Ethnic differences on incidence of PCOS based on the Rotterdam, Japanese and Chinese criteria

The prevalence of PCOS in largely Western populations was reported to be as high as 15% based on the Rotterdam criteria of 2003 . By contrast, a study of >15,000 Chinese women noted that the prevalence of PCOS was only 5.6% . The prevalence of PCOS among the Asian populations has been reported to vary from 5.6% in the Southern Chinese population , 5.7% in Thai women , 6.3% in Sri Lankan women and 14.3% in Iranian women . Even in studies on Caucasian subjects, the prevalence of PCOS ranges from 6% to 15% .

One reason for the wide range of reported prevalence of PCOS is likely due to differences in the criteria used in the diagnosis of the condition, an enduring controversy further explored in Chapter 1. Other reasons may be technical, such as subjective variation in clinical assessment of hirsutism, lack of reference standards for androgen assays and inter-observer differences in the measurement of ovarian morphology. Nevertheless, the divergent prevalence of PCOS in different ethnic groups suggests the possibility that there may be true ethnic differences ( Table 1 ). In particular, East Asian subjects (Korean, Chinese and Thai) appear to have a lower prevalence of PCOS (about 5%) using currently accepted diagnostic criteria when compared with Caucasian women (11–20%). Responses to treatment modalities may differ and therefore ethnic differences may have important therapeutic consequences.

Hirsutism: distinct ethnic differences

Hirsutism, or male-pattern hair growth, is sufficient to label a woman as hyperandrogenic, meeting a key diagnostic criterion in both Rotterdam and AE-PCOS definitions. Hirsutism is clinically identified by the presence of excessive coarse terminal hair in androgen-responsive areas of the female body. The modified Ferriman–Gallwey (mFG) is commonly used to score nine body areas (upper lip, chin, chest, upper and lower back, upper and lower abdomen, arm, forearm, thigh and lower leg), according to a scale ranging from 0 (no hair) to 4 (similar to that of a well-virilised adult male) . It has to be kept in mind that the severity of hirsutism does not correlate well with the magnitude of androgen excess and that hair follicle response to circulating androgens varies considerably between individuals, explaining why some women with clearly elevated androgen levels do not exhibit hirsutism . Despite publication of visual scoring aids, inter-observer differences may arise due to the subjective nature of visual assessments . An added complexity is that hirsutism itself may vary with age, with hirsutism scores decreasing with increasing age . Previous treatments for unwanted hair growth may have confounded the results of some studies .

The original study by Ferriman and Gallwey defined hirsutism as a score >8 . In most Caucasian studies using this definition, the average hirsutism rate was 74.6% in a review of >6000 subjects with a range from 17% to 100% . There is evidence that this criterion may not be appropriate for all ethnicities. East Asian women (of Chinese, Korean, Thai and Japan ethnicities) have a lower prevalence of hirsutism ( Table 2 ). One of the earliest comparative studies indicated that Japanese PCOS patients have lower mean hirsutism scores than Italian and American women in USA .

Using systematic cluster random sampling and cluster analysis of about 3000 subjects in a Southern Chinese population, an mFG score of ≥5 was found to be indicative of hair growth above the norm . Hirsutism in these women was correlated with higher incidences of acne, menstrual irregularity, polycystic ovaries and acanthosis nigricans. Interestingly, this cut-off value decreased with increasing age, being 14.4%, 10.7%, 7.9%, 3.6% and 1.5%, respectively, in women aged 20–25, 26–30, 31–35, 36–40 and 41–45 years, respectively. A large community-based study involving 10,120 Chinese women concluded that mFG score >4 could lead to a diagnosis of hirsutism . An mFG score of ≥5 indicates hair growth above the norm among Chinese women in the general Southern Chinese population, a cut-off value that decreases with increasing age. Based on menstrual irregularities and the presence of polycystic ovaries, an mFG score ≥5 has been proposed to define hirsutism in Chinese women . Women diagnosed to have PCOS living in China were reported to be significantly less hirsute than Caucasian PCOS subjects living in the Netherlands, with mFG scores of 3.6 ± 4.9 versus 5.2 ± 5.4, respectively . Differences also exist within East Asian populations. In Thai women, mFG score >3 has been proposed based on the 97.5th percentile of a community-based population . In Japanese women, it was suggested to use mFG score of >6 to define hirsutism . However, some studies involving small number of Asians have found no differences in hirsutism rates comparing Caucasian with Asians .

In South Asians, PCOS subjects were reported to have a higher mean mFG score, and in one study the mean score was as high as 18 . Another study from an endocrinology clinic in the United Kingdom demonstrated that in subjects diagnosed with PCOS, women of South Asian descent exhibited a rate of hirsutism comparable to White women (88% vs. 77%, respectively) . A study in New Zealand indicates about two-thirds of women of South Indian descent diagnosed with PCOS have hirsutism, similar to the prevalence in women of European, Maori and Pacific Island descent . Mediterranean and Middle Eastern women also tend to have higher prevalence of hirsutism, and this was observed in a study comparing Middle Eastern and Caucasian women with PCOS .

Even within Caucasian populations, ethnic differences have been reported. The Ferriman–Gallwey score was lower (7.1 ± 6.0 vs. 15.4 ± 8.5, p < 0.001) in Icelandic women with PCOS than in Boston women with PCOS . However, no significant differences in mFG scores were noted between White and Black women (195 Black and 174 White) in a study conducted in Southeastern USA using an mFG score of ≥6 as indicative of hirsutism .

