If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
This systematic review aimed to assess variations in the clinical presentation and treatment outcomes of patients with polycystic ovary syndrome (PCOS) belonging to different ethnicities. A search was performed for studies comparing various clinical aspects of PCOS in two or more different ethnic groups. After screening 2264 studies, 35 articles were included in the final analysis. In comparison with White women with PCOS (wPCOS), East Asian women with PCOS (eaPCOS) were less hirsute, whereas Hispanic women with PCOS (hPCOS), South Asian women with PCOS (saPCOS) and Middle Eastern women with PCOS (mePCOS) were more hirsute. saPCOS had higher androgen and lower sex hormone-binding globulin (SHBG) concentrations, mePCOS had higher DHEAS concentrations, and hPCOS and Black women with PCOS (bPCOS) had lower SHBG and DHEAS measures than wPCOS. Menstrual disturbances were more frequent in eaPCOS. Both saPCOS and eaPCOS had lower body mass index with increased central adiposity. hPCOS and bPCOS were more obese. saPCOS, mePCOS, hPCOS and bPCOS had a higher prevalence of insulin resistance than wPCOS. bPCOS had a better lipid profile but higher blood pressure and cardiovascular risk. Indigenous Australian women with PCOS were more obese and more insulin resistant with higher androgen concentrations. The clinical phenotype of PCOS therefore shows a wide variation depending on ethnicity.
Polycystic ovary syndrome (PCOS) is a common endocrine disorder that is characterized by clinical and/or biochemical androgen excess, ovulatory dysfunction and polycystic ovarian morphology (PCOM). Women with PCOS have an increased risk of metabolic and cardiovascular comorbidities, infertility, pregnancy complications, psychological disorders and cancer (
). Considering these pathophysiological factors, ethnicity might affect the pathogenesis of PCOS. Although many studies have shown that ethnicity has an impact on the prevalence and clinical manifestations of PCOS, information regarding the variability of PCOS in different ethnic groups is scant (
). In 2018, by defining clinical consensus recommendations (CCR), the International PCOS Network recommended that health professionals should consider the individual's ethnicity when assessing a patient with PCOS (
). However, high-quality evidence regarding the variability of PCOS in individuals from different ethnic backgrounds is currently lacking.
Objectives
The main objective of this review was to summarize the available data on the impact of ethnicity on the different aspects of PCOS in patients presenting to clinics. The main research questions were as follows:
1.
Is there any ethnic difference in women with PCOS in terms of clinical or biochemical hyperandrogenism, menstrual dysfunction and PCOM?
2.
Is there any ethnic difference in women with PCOS in terms of metabolic and cardiovascular comorbidities?
3.
Is there any ethnic difference in women with PCOS in terms of infertility and obstetric complications?
4.
Is there any ethnic difference in women with PCOS in terms of mood disorders and quality of life (QoL)?
5.
Is there any ethnic difference in women with PCOS in terms of treatment outcomes?
Materials and methods
This review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (
A literature search was performed in the PubMed database. Bibliographies of relevant studies were also searched to identify additional sources. Three different key word sets were used for the search:
1.
(polycystic ovary syndrome OR pcos OR stein-leventhal syndrome OR hirsutism OR oligomenorrhea OR polycystic ovaries OR hyperandrogenism OR hyperandrogenemia) AND (ethnicity)
2.
(polycystic ovary syndrome OR pcos OR stein-leventhal syndrome OR hirsutism OR oligomenorrhea OR polycystic ovaries OR hyperandrogenism OR hyperandrogenemia) AND (race)
3.
(polycystic ovary syndrome OR pcos OR stein-leventhal syndrome OR hirsutism OR oligomenorrhea OR polycystic ovaries OR hyperandrogenism OR hyperandrogenemia) AND (ancestry OR hispanic OR american OR oceanic OR brazilian OR russian OR middle eastern OR turkish OR iranian OR european OR australia OR african OR asian OR black OR white OR thai OR caucasian OR mexican OR latin OR chinese OR arab OR japanese OR indian OR pakistani OR korean OR african american OR american indian OR alaska native OR asian indian OR filipino OR vietnamese OR native hawaiian OR chamorro OR samoan OR jews OR inuits OR other pacific islander OR amish OR other asian OR some other races OR migrant OR indigenous OR aboriginal).
Inclusion and exclusion criteria
Articles that were designed to compare two or more different ethnic groups in the same study and were published between April 1990 and February 2020 were included. Only studies that assessed the clinical characteristics of patients with PCOS who were referred (i.e. patients presenting to the clinic but not those in unselected populations) were included. Studies with non-human subjects, those published before 1990 and in languages other than English, and those that were meta-analyses, case reports, case series, editorials or reviews were excluded.
Data extraction
Data were extracted on the following: the author of the publication; publication year; number, age and ethnicity of the participants; mean modified Ferriman–Gallwey (mFG) scores or prevalence of subjects with hirsutism; serum androgen concentrations; prevalence of acne; prevalence and type of menstrual disturbances; ovarian morphology; fertility status; mean body mass index (BMI), waist–hip ratio (WHR) or prevalence of obesity; surrogate markers of glucose metabolism such as fasting insulin concentrations and Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) index; prevalence of acanthosis nigricans; and lipid profile and systolic and diastolic blood pressure (SBP and DBP, respectively).
Results
A total of 2916 papers were identified. Of these, 652 were removed because they were duplicates. A further 2220 articles were excluded after the title/abstract screen due to their irrelevant content. Nine more articles were also excluded after a full-text screening due to lack of relevant data. Thirty-five articles were included for the final analyses (Figure 1).
The data available from 26 studies regarding clinical and biochemical hyperandrogenism in various ethnic populations are summarized in Table 1. In two studies, it was shown that East Asian women with PCOS (eaPCOS) had lower mFG scores than White women with PCOS (wPCOS) (
). In the other two studies comparing eaPCOS and wPCOS, eaPCOS appeared to have lower mFG scores, although this did not reach statistical significance (
). South Asian women with PCOS (saPCOS) and Middle Eastern women with PCOS (mePCOS) had consistently higher mFG scores than their Caucasian counterparts (
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
). mFG scores varied even in different Caucasian groups (Icelandic Caucasian versus Boston Caucasian, White American versus Finnish and Norwegian). European Caucasians were less hirsute than their American counterparts (Table 1).
Table 1Comparison of clinical and biochemical hyperandrogenism between various ethnic groups
Study (first author, year)
DC
Study population
Age (years)
mFG
Acne
Serum androgens
Comments
Carmina (1992)
NIH
25 Hispanic American 25 Japanese 25 Italian For each ethnic group 10 age-matched controls
30 ± 2 24 ± 1 24 ± 1
Patients: Hispanic American 12 ± 1 Japanese 3.5 ± 0.2 Italian 12.5 ± 1 Controls: All <8 Subjects with hirsutism: Hispanic American 60% Japanese ns Italian 75% Controls none
Serum T and DHEAS concentrations were similar in patients
The measurements were made in the same laboratory.
In spite of similar serum androgen concentrations, Japanese patients had a lower mFG score mFG scores of Japanese patients and controls were similar 3α-Androstanediol glucuronide, a marker of skin 5α-reductase activity, was lower in Japanese patients
Dunaif (1993)
NIH
13 Caribbean Hispanic 10 non-Hispanic White 5 Caribbean Hispanic control 8 non-Hispanic White control
25.1 ± 1 27 ± 2 28 ± 2 27 ± 1
Serum T, fT, A4, DHEAS and SHBG concentrations were similar in patients
In this study age was not reported for whole PCOS cohort; however, the mean age of White and Mexican American participants with PCOS was approximately 26 years. White healthy controls were older than Mexican American healthy controls. A4, androstenedione; BMI, body mass index; DC, diagnostic criteria; DHEAS, dehydroepiandrosterone sulfate; FAI, free androgen index; fT, free testosterone; mFG, modified Ferriman–Gallwey score; NIH, National Institutes of Health criteria; ns, not specified; PCOS, polycystic ovary syndrome; R, Rotterdam criteria; SHBG, sex hormone-binding globulin; T, total testosterone.
∼30 ∼25
Serum T, fT and DHEAS concentrations were similar in patients
The measurements were made in the same laboratory.
SHBG concentrations were similar in controls
Despite comparable T concentrations, South Asian women with PCOS had higher mFG scores and acne was more frequent in South Asian women Compared with White controls, South Asian controls also had higher mFG scores Hirsutism started to present at earlier ages in South Asian patients (∼18 versus ∼22 years)
Carmina (2003)
NIH
20 non-Hispanic White American 20 Italian
29 ± 2 26.6 ± 2
Serum T, fT and DHEAS concentrations were similar
Similar androgen concentrations
Hashimoto (2003)
NIH
102 Brazilian (predominantly Black) 31 Austrian
25.5 ± 3.9 23.8 ± 4.7
Brazilians were more hirsute, and hirsutism had a greater impact on the quality of life in Brazilian patients
Wijeyaratne (2004)
ns
74 Sri Lankan 47 British Asian 40 White 45 Sri Lankan control 11 British Asian control 22 White control
Serum T was similar SHBG was lower in Sri Lankan patients but similar in controls Despite similar T concentrations, FAI was greater in Sri Lankan patients
South Asian patients with PCOS had higher FAI measures
Ehrmann (2005)
NIH
303 White 51 Black 38 Other
28.8 ± 0.3 27.6 ± 0.8 26.7 ± 0.9
SHBG concentrations in White women with PCOS were higher but the difference was not statistically significant
Iceland Caucasian 7.1 ± 6.0 Boston Caucasian 15.4 ± 8.5 Boston African American 18.5 ± 8.9 Boston Hispanic 18.2 ± 9.4 Boston Asian 15.7 ± 11.0 Iceland controls 3.0 ± 1.4
Iceland Caucasian 62.1% Boston Caucasian 84.8% Boston African American 86.0% Boston Hispanic 87.0% Boston Asian 85.7%
When compared with Boston Caucasians, A4 concentrations were higher in Icelandic women with PCOS, and T, fT and DHEAS concentrations were lower While SHBG concentrations were lower in African Americans and Hispanics, fT concentrations were highest in these ethnic groups
Icelandic Caucasian PCOS participants had a lower mean mFG score than Boston Caucasian participants The percentage of participants with acne was lower in Icelandic Caucasians with PCOS
Glintborg (2010)
R
784 Caucasian 190 Middle Eastern
32 25
Caucasian 11 Middle Eastern 16
Serum T, SHBG and fT concentrations were higher in Caucasian women but DHEAS concentrations were lower
Middle Eastern 12.3 ± 8.6 Ashkenazi Jewish 7.4 ± 5.3 Caucasian 7.4 ± 5.7 East Asian 7.9 ± 6.4 South Asian 9.6 ± 1.5 Hispanic 12.6 ± 7.5 African American 9.4 ± 4.4 Native American 4 ± 4.2 Other 9.1 ± 6.6
Hispanic, Middle Eastern, South Asian and African American women with PCOS had higher mFG scores Hispanic and African American women had higher facial mFG scores Middle Eastern and Hispanic patients had higher truncal and extremity mFG scores
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
476 non-Hispanic White 98 non-Hispanic Black 128 Hispanic American
28.8 ± 4.2 28.7 ± 4.9 29.2 ± 4.1
Non-Hispanic White 17.0 ± 8.7 Non-Hispanic Black 15.8 ± 8.5 Hispanic American 17.6 ± 7.5 Participants with hirsutism: Non-Hispanic White 86.8% Non-Hispanic Black 82.7% Hispanic American 93.8%
Non-Hispanic White 35.2% Non-Hispanic Black 25.5% Hispanic American 64.1
Participants with hirsutism: Non-Hispanic American 49% Hispanic American 75.6%
Non-Hispanic American 63.4% Hispanic American 70.5%
Serum T, A4, DHEAS, SHBG and FAI measures were comparable
Hirsutism was more common in Hispanic women The prevalence of acne was similar
Age: values are mean ± SD, mean or median; mFG: values are mean ± SD.
a The measurements were made in the same laboratory.
b In this study age was not reported for whole PCOS cohort; however, the mean age of White and Mexican American participants with PCOS was approximately 26 years. White healthy controls were older than Mexican American healthy controls.A4, androstenedione; BMI, body mass index; DC, diagnostic criteria; DHEAS, dehydroepiandrosterone sulfate; FAI, free androgen index; fT, free testosterone; mFG, modified Ferriman–Gallwey score; NIH, National Institutes of Health criteria; ns, not specified; PCOS, polycystic ovary syndrome; R, Rotterdam criteria; SHBG, sex hormone-binding globulin; T, total testosterone.
A comparison of total mFG scores was available in all studies, and two studies also reported on regional mFG score comparisons. Wang and colleagues showed that despite the similar total mFG scores, eaPCOS had less hair on the chest (
). Moreover, Afifi and co-workers found that while hPCOS and bPCOS had higher facial mFG scores, mePCOS and hPCOS had higher truncal and extremity mFG scores (
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
Controversial results were available from seven studies that assessed acne prevalence among various ethnic groups. Acne was found to be more frequent in saPCOS than wPCOS in one study, whereas in another study the percentage of subjects with acne was higher in wPCOS compared with saPCOS (
Androgen concentrations were assessed in 23 studies. In 15 out of the 23, hormone measurements were made in the same laboratory. Twelve studies did not reveal any differences in serum androgen based on ethnicity (Table 1) (
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
). In two studies that assessed wPCOS and saPCOS, serum testosterone concentrations were comparable but sex hormone-binding globulin (SHBG) concentrations were lower and free androgen index was higher in saPCOS (
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
). Legro and colleagues found that, compared with eaPCOS, testosterone concentrations were higher in wPCOS and bPCOS but the free androgen index and SHBG measures were similar (
). Differences were reported even within various Caucasian populations. Whereas serum testosterone, free testosterone and DHEAS concentrations were lower, serum androstenedione measures were higher in Icelandic Caucasian women with PCOS when compared with Boston Caucasian women with PCOS (
The data regarding menstrual disturbances among various ethnic populations are summarized in Table 2. There were only seven studies that compared menstrual patterns in women with PCOS in different ethnic groups. Generally, no differences were noted (
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
Table 2Comparison of menstrual disturbances between various ethnic groups
Study (first author, year)
DC
Study population
Age (years) mean± SD, mean or median
Criteria
Comments
Wijeyaratne (2002)
ns
47 South Asian 40 Caucasian 11 South Asian control 22 Caucasian control
26 ± 4 30.1 ± 5 31.3 ± 2 32.9 ± 3
Menstrual disturbance: cycle length >35 days and lack of ovulation
Age at menarche and number of periods per year were similar (∼13 years and ∼5 periods/year, respectively) Menstrual irregularity tended to be seen in earlier ages in South Asian women (16.2 versus 18.4 years)
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
Menstrual disturbance frequencies were similar (∼70%) but had a greater impact on quality of life in Brazilian women
Guo (2012)
R
547 Chinese 427 Dutch
28.3 ± 3.4 29.0 ± 5.2
Oligomenorrhoea: cycles between 35 and182 days Amenorrhoea: cycles ≥182 days
Chinese women presented more often with amenorrhoea (75% versus 27%)
Wang (2013)
R
121 Caucasian 28 Asian
28.0 ± 5.4 29.6 ± 5.9
Oligomenorrhoea was more frequent in Asian women
Hilman (2014)
R
413 non-Hispanic White 106 non-Hispanic Black
25.0 ± 6.7 26.3 ± 7.3
Menstrual disturbance: 9 or fewer menses/year
Menses per year were similar (∼4 ± 3 menses/year)
Mani (2015)
ns
929 White 381 South Asian
27.1 ± 7.4 24.3 ± 6.7
Oligomenorrhoea: 9 or fewer menses/year Amenorrhoea: not specifically stated
Whereas oligomenorrhoea was more common in South Asian women, the percentage of women with amenorrhoea was higher in White women even though the menstrual disturbance rate was similar
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
In 10 studies, ovarian morphology was compared between different ethnic groups. The data regarding ovarian morphology among various ethnic populations are summarized in Table 3. Despite the selection of different criteria to determine PCOM and PCOS, in most studies there was no difference between ethnic groups in terms of ovarian morphology. Although eaPCOS had a lower ovarian volume and follicle count, the prevalence of PCOM did not differ (
Prevalence and predictors of risk for type 2 diabetes mellitus and impaired glucose tolerance in polycystic ovary syndrome: A prospective, controlled study in 254 affected women.
The Journal of clinical endocrinology and metabolism.1999; 84: 165-169
). Interestingly, the frequency of PCOM, ovarian volume and follicle number differed in various Caucasian populations. Icelandic subjects with PCOS had smaller ovaries and fewer follicles (although the frequency of PCOM was similar) and the prevalence of PCOM was higher in Norwegian and Finnish women with PCOS when compared with American White participants (
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
). In the only study that compared the prevalence of PCOM between wPCOS and mePCOS, Glintborg and colleagues found a higher frequency of PCOM in Caucasians (
Table 3Comparison of ovarian morphology between various ethnic groups
Study (first author, year)
DC
Study population
Age (years)
PCO criteria
Percentage of PCOS women with PCO
Comments
Carmina (1992)
NIH
25 Hispanic American 25 Japanese 25 Italian For each ethnic group 10 age-matched controls
30 ± 2 24 ± 1 24 ± 1
10 or more cysts 2–8 mm in diameter arranged either peripherally around a dense core of stroma or scattered throughout an increased amount of stroma
Hispanic American 80% Japanese 68% Italian 76%
Similar frequency of PCO (∼75%)
Wijeyaratne (2002)
ns
47 South Asian 40 Caucasian 11 South Asian control 22 Caucasian control
26 ± 4 30.1 ± 5 31.3 ± 2 32.9 ± 3
10 or more cysts 2–8 mm in diameter with a thickened echo dense stroma
Mean ovarian volumes were comparable (∼12 ml)
Legro (2006)
NIH
435 Caucasian 109 African American 17 Asian 72 American Indian or Alaska Native
28.2 ± 3.9 27.9 ± 4.3 30.4 ± 3.0 27.6 ± 4.1
Multiple cysts (≥10) of 2–8 mm in diameter distributed evenly around the ovarian periphery with an increased amount of stroma, or multiple small cysts 2–4 mm in diameter distributed throughout abundant stroma
Caucasian 89.4% African American 97.2% Asian 100% American Indian or Alaska Native 100% Mean right ovarian volume (ml): Caucasian 12.1 ± 6.7 African American 13.4 ± 9.2 Asian 10.9 ± 5.5 American Indian or Alaska Native 10.1 ± 4.3 Mean left ovarian volume (ml): Caucasian 11.6 ± 6.7 African American 11.3 ± 6.1 Asian 7.7 ± 5.3 American Indian or Alaska Native 8.7 ± 4.3
PCO was less frequent in Caucasian women with PCOS Asians and Native Americans had lower ovarian volumes Right and left ovary volumes were different
Welt (2006)
NIH
105 Iceland Caucasian 172 Boston Caucasian 44 Boston African American 25 Boston Hispanic 21 Boston Asian 32 Iceland controls
Presence of ≥12 follicles in each ovary measuring 2–9 mm in diameter, and/or increased ovarian volume (>10 ml)
Iceland Caucasian 92.5% Boston Caucasian 99.3% Boston African American 97.4% Boston Hispanic 95.0% Boston Asian 100% Maximum ovarian volume (ml): Iceland Caucasian 12.2 ± 5.6 Boston Caucasian 15.8 ± 7.2 Boston African American 18.2 ± 7.4 Boston Hispanic 14.4 ± 4.4 Boston Asian 13.3 ± 3.8 Maximum follicle no (count): Iceland Caucasian 11.6 ± 3.9 Boston Caucasian 14.5 ± 3.8 Boston African American 14.8 ± 4.6 Boston Hispanic 13.7 ± 3.8 Boston Asian 12.0 ± 2.9
When compared with Boston Caucasian subjects, Icelandic women with PCOS had smaller ovaries and fewer follicles but the frequency of PCO was similar Ovarian volume and follicle number were lower in Asian women
Glintborg (2010)
R
784 Caucasian 190 Middle Eastern
32 25
10 or more cysts 2–8 mm in diameter with a thickened echo-dense stroma
Caucasian 47% Middle Eastern 29%
PCO was less frequent in Middle Eastern patients
Ladson (2011)
NIH
43 Black 77 White 87 Black control 35 White control
27.9 ± 5.0 26.0 ± 6.9 Age not specified
Left and right ovarian volumes were comparable
Guo (2012)
R
547 Chinese 427 Dutch
28.3 ± 3.4 29.0 ± 5.2
Presence of ≥12 follicles in each ovary measuring 2–9 mm in diameter, and/or increased ovarian volume (>10 ml)
The prevalence of PCO did not differ
Wang (2013)
R
121 Caucasian 28 Asian
28.0 ± 5.4 29.6 ± 5.9
Presence of ≥12 follicles in each ovary measuring 2–9 mm in diameter, and/or increased ovarian volume (>10 ml)
Percentage of PCOS women with antral follicle count ≥12: Caucasian 85.1% Asian 82.1% Percentage of PCOS women with ovarian volume ≥10 ml: Caucasian 44.6% Asian 50.0%
Both groups had similar ovarian morphology
Chan (2017)
R
184 American White 100 American Black 220 Indian 233 Brazilian 94 Finnish 258 Norwegian
29 29 25 26 33 28.5
Presence of ≥12 follicles in each ovary measuring 2–9 mm in diameter, and/or increased ovarian volume (>10 ml)
American White 68.5% American Black 76% Indian 78.6% Brazilian 67.4% Finnish 100% Norwegian 90.3%
When compared with American White patients with PCOS, the prevalence of PCO was higher in Indian, Finnish and Norwegian participants
Thirty studies evaluated and compared the metabolic and cardiovascular properties of individuals with PCOS belonging to different ethnicities. There was wide variability in the metabolic and cardiovascular comorbidities between different ethnic groups (Table 4). When compared with wPCOS, hPCOS and bPCOS were more obese. saPCOS and eaPCOS had lower BMI measures (
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
). The results were contradictory in terms of obesity among various Caucasian populations. While Northern Europeans had a similar BMI to American White participants, Southern Europeans were leaner (
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
Table 4Comparison of metabolic and cardiovascular risk factors between various ethnic groups
Study (first author, year)
DC
Study population
Age (years)
Obesity
IR/prediabetes/diabetes
Acanthosis nigricans
Dyslipidaemia
Hypertension
Comments
Carmina (1992)
NIH
25 Hispanic American 25 Japanese 25 Italian For each ethnic group 10 age-matched controls
30 ± 2 24 ± 1 24 ± 1
Percentage of ideal body weight: Hispanic American 122% Japanese 111% Italian ns
FIC were lower in Japanese Similar IR
Hispanic patients were more obese FIC were lower in Japanese women Despite lower weight in Japanese women, all groups were similar in terms of IR
Dunaif (1993)
NIH
13 Caribbean Hispanic 10 non-Hispanic White 5 Caribbean Hispanic control 8 non-Hispanic White control
25.1 ± 1 27 ± 2 28 ± 2 27 ± 1
mBMI: Hispanic 35.6 ± 1.5 White 37.1 ± 1.8
In the euglyaemic clamp test, Caribbean Hispanic women had lower IS than non-Hispanic White women
In an age-, weight- and body composition-matched cohort, Caribbean Hispanic patients were more insulin resistant
Norman (1995)
NIH
11 Indian 11 White 11 Indian control 11 White control
∼25 Age was similar for all groups
In an age- and BMI-matched cohort, Indian PCOS patients had higher insulin responses than White patients on an OGTT
Ehrmann (1999)
NIH
63 Caucasian 44 African American 10 Asian 5 Hispanic
ns
Frequency of IGT or type 2 DM: Caucasian 41% African American 50% Asian 40% Hispanic 50%
The study was not specifically designed to assess ethnic differences Numbers of Asian and Hispanic participants were low African Americans had a slight preponderance of type 2 DM No difference between ethnicities in terms of glucose tolerance
Williamson (2001)
ns
112 White 16 Maori 15 Pacific Islander 19 Indian 4 Asian 6 other There was an inconsistency between number and percentage of PCOS patients
In this study age was not reported for whole PCOS cohort; however the mean age of White and Mexican American participants with PCOS was approximately 26 years. White healthy control participants were older than Mexican American healthy controls. AUC, area under the curve; BMI, body mass index (kg/m2); DBP, diastolic blood pressure; DC, diagnostic criteria; DM, diabetes mellitus; FIC, fasting insulin concentration; FPG, fasting plasma glucose; HbA1c, glycated haemoglobin; HDL, high-density lipoprotein; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; HT, hypertension; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; IR, insulin resistance; IS, insulin sensitivity; LDL, low-density lipoprotein; mBMI, mean body mass index (kg/m2); mDBP, mean diastolic blood pressure; mSBP, mean systolic blood pressure; NIH, National Institutes of Health criteria; ns, not specified; OGTT, oral glucose tolerance test; PCOS, polycystic ovary syndrome; QUICKI, quantitative insulin sensitivity check index; R, Rotterdam criteria; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WHR, waist–hip ratio.
∼30 25
mBMI was higher in insulin-resistant Mexican American women with PCOS (40 versus 36)
FIC and HOMA-IR measures were higher in Mexican American women while glucose/insulin ratios were lower IR White 43.8% IR Mexican American 73.1%
After adjustment for BMI, Mexican American women still had higher FIC and HOMA-IR values but lower glucose/insulin ratios These ethnic difference were negligible at BMI values of ≤24 Ethnicity influenced the normative values of IR screening
Wijeyaratne (2002)
ns
47 South Asian 40 Caucasian 11 South Asian control 22 Caucasian control
26 ± 4 30.1 ± 5 31.3 ± 2 32.9 ± 3
mBMI: South Asian 30.6 ± 7.5 Caucasian 32.1 ± 5.9 WHR: South Asian 1.04 ± 0.02 Caucasian 0.92 ± 0.01
FIC in South Asian patients were higher IS calculated by the QUICKI method was lower in South Asians
South Asian 55% Caucasian 7%
Serum TC, TC/HDL and TG concentrations were similar
After adjustment for age, SBP and DBP were higher in Caucasians
Despite similar BMI and WHR, South Asians had lower IS than Caucasians Almost half of the South Asian women with PCOS had acanthosis nigricans
Carmina (2003)
NIH
20 non-Hispanic White American 20 Italian
29 ± 2 26.6 ± 2
mBMI: White American 40.3 ± 1 Italian 29.7 ± 1 WHR: White American 0.85 ± 0.02 Italian 0.83 ± 0.04
FIC were higher in Americans American women had lower glucose/insulin ratios
TC and LDL were comparable Serum TG concentrations were higher in American women while HDL concentrations were lower
SBP and DBP were similar
After adjustment for BMI, American women still had a worse metabolic profile Consumption of saturated fat was higher in Americans
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
mBMI: Sri Lankan 26.3 ± 0.9 British Asian 30.6 ± 7.5 White 32.1 ± 5.9 WHR: Sri Lankan 0.97 ± 0.01 British Asian 1.04 ± 0.02 White 0.92 ± 0.01
Sri Lankan women with PCOS had higher FPG and FIC but lower IS assessed by QUICKI
TC and TG concentrations were higher in Sri Lankan patients
South Asian women were younger Sri Lankan women had lower BMI Sri Lankan women had the highest homocysteine concentrations Despite lower BMI and comparable WHR, Sri Lankan women had a worse metabolic profile Sri Lankan women with PCOS may have increased cardiovascular disease risk compared with their White counterparts
mBMI: Western European 30.2 ± 4.5 Middle Eastern 31.4 ± 5.0 African American 29.9 ± 4.8 East Indian 27.3 ± 5.5 South American 28.8 ± 4.3
Middle Eastern women had higher glucose and insulin concentrations than Western European women
Independent of BMI, Middle Eastern women had worse glucose tolerance than Western Europeans
Ehrmann (2005)
NIH
303 White 51 Black 38 Other
28.8 ± 0.3 27.6 ± 0.8 26.7 ± 0.9
mBMI: White 36.3 ± 0.5 Black 37.1 ± 1.2 Other 32.6 ± 1.1 WHR: White 0.87 ± 0.01 Black 0.91 ± 0.01 Other 0.88 ± 0.01
Black women had higher FIC, higher HbA1c and were more IR by HOMA-IR
Black women with PCOS were more IR
Kumar (2005)
NIH
186 White 27 Black 94 White control 88 Black control
27.3 ± 6.6 28.7 ± 6.2 28.7 ± 9.1 30.2 ± 9.1
mBMI: White 36.0 ± 9.3 Black 36.0 ± 8.9 WHR: White 0.84 ± 0.08 Black 0.84 ± 0.10
FIC and FPG were similar
No differences in terms of carbohydrate metabolism
Kauffman (2006)
R
111 White 50 Mexican American
27.3 26.9
mBMI: White 33.7 Mexican American 35.4
FPG concentrations were similar Mexican American patients had higher FIC and were higher IR
Mexican American patients had a worse metabolic profile
Legro (2006)
NIH
435 Caucasian 109 African American 17 Asian 72 American Indian or Alaska Native
28.2 ± 3.9 27.9 ± 4.3 30.4 ± 3.0 27.6 ± 4.1
mBMI: Caucasian 35.4 ± 8.8 African American 36.0 ± 8.4 Asian 29.1 ± 6.3 American Indian or Alaska Native 34.3 ± 8.0
Glucose/insulin ratios were lowest in Asian women, and Asian women tended to have more normal HOMA-IR values
mSBP: Caucasian 122.2 ± 13.7 African American 123.6 ± 12.6 Asian 111.4 ± 9.1 American Indian or Alaska Native 122.5 ± 15.0 mDBP: Caucasian 77.6 ± 9.9 African American 76.8 ± 10.5 Asian 73.1 ± 6.8 American Indian or Alaska Native 74.4 ± 9.2
Asian women with PCOS tended to be older and lighter After adjustment for weight, blood pressures were similar Asian women had a better metabolic profile
Lo (2006)
ns
3778 White 552 Black 1117 Asian 1324 Hispanic 432 other
BMI ≥30: White 67.5% Black 80.3% Asian 45.1% Hispanic 73.8% Other 68.9%
DM: White 9.5% Black 8.9% Asian 11.9% Hispanic 11.9% Other 13.9%
HT: White 13.9% Black 21.7% Asian 14.2% Hispanic 12.2% Other 13.9%
Asian women with PCOS had the lowest BMI Black and Hispanic women were more obese After adjustment for BMI and age, Asian and Hispanic women had an increased risk of DM After adjustment for BMI, age and DM status, Black women had an increased risk of HT
Welt (2006)
NIH
105 Iceland Caucasian 172 Boston Caucasian 44 Boston African American 25 Boston Hispanic 21 Boston Asian 32 Iceland controls
mBMI: Iceland Caucasian 31.5 ± 7.7 Boston Caucasian 30.7 ± 9.2 Boston African American 36.3 ± 7.9 Boston Hispanic 32.3 ± 10.3 Boston Asian 26.3 ± 5.9
DM: Iceland Caucasian na Boston Caucasian 2% Boston African American 11.9% Boston Hispanic 4.3% Boston Asian 0% FIC and HOMA-IR measures were higher in African American and Hispanic women
Iceland Caucasian 47.4% Boston Caucasian 62.9% Boston African American 76.7% Boston Hispanic 69.6% Boston Asian 70%
TC, LDL and TG concentrations were comparable across all ethnic groups but HDL concentrations were lower in Icelandic Caucasians
mSBP of Icelandic PCOS subjects was higher
African American women had the highest BMI and Asian women had the lowest BMI African American and Hispanic participants had a worse metabolic profile The percentage of participants with acanthosis nigricans was lower in Icelandic Caucasians
Essah (2008)
NIH
106 American (92% White) 108 Italian
29.9 ± 7.5 24.7 ± 5.2
mBMI: American 36.1 ± 8.6 Italian 28.1 ± 5.8 BMI >30: American 73.6% Italian 30.6%
While HDL concentrations were lower, TC, LDL and TG concentrations were higher in American women
mSBP of Italian PCOS patients was higher
After adjustment for BMI and age, serum TG concentrations remained higher in American participants
Glintborg (2010)
R
784 Caucasian 190 Middle Eastern
32 25
mBMI: Caucasian 27.0 Middle Eastern 25.7
Middle Eastern women had higher FIC Middle Eastern PCOS patients had higher 2 h glucose and AUC insulin during OGTT
After adjustment for BMI and age, lipid profiles were comparable
mSBP and mDBP were higher in Caucasian women even after correcting for BMI and age
When compared with Middle Eastern women, Caucasian women with PCOS had higher IS but an increased cardiovascular disease risk
Koval (2010)
NIH
94 Caucasian 32 African American
30.5 ± 6.8 30.6 ± 7.6
mBMI: Caucasian 37.0 ± 7.1 African American 41.0 ± 9.6
After adjustment for BMI, age and HOMA-IR, African American women had higher HDL, lower TG and non-HDL cholesterol
African American women with PCOS had a more favourable lipid profile
Kauffman (2011)
R
120 White 71 Mexican American
27.4 26.8
mBMI: White 34.8 Mexican American 34.3
HOMA-IR, 2 h insulin concentrations and AUC insulin during OGTT were higher in Mexican American participants with PCOS Matsuda and Strumvoll IS index values were higher in Whites
TC, TG, HDL, LDL and non-HDL cholesterol concentrations were similar
In an age- and BMI-matched cohort, Mexican American women had higher IR but lipid concentrations were similar
Ladson (2011)
NIH
43 Black 77 White 87 Black control 35 White control
27.9 ± 5.0 26.0 ± 6.9 Age not specified
mBMI: Black 39.0 ± 9.3 White 37.7 ± 6.3 WHR: Black 0.88 ± 0.08 White 0.88 ± 0.06
FIC and HOMA-IR values were higher in Black participants
Black women with PCOS had higher HDL and lower TG concentrations
mSBP and mDBP were comparable
Lipid profile was more favourable in Black women with PCOS
Guo (2012)
R
547 Chinese 427 Dutch
28.3 ± 3.4 29.0 ± 5.2
mBMI: Chinese 25.3 ± 4.3 Dutch 26.3 ± 6.9 WHR: Chinese 0.85 ± 0.06 Dutch 0.82 ± 0.08
FPG was lower but FIC were higher in Chinese HOMA-IR values were similar
TC and TG concentrations were similar Chinese participants had higher LDL and lower non-HDL cholesterol concentrations
mSBP and mDBP were higher in Dutch women
Although mBMI values of Chinese women were lower, WHR of Chinese women was higher Chinese women were more prone to central obesity
Wang (2013)
R
121 Caucasian 28 Asian
28.0 ± 5.4 29.6 ± 5.9
mBMI: Caucasian 30.4 ± 8.1 Asian 30.1 ± 7.4
FPG, 2 h glucose concentrations and FIC were similar
TC, TG, HDL and LDL concentrations were similar
Groups were similar in terms of metabolic profile
Hilman (2014)
R
413 non-Hispanic White 106 non-Hispanic Black
25.0 ± 6.7 26.3 ± 7.3
BMI >30: White 47% Black 72.3%
FIC and FPG were higher in Black women
TC, TG and HDL concentrations were lower in Black women
mSBP and mDBP were higher in Black women
Black women were more obese and had higher IR, had higher blood pressure values Despite the relatively favourable lipid profile, Black individuals with PCOS had a higher prevalence of metabolic syndrome and 10-year cardiovascular disease risk After adjustment for age and BMI, Black women had low HDL and high glucose concentrations
Mani (2015)
ns
929 White 381 South Asian
27.1 ± 7.4 24.3 ± 6.7
mBMI: White 31.5 ± 7.9 South Asian 29.3 ± 6.8
DM: White 5.6% South Asian 8.1%
White 3.1% South Asian 16.8%
mSBP and mDBP were higher in White women
South Asian women presented at earlier ages Despite lower BMI, acanthosis nigricans and DM were more common in South Asian women
62 Black 32 White 23 Hispanic 113 Black control 54 White control 37 Hispanic control
41 41 39 40 40 39
mBMI: Black 32.3 White 28.2 Hispanic 31.7 WHR: Black 0.87 White 0.83 Hispanic 0.86
FIC were lower in White women with PCOS even after adjustment for BMI Hispanic women had a higher frequency of IFG
Prevalence of hypertriglyceridaemia was higher in White women TC concentrations were lower in Black women
mSBP and the prevalence of HT were higher in Black individuals
Black and Hispanic women were more obese Black women had higher FIC Hypertension was more common in Black women Hispanic women had a higher frequency of IFG The lipid profile of Black women was more favourable Ethnicity influences cardiovascular risk factors in both individuals with and without PCOS but no synergistic effect on cardiovascular risk factors that varied by ethnicity was noted
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
184 American White 100 American Black 220 Indian 233 Brazilian 94 Finnish 258 Norwegian
29 29 25 26 33 28.5
Median BMI: American White 30.6 American Black 37.5 Indian 26.7 Brazilian 29.3 Finnish 29.4 Norwegian 31.1
Despite lower BMI, FPG concentrations were highest in Indian participants
When compared with American White women with PCOS, Black women had lower and Indian women had higher TG concentrations HDL concentrations were lower both in Indian and Black women
When compared with American White women with PCOS, Black participant had higher mSBP and mDBP mDBP was also higher in Brazilian, Finnish and Norwegian women with PCOS
Black individuals had the highest rate of obesity The prevalence of metabolic syndrome was higher in Black (52%) and Norwegian (41.1%) women compared with American White women (28.3%) In an age- and BMI-adjusted analysis, Indian and Norwegian women had elevated odds of metabolic syndrome
476 non-Hispanic White 98 non-Hispanic Black 128 Hispanic American
28.8 ± 4.2 28.7 ± 4.9 29.2 ± 4.1
mBMI: non-Hispanic White 35.1 ± 9.8 non-Hispanic Black 35.7 ± 7.9 Hispanic American 36.4 ± 7.9
FIC, FPG and HOMA-IR measures were higher in Hispanic participants
TG concentrations were highest in Hispanic and lowest in Black women LDL concentrations were comparable
Black women had higher mSBP
Metabolic syndrome was more prevalent in Hispanic women Despite the comparable BMI and waist circumference, Hispanic women had the most severe metabolic disturbance
median BMI: non-Hispanic American 26.0 Hispanic American 31.2
HOMA-IR measures were higher in Hispanic women
Non-Hispanic American 34.3% Hispanic American 65.2%
Whereas TG concentrations were higher, HDL concentrations were lower in Hispanic women
Hispanic women had an unfavourable metabolic profile Non-alcoholic steatohepatitis was more common in Hispanic women, and Hispanic ethnicity was an independent risk factor for this in multivariate analysis
Age: values are mean ± SD, mean or median.
a In this study age was not reported for whole PCOS cohort; however the mean age of White and Mexican American participants with PCOS was approximately 26 years. White healthy control participants were older than Mexican American healthy controls.AUC, area under the curve; BMI, body mass index (kg/m2); DBP, diastolic blood pressure; DC, diagnostic criteria; DM, diabetes mellitus; FIC, fasting insulin concentration; FPG, fasting plasma glucose; HbA1c, glycated haemoglobin; HDL, high-density lipoprotein; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; HT, hypertension; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; IR, insulin resistance; IS, insulin sensitivity; LDL, low-density lipoprotein; mBMI, mean body mass index (kg/m2); mDBP, mean diastolic blood pressure; mSBP, mean systolic blood pressure; NIH, National Institutes of Health criteria; ns, not specified; OGTT, oral glucose tolerance test; PCOS, polycystic ovary syndrome; QUICKI, quantitative insulin sensitivity check index; R, Rotterdam criteria; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WHR, waist–hip ratio.
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
). Conflicting results were available for eaPCOS. In general, eaPCOS were leaner but more prone to central adiposity. Three studies revealed no difference between eaPCOS and wPCOS. While Carmina and colleagues used the percentage ideal body weight for comparison, BMI values were measured in other studies (
). Nevertheless, in the largest study based on a health database review including 3778 wPCOS, 552 bPCOS, 1117 eaPCOS, 1324 hPCOS and 432 individuals with PCOS belonging to other ethnic groups, Lo and colleagues showed that, after adjustment for BMI and age, eaPCOS had an increased risk of diabetes mellitus (
). Four studies evaluating acanthosis nigricans reported that this finding was more common in ethnic populations that were more insulin resistant (hPCOS and saPCOS; Table 4) (
Seventeen studies evaluated lipids in different ethnic groups with PCOS. It was not the ethnicity but the fat mass (especially the central fat mass) that determined serum triglyceride and high-density lipoprotein (HDL) concentrations. In association with BMI or WHR, serum triglyceride concentrations were higher and serum HDL concentrations were lower in more obese (especially centrally obese) ethnic groups (
). Exceptionally, bPCOS had a more favourable lipid profile. In particular, serum triglyceride concentrations were lower in bPCOS. Moreover, bPCOS had higher HDL concentrations when compared with their White counterparts (
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
Although bPCOS had a low prevalence of dyslipidaemia, they presented with higher blood pressure. When compared with wPCOS, bPCOS had higher SBP and DBP values (
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
). Blood pressure measurements were higher among European White women, followed by American White participants, with the values for other ethnic groups (saPCOS, mePCOS and eaPCOS) being lower than in White participants (
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
Sub-infertility and pregnancy obstetric complications
No data currently exist regarding fertility outcomes in PCOS across various ethnic groups. In the only study that assessed the risk of gestational diabetes mellitus during pregnancy in PCOS including diverse ethnic groups (White, Black, Hispanic, Asian and others), Lo and co-workers reported East Asian ethnicity as a predictor for gestational diabetes mellitus with an odds ratio of 3.5 (95% confidence interval 2.3–5.5) in relation to the White population (
Only two studies compared QoL parameters across different ethnicities. Hashimoto and colleagues compared Brazilian (Predominantly black) women with PCOS and Austrian women with PCOS. The rates of infertility and menstrual disturbances were similar. Hirsutism and obesity were more common among Brazilian participants. Overall, Brazilian women with PCOS had lower QoL scores. When compared with Austrian women with PCOS, hirsutism, infertility and menstrual disturbances had a more negative impact on QoL in Brazilian women with PCOS. Austrian women with PCOS were leaner. However, despite lower BMI measures, obesity had a greater impact on QoL in Austrian women with PCOS (
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
). Although it was not a direct comparison of different ethnic groups, Schmid and co-workers illustrated a decreased QoL in Moslem immigrants in Austria. Moslem women had more concerns about infertility (
No studies were found regarding treatment outcomes of PCOS in different ethnic populations.
Discussion
For this review, the literature was systematically searched to identify ethnic differences in the clinical presentation of PCOS and it was found that the phenotype of PCOS varies widely depending on the ethnicity.
Clinical hyperandrogenism is included in the definition of PCOS in all three sets of recommended diagnostic criteria (
Positions statement: Criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: An androgen excess society guideline.
The Journal of clinical endocrinology and metabolism.2006; 91: 4237-4245
). Hirsutism is used to define clinical hyperandrogenism, although acne or androgenic alopecia can accompany hirsutism in PCOS. Along with serum androgen concentrations, individual sensitivity of the pilosebaceous unit to androgens determines the degree of hair growth (
). The mFG system is the gold standard method to diagnose and quantify hirsutism, scoring hair density in the nine body areas from 0 to 4. In the original study conducted in Caucasian women, a cut-off score of ≥8 was determined as defining hirsutism (
). However, subsequent studies performed in diverse ethnic populations showed ethnic variations in the mFG cut-off. While lower cut-off values were recommended for East Asian women, higher mFG cut-off values have been proposed even in healthy Hispanic and Middle Eastern subjects owing to the higher hair density (
). The findings of this review suggest lower mFG scores in eaPCOS and higher mFG scores in hPCOS, saPCOS and mePCOS compared with Caucasians (Figure 2). All available data regarding bPCOS came from studies conducted in the USA and in these studies, in general, bPCOS had comparable mFG scores to their American non-Hispanic White counterparts. It is also worth emphasizing that there were differences in hirsutism even within Caucasian populations (European versus American) (Table 1).
Figure 2Variations in the clinical presentation of polycystic ovary syndrome patients belonging to different ethnicities. AN, acanthosis nigricans; BP, blood pressure; DHEAS, dehydroepiandrosterone sulphate; FAI, free androgen index; fT, free testosterone; GDM, gestational diabetes mellitus; HDL, high-density lipoprotein; IR, insulin resistance; SHBG, sex hormone-binding globulin; T, testosterone; TG, triglyceride.
Total mFG scores were used for the comparison of hirsutism between different ethnic groups in all but two studies. Although total mFG scores give an estimation of the total amount of body hair, excessive hair growth might occur only in some parts of the body. Some body areas may be more sensitive to androgen action and this might be more evident in some ethnic groups. In support of this idea, in the only two studies that also compared site-specific mFG scores, researchers pointed out ethnic differences in site-specific mFG scores (Table 1). mFG scores decrease with age (
), but not all studies directly comparing different ethnicities in the current review were age matched, and age groups were heterogeneous (Table 1). This may lead to errors of interpretation. Moreover, there is no comparative study investigating how hair density changes with ageing in different ethnicities. Therefore, when using the mFG score to determine clinical androgen excess, the individual's ethnicity and age should be taken into consideration.
Acne can be observed in patients with hyperandrogenism. Although the available data are contradictory, some differences may exist in the biological characteristics of skin among various ethnic groups (
). Hence, the prevalence of acne may differ in various ethnic groups. In this review, the results for acne were discordant and not sufficient to conclude whether ethnicity has any impact on the prevalence of acne among different ethnicities in PCOS. It should also be noted that there are no universally accepted visual tools to evaluate acne.
Evidence for ethnic differences in hyperandrogenemia was scarce and unclear. There are no studies specifically evaluating and rogens and their cut-off levels in different ethnic groups with different ages. Most studies had a limited sample size and did not assess the same androgens. Despite these limitations, androgen concentrations seemed to be similar in diverse ethnic populations, with a few exceptions for ethnic differences in SHBG and adrenal androgens (Table 1).
Ovulatory dysfunction is a diagnostic criterion for PCOS, and the clinical detection of ovulatory dysfunction is based on menstrual irregularity. Only seven studies were identified that compared the menstrual patterns of women with PCOS belonging to different ethnicities (Table 2). Menstrual disturbances appeared to be more common in eaPCOS. eaPCOS has a mild androgenic phenotype and menstrual disturbances may be a major complaint in clinical presentation. Considering that all studies enrolled patients presenting to clinics, this finding might be due to selection bias. Whether there are frank differences in menstrual irregularity between eaPCOS and other ethnic groups warrants further multiethnic comparative studies in unselected populations.
The 2003 Rotterdam consensus proposed PCOM as a diagnostic criterion for PCOS, defining it as the presence of 12 or more follicles in each ovary measuring 2–9 mm in diameter, and/or increased ovarian volume (>10 ml) on ultrasound (
Positions statement: Criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: An androgen excess society guideline.
The Journal of clinical endocrinology and metabolism.2006; 91: 4237-4245
). Accordingly, using NIH criteria to diagnose PCOS may exclude some individuals with PCOM and may yield an underrepresentation of this group. In the current review, 15 studies had used NIH criteria. In general, the frequency of PCOM did not show a difference for different ethnic groups in the available studies. However, there were subtle differences in ovarian morphology even though the prevalence of PCOM was similar between ethnic groups. For instance, eaPCOS had lower follicle counts and ovarian volume in two studies (see Table 3). Further studies are needed to determine whether PCOM differs in different ethnic groups.
Obesity is prevalent in women with PCOS presenting to clinics. In this review, obesity was more frequent in hPCOS and bPCOS compared with Caucasian patients. On the other hand, eaPCOS and saPCOS had lower BMI values with increased central fat and a comparable or higher metabolic risk compared with Caucasian PCOS women. The most common method for assessing excessive fat is BMI calculation. A BMI value higher than 30 kg/m2 was proposed to diagnose obesity in Caucasian populations. However, BMI is a crude indicator and does not reflect the increased metabolic and cardiovascular disease risk in certain ethnic populations (
). Accordingly, different BMI thresholds were determined to define obesity in various ethnic groups. BMI cut-offs have been lowered to 25 and 23 kg/m2 to reflect the risk in South Asian and East Asian populations, respectively (
Consensus statement for diagnosis of obesity, abdominal obesity and the metabolic syndrome for asian indians and recommendations for physical activity, medical and surgical management.
The Journal of the Association of Physicians of India.2009; 57: 163-170
). Considering the propensity of East Asian and South Asian individuals to develop central adiposity, the lowered thresholds of BMI may still not be appropriate for these ethnic populations. Therefore, for evaluating the increased adiposity in women of various ethnic origins with PCOS, the use of population-specific anthropometric measures such as WHR may provide a better assessment of metabolic risk.
Many women with PCOS are insulin resistant and the incidence of prediabetes and diabetes mellitus is increased in women with PCOS independent of age and BMI (
). Thus, one might expect variable degrees of insulin resistance in PCOS patients with different ethnicities. In this systematic review, it was found that hPCOS, bPCOS, mePCOS and saPCOS were more insulin resistant and predisposed to glucose intolerance compared with wPCOS. Moreover, the prevalence of acanthosis nigricans was higher in the ethnic groups with a higher level of insulin resistance. However, the methods to assess glucose homeostasis showed variations (Table 4). Other cardiovascular risk factors, including dyslipidaemia and hypertension, also varied among different ethnic populations. Dyslipidaemia was more prevalent in more obese ethnic groups. bPCOS had better lipid measures but higher blood pressure. High blood pressure values were also noted in European and American Caucasians when compared with saPCOS, mePCOS and eaPCOS (Table 4). Hence, the components of the metabolic syndrome were widely variable in women with PCOS of different ethnicities. The current data regarding the differences in clustering components of metabolic syndrome in different ethnic populations emphasize that ethnic variables should also be taken into consideration when evaluating patients’ metabolic status.
PCOS is the main cause of anovulatory subfertility. Evidence suggests that both natural and assisted fecundity rates differ between various ethnic groups and that ethnicity may affect the success of fertility treatments and their outcomes (
). However, this literature search did not reveal any study comparing fertility outcomes across various ethnic groups that included only women with PCOS. Women with PCOS are at increased risk of pregnancy complications and adverse obstetric outcomes (
). The diverse reproductive and metabolic presentation of PCOS in diverse ethnic populations may persist during pregnancy and influence the obstetric and neonatal outcomes. Therefore, it can be expected that pregnancy-associated complications may vary in different ethnic groups. In the current review, only one study was found showing that eaPCOS had a higher risk of gestational diabetes mellitus during pregnancy compared with wPCOS (
Androgen excess- polycystic ovary syndrome society: Position statement on depression, anxiety, quality of life, and eating disorders in polycystic ovary syndrome.
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
). Two studies were found that assessed the QoL scores of PCOS patients with different ethnicities, and in a separate study individuals from different religions were compared (
). There was no study that compared the outcomes of lifestyle intervention and medical treatment among different ethnic populations with PCOS. Ethnic variations can be important for treatment outcomes. For instance, interindividual variations in metformin response due to genetic heterogeneity have been reported in patients with diabetes (
Genetic polymorphisms of organic cation transporter 1 (oct1) and responses to metformin therapy in individuals with type 2 diabetes: A systematic review.
). Moreover, the ethnic or cultural background of an individual may affect drug preferences. Rocca and colleagues reported significant ethnic disparities in use of the contraceptive pill (
Rocca, C.H., Harper, C.C., 2012. Do racial and ethnic differences in contraceptive attitudes and knowledge explain disparities in method use? Perspectives on sexual and reproductive health. 44, 150-8
). Accordingly