high cholesterol, for example. However, if her periods are irregular, she may need help with fertility. So, personalized care is really what is needed in PCOS.
We don’t have as much information in PCOS about anxiety specifically in those who have a mild phenotype or related to overdiagnosis, but for other diseases, there is more information about how overdiagnosis can affect a patient’s psychological health. There are two groups of women who have a potential to be overdiagnosed: young girls and those women who mimic functional hypothalamic amenorrhea. In young girls going through puberty, we do not have good cutoffs for male hormone levels or ultrasound criteria. Because of the limitations with accurate diagnoses and the overlap with puberty symptoms, this is a group we should be cautious to avoid over-labeling.
Anuja Dokras, MD, PhD, is a professor of obstetrics and gynecology at the University of Pennsylvania and director of PENN PCOS Program. Disclosure: Dokras reports she is a consultant for Fractyl, Medtronic and has grant funding from the NIH.
Overall, PCOS is not overdiagnosed, and early diagnosis with targeted counseling can prevent complications and improve patient satisfaction.
In recent publications, we have seen the prevalence of PCOS swell to 20% of reproductive-age women. There are several possible explanations for this observed epidemiologic trend: using the broader Rotterdam criteria may be a driver as this widens the diagnostic net; the rate of PCOS may be rising secondary to increased obesity rates; more sensitive androgen assays allow more timely and accurate diagnosis of PCOS; and diagnoses may result from improved patient and physician awareness.
Several distinct phenotypes meet diagnostic criteria for PCOS. The universal adoption of the Rotterdam criteria means the definition of PCOS has been broadened to include women with a phenotype of PCOS that is non-hyperandrogenic (ovulatory dysfunction plus polycystic ovarian morphology). In the non-hyperandrogenic phenotype, the degree of endocrine and metabolic dysfunction is considered mild. The metabolic features and risks for PCOS cluster in the phenotype defined by the classic NIH criteria (hyperandrogenism plus ovulation dysfunction). Approximately two-thirds of patients who are diagnosed with PCOS present with clinical or biochemical hyperandrogenemia, as well as some form of ovulatory dysfunction. These patients with the “classic” phenotype should be counseled extensively with the goal of early prevention of long-term reproductive, dermatologic and metabolic sequelae.
Whether patients with the non-hyperandrogenic phenotype are mislabeled as having PCOS is controversial. These patients should certainly be counseled regarding their likely low risk for metabolic complications of PCOS to avoid unnecessary concern. Providing personalized counseling to target the specific phenotype is indicated.
Labeling adolescents and young adult women who may have only transient symptoms with a diagnosis of PCOS represents a potential pathway for overdiagnosis and unnecessary anxiety. Although the diagnostic criteria are the same for adolescents, delaying diagnosis of PCOS for at least 2 years after menarche helps avoid this pitfall. When there are persistent menstrual irregularities in the setting of increased serum androgen levels and/or progressive hirsutism, PCOS is an appropriate diagnosis.
Early diagnosis of PCOS is helpful for most women with bothersome signs and symptoms. A recent study showed that most women are frustrated by the length of time it takes to reach diagnosis. Delayed diagnosis contributes to patient dissatisfaction due to the feeling that the diagnosis has been missed and their symptoms are being dismissed as “normal.” Many women with PCOS experience relief and satisfaction with the diagnosis of PCOS. Early diagnosis also helps alleviate anxiety and depression by enabling more directed treatment for acne, hirsutism and menstrual problems.
PCOS is an extremely prevalent condition that likely affects 1 in 5 women to some degree. Our job as physicians is to include women in this discussion and explain accurately where they fit in the spectrum of metabolic risk so that together we can devise a personalized management strategy.
- Gibson-Helm M, et al. J Clin Endocrinol Metab. 2017;doi:10.1210/jc.2016-2963.
Jennifer Glueck, MD, is an assistant professor in the division of endocrinology at Rush University Medical Center in Chicago. Disclosure: Glueck reports no relevant financial disclosures.