Studies in the literature suggest that the appropriate mFG score to define hirsutism in Asian women, excluding those from the Indian subcontinent, is <8. Recognising these differences, investigators have used mFG cut-off threshold for hirsutism from 3 to 6 in East Asian populations ( Table 2 ). By using these race-specific cut-offs, the mean prevalence of hirsutism in Caucasian populations was found to be 2.7-fold higher in Caucasian than in East Asian PCOS subjects (76.3% vs. 28.4%, respectively; Table 2 ). For these reasons, it is clear that ethnic differences in hirsutism among PCOS do exist, and such differences may contribute to varying rates of PCOS in different populations.

Hirsutism: distinct ethnic differences

Hirsutism, or male-pattern hair growth, is sufficient to label a woman as hyperandrogenic, meeting a key diagnostic criterion in both Rotterdam and AE-PCOS definitions. Hirsutism is clinically identified by the presence of excessive coarse terminal hair in androgen-responsive areas of the female body. The modified Ferriman–Gallwey (mFG) is commonly used to score nine body areas (upper lip, chin, chest, upper and lower back, upper and lower abdomen, arm, forearm, thigh and lower leg), according to a scale ranging from 0 (no hair) to 4 (similar to that of a well-virilised adult male) . It has to be kept in mind that the severity of hirsutism does not correlate well with the magnitude of androgen excess and that hair follicle response to circulating androgens varies considerably between individuals, explaining why some women with clearly elevated androgen levels do not exhibit hirsutism . Despite publication of visual scoring aids, inter-observer differences may arise due to the subjective nature of visual assessments . An added complexity is that hirsutism itself may vary with age, with hirsutism scores decreasing with increasing age . Previous treatments for unwanted hair growth may have confounded the results of some studies .

The original study by Ferriman and Gallwey defined hirsutism as a score >8 . In most Caucasian studies using this definition, the average hirsutism rate was 74.6% in a review of >6000 subjects with a range from 17% to 100% . There is evidence that this criterion may not be appropriate for all ethnicities. East Asian women (of Chinese, Korean, Thai and Japan ethnicities) have a lower prevalence of hirsutism ( Table 2 ). One of the earliest comparative studies indicated that Japanese PCOS patients have lower mean hirsutism scores than Italian and American women in USA .

Using systematic cluster random sampling and cluster analysis of about 3000 subjects in a Southern Chinese population, an mFG score of ≥5 was found to be indicative of hair growth above the norm . Hirsutism in these women was correlated with higher incidences of acne, menstrual irregularity, polycystic ovaries and acanthosis nigricans. Interestingly, this cut-off value decreased with increasing age, being 14.4%, 10.7%, 7.9%, 3.6% and 1.5%, respectively, in women aged 20–25, 26–30, 31–35, 36–40 and 41–45 years, respectively. A large community-based study involving 10,120 Chinese women concluded that mFG score >4 could lead to a diagnosis of hirsutism . An mFG score of ≥5 indicates hair growth above the norm among Chinese women in the general Southern Chinese population, a cut-off value that decreases with increasing age. Based on menstrual irregularities and the presence of polycystic ovaries, an mFG score ≥5 has been proposed to define hirsutism in Chinese women . Women diagnosed to have PCOS living in China were reported to be significantly less hirsute than Caucasian PCOS subjects living in the Netherlands, with mFG scores of 3.6 ± 4.9 versus 5.2 ± 5.4, respectively . Differences also exist within East Asian populations. In Thai women, mFG score >3 has been proposed based on the 97.5th percentile of a community-based population . In Japanese women, it was suggested to use mFG score of >6 to define hirsutism . However, some studies involving small number of Asians have found no differences in hirsutism rates comparing Caucasian with Asians .

In South Asians, PCOS subjects were reported to have a higher mean mFG score, and in one study the mean score was as high as 18 . Another study from an endocrinology clinic in the United Kingdom demonstrated that in subjects diagnosed with PCOS, women of South Asian descent exhibited a rate of hirsutism comparable to White women (88% vs. 77%, respectively) . A study in New Zealand indicates about two-thirds of women of South Indian descent diagnosed with PCOS have hirsutism, similar to the prevalence in women of European, Maori and Pacific Island descent . Mediterranean and Middle Eastern women also tend to have higher prevalence of hirsutism, and this was observed in a study comparing Middle Eastern and Caucasian women with PCOS .

Even within Caucasian populations, ethnic differences have been reported. The Ferriman–Gallwey score was lower (7.1 ± 6.0 vs. 15.4 ± 8.5, p < 0.001) in Icelandic women with PCOS than in Boston women with PCOS . However, no significant differences in mFG scores were noted between White and Black women (195 Black and 174 White) in a study conducted in Southeastern USA using an mFG score of ≥6 as indicative of hirsutism .

Studies in the literature suggest that the appropriate mFG score to define hirsutism in Asian women, excluding those from the Indian subcontinent, is <8. Recognising these differences, investigators have used mFG cut-off threshold for hirsutism from 3 to 6 in East Asian populations ( Table 2 ). By using these race-specific cut-offs, the mean prevalence of hirsutism in Caucasian populations was found to be 2.7-fold higher in Caucasian than in East Asian PCOS subjects (76.3% vs. 28.4%, respectively; Table 2 ). For these reasons, it is clear that ethnic differences in hirsutism among PCOS do exist, and such differences may contribute to varying rates of PCOS in different populations.