In women, the definition of polycystic ovary syndrome (PCOS) has become broad and includes several possible phenotypes. Because several features of PCOS may be in evolution in adolescents, we suggest that only firm criteria should be used to make a diagnosis of PCOS during adolescence. Hyperandrogenism, oligomenorrhea, and ovarian morphology change during adolescence and are discussed individually. Adolescents with incomplete criteria for a firm diagnosis of PCOS should be followed up carefully and may be diagnosed at a later time.
In recent years, the diagnosis of polycystic ovary syndrome (PCOS) has broadened to allow for several possible phenotypes. At the present time there are 3 different criteria for the diagnosis of PCOS. All require the exclusion of other disorders such as congenital adrenal hyperplasia and tumors. The National Institutes of Health criteria require evidence of hyperandrogenism and menstrual irregularity, without knowledge of ovarian findings. The European Society of Human Reproduction and Embryology/American Society for Reproductive Medicine criteria, often called Rotterdam, includes ultrasound findings of a polycystic ovary and allows for various phenotypes based on a combination of any 2 of the 3 findings of hyperandrogenism, menstrual irregularity, and polycystic ovaries on ultrasound. Finally, criteria set forth by the Androgen Excess and PCOS Society stress hyperandrogenism, with the associated findings of either menstrual irregularity and/or polycystic ovaries on ultrasound.
While in the adult the diagnostic criteria have been broadened, allowing for several different phenotypes, we believe this is different for the adolescent. In adolescents, using a broader definition may be problematic in that during the transition of girls into adulthood, several features may be in evolution, or in fact may only be transitory findings. Thus, prematurely assigning a diagnostic label of PCOS to an adolescent may be problematic and may result in unnecessary treatments and impose psychological distress, which by itself may be associated with several concerns including those of body image and concerns over future fertility. An incorrect diagnosis may also jeopardize clinical and basic studies of PCOS that require greater homogeneity and certainty about the diagnosis. Nevertheless, most recently there has been growing attention to diagnosing PCOS in adolescence, due in part to the importance of early intervention of this disorder in young women. Furthermore, there is no agreement concerning how to diagnose PCOS in adolescence. Some authors have even suggested avoiding making the diagnosis until the age of 18 years while others have proposed specific and very strict criteria for the diagnosis. One suggests the need for at least 4 of the 5 following criteria: clinical hyperandrogenism, biological hyperandrogenemia, insulin resistance and hyperinsulinism, oligoamenorrhea persisting for 2 years postmenarche, and polycystic ovaries on ultrasound. For us, this approach raises some significant questions because of the variability and inconsistency of a number of these findings during the periadolescent years. Hence we sought to critically review elements thought to encompass the diagnosis of PCOS that may emerge during adolescence, and attempt to establish firm criteria upon which to base the diagnosis.
Hyperandrogenism
For the purpose of this discussion we will exclude adolescents from the diagnosis of PCOS, who have hyperandrogenism due to 21 hydroxylase deficiency nonclassic adrenal hyperplasia, androgen secreting neoplasms, androgenic/anabolic drug use, Cushing syndrome, hyperandrogenic-insulin resistance-acanthosis nigricans syndrome, thyroid dysfunction, and hyperprolactinemia. Further, even though adolescents with these disorders may have findings associated with PCOS, we prefer to view PCOS as a separate entity. Additionally, the scope of this review does not allow us to comment on the predisposing factors that lead to development of PCOS in adolescents such as obesity, early pubarche/adrenarche, or frank insulin resistance. However, data do suggest that early exposure to elevated androgens (usually from an adrenal source) may result in findings consistent with PCOS. Also to be kept in mind is that if the diagnosis is suspected, these young women should be prospectively monitored since it has been clearly demonstrated that this population with increased androgen levels is at increased risk of development of the metabolic syndrome, independent of obesity and insulin resistance.
Definition of increased androgens
Recently the Endocrine Society issued a position statement discussing the utility, limitations, and pitfalls in measuring testosterone. It was concluded that in children as in women, total testosterone determinations should be carried out only with “assays of sufficient sensitivity” and most importantly “in conjunction with appropriate normative data.” It is suggested that when assessing total testosterone in children, one should use immunoassays after extraction or ligand chromatography/mass spectrometry-mass. Routine clinical assays, which use a “direct” assay method, without organic solvent extraction or chromatography, are often inaccurate and not sufficiently precise for this purpose. In adolescents with some menstrual cyclicity, it is suggested that testosterone be sampled in the early follicular phase. Early morning (<8:30 am ) is preferred for the measurement of serum testosterone and particularly for 17-hydrox progesterone, which should be drawn as part of a screen to rule out nonclassic adrenal hyperplasia, which is as common as 1/1000 in certain at-risk populations. Serum 17-hydrox progesterone has a diurnal variation and thus could appear as nondiagnostic if sampled in the mid-afternoon. An early morning screening value of <200 ng/dL is sufficient to rule out this disorder as the cause of oligomenorrhea. Further, measures of androstenedione are also higher in the morning due to the strong adrenal component of this hormone, which is under adrinocorticotropic hormone regulation. Thus we would suggest that screening blood be obtained in the early morning. Very high testosterone levels (>200 ng/dL) may suggest the possibility of the presence of an androgen-secreting tumor while measurement of prolactin and thyroid hormones may be useful in initial screening of hirsute patients with irregular menses.
Clinical suspicion of Cushing syndrome requires several assays of urinary cortisol and often a dexamethasone test. These tests should be performed in patients presenting with rapidly progressing hirsutism associated with abdominal obesity and clinical signs of hypercortisolism.
Elevated androgens
The most frequently agreed upon hormone used to define PCOS is an elevated testosterone (>2 SD above the mean) for the assay that is used. For an assay performed after appropriate extraction, an elevated value should be >55-58 ng/dL. A free testosterone (non-sex hormone binding globulin bound or unbound testosterone) of >15 ng/dL is also commonly used but most commercial assays have poor validity in females and this probably should not be used. Delta 4–androstenedione and dehydroepiandrosterone may also be assessed and values of >244 ng/dL and >248 μg/dL, respectively, are suggested to be upper normal range values.
Clinical markers of hyperandrogenism
Most diagnostic criteria include hirsutism, acne, or androgenic alopecia as markers of hyperandrogenism. However, acne is quite common during the adolescent years and in most subjects is a transitory phenomenon. In addition, very few data exist about adolescent androgenic alopecia. Accordingly, we do not suggest the use of these clinical markers as criteria for the diagnosis of PCOS in adolescents. Hirsutism may represent a better marker of hyperandrogenism and some authors have reported that progressive hirsutism during adolescent years may be an important sign of PCOS. However, we believe it is more prudent to rely on the presence of hyperandrogenemia, a classic character of PCOS that generally appears at puberty, rather than on biological signs for the diagnosis of PCOS in adolescents.
In fact, it has been reported that documented hyperandrogenemia is relatively constant and may represent the most important symptom of PCOS during adolescence.
Chronic anovulation and menstrual irregularities
Chronic anovulation, generally presenting with oligomenorrhea or secondary amenorrhea, is one of the key elements for the diagnosis of PCOS in the adult. However, in adolescence, chronic anovulation and menstrual irregularities are very common and approximately 40-50% of adolescent girls (gynecologic age, 0.5-5) have anovulatory cycles. Some girls may have chronic anovulation in spite of regular cycles. There is a progression toward more ovulatory cycles with increasing gynecological age; the prevalence of ovulatory cycles increases from 23-35% during the first year after menarche to 63-65% in the fifth year after menarche, and to 70-80% of women with a gynecological age of 6-10 years. It also has been reported that in patients having delayed menarche, normalization of menstrual cycles may take up to 8-12 years. At 13 years of age, the 5th-95th percentile in days between consecutive menses ranges from 19–64 days.
It has been suggested that menstrual cycles lasting 40-45 days be considered normal until 2-3 years after menarche. However, 35 days for an upper limit of the menstrual interval may be more correct since prospective studies have shown that 98% of girls with cycles between 21 and 34 days have normal cycles during adult age whereas the same is true only for 66% of adolescent girls with cycles lasting 35-40 days. The cause of the longer intercycle intervals during the first 12-18 months postmenarche may be due to both inadequate follicular development as well as complete anovulation.
Adolescent menstrual disorders as predictors of PCOS
Since most adolescent girls become ovulatory with age, it has been suggested that it is unnecessary to carry out any investigation at this time. The dilemma exists however, since 10-20% of anovulatory girls will remain anovulatory and 26-62% of these girls will have this pattern persist into adulthood. Because half of adolescent girls who have oligomenorrhea or secondary amenorrhea are affected by a permanent anovulatory disorder, it would be important to attempt to distinguish these girls from those who will progress eventually toward ovulatory cycles. Interestingly, approximately half of adolescent girls who have oligoamenorrhea also have elevations in serum androgens and/or lutenizing hormone (LH); and half also have polycystic ovaries.
On the other hand, Van Hooff et al assessed hormone and ovarian morphologic parameters at the age of 15 years and then again at age 18 years. The authors showed that persisting oligomenorrhea was not predicted by increased serum LH, androgens, or the polycystic ovary. Persisting oligomenorrhea was also present in 43% of girls who had normal androgens, in 40% of girls having normal LH, and in 44% of girls who had polycystic ovaries. Similarly insulin resistance (assessed by the glucose/insulin ratio) and body mass index did not predict persisting oligomenorrhea; in this study increased body mass index was the major risk factor for the persistence of anovulation. Yet, a further analysis of the data of Van Hooff et al suggests that the association of oligomenorrhea with increased androgens or polycystic ovaries may help predict the persistence of oligomenorrhea. In this small study, the additional findings of elevated androgen (testosterone) and polycystic ovaries, more so than increases in serum LH, were highly predictive of the persistence of oligomenorrhea at age 18 years.
Thus, although it has been suggested that oligomenorrhea persisting 2 years after menarche should be used as a criterion for the diagnosis of PCOS in adolescents, we suggest from the data presented that this finding, although suggestive, should not be used in isolation as a firm criterion.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
