Home / Specialties / Women's Health / Andrea Dunaif Part 4, Is It Hard To Diagnose PCOS?

Andrea Dunaif Part 4, Is It Hard To Diagnose PCOS?

In part 4 of this Exclusive Interview, Andrea Dunaif talks with Diabetes in Control Medical Editor Joy Pape about the ease of diagnosing PCOS.

Andrea Dunaif, MD is system chief of endocrinology, diabetes & bone disease for Icahn School of Medicine at Mount Sinai Health System in New York, NY.

Transcript of this video segment:

Pape: So, you can answer the question, is PCOS difficult to diagnose? I know a lot of people say it is. What do you say?

Dunaif: It’s really very easy to diagnose. The symptom of infrequent menstrual periods, eight or fewer per year, six or fewer per year. Women are aware of that. It’s a big deal when you miss a period. And so, women can really accurately self record their menstrual frequency. And 80% of women with that symptom, six or eight or fewer menses per year turn out to have PCOS just on a single blood test. So, it’s not that you have to repeat blood tests or there’s any difficulty and that’s just an elevation of testosterone or free testosterone, so in that group it’s very easy to make the diagnosis. Most women who have increased hair will turn out to have PCOS 80% or 90%. So, it’s not difficult at all. I think what’s happened is physicians say, “Oh, it’s such a complicated and heterogeneous disorder,” I say that’s the dark and stormy night of the field — PCOS as a complex and heterogeneous disorder that, oh, no, they can’t diagnose it, and yet the women are going online and reading the symptom online and saying, “Gee, I think that’s what I have.” And they’re right most of the time.

Pape: Mm-hmm. So, what other tests are needed to diagnose? What you’d mentioned? Do you do any other testing?

Dunaif:  So, we do the testosterone, the total, and what we call, the best one is the bioavailable. And it makes a difference how that’s done, there’s now a wider standardization of the high quality method which is liquid chromatography, Tandem Mass Spec. And that the CDC is now requiring laboratories to do that and providing standards. And then, we mainly have to rule out other conditions. And the conditions that we rule out are pituitary conditions, though they’re really very uncommon, but high prolactin, thyroid conditions, again, very uncommon presenting as PCOS but inexpensive to exclude. We make sure that the ovaries are functioning normally, that there’s no premature ovarian failure, and we make sure that there’s not an inherited condition of the adrenal gland production of male hormones, 21 hydroxylase deficiency, but again that’s uncommon except in certain ethnic groups. So, most of the time it’s PCOS.

Pape: And so, what’s the best test to screen if you were random screening?

Dunaif:  So, well done testosterone, a total, and a free, because 90% of the time it will be some measure of the free testosterone. And by that we mean the testosterone that’s not bound to sex hormone-binding globulin, it’s either circulating free or it’s weakly associated with albumin. And that’s very easy to do if you measure the total testosterone and the sex hormone-binding globulin. You can Google free testosterone calculator, and you plug your numbers in, and it just by the equations and mass action laws will calculate for you what’s free.

Pape: So, I think you’ve answered this question, but does someone need to obtain an ovarian ultrasound to make the diagnosis of PCOS?

Dunaif: Absolutely not. What I call the endocrinologist form of PCOS , the old NIH criteria, and that’s the women with the irregular, infrequent menstrual cycle and the high androgen levels, and that’s the form that’s associated with the increased metabolic risk. The women who fall into the other diagnostic criteria, the so-called Rotterdam Criteria, which include women with polycystic ovarian changes on ultrasound and high androgen levels or polycystic ovarian changes on ultrasound and irregular periods. Both those other groups don’t have increased metabolic risk or if they do it’s very slight. So, when we’re talking as diabetes health care providers about the form of PCOS that we should be worried about, it’s the old classic NIH PCOS that’s easy to diagnose without an ultrasound.

Return to the main page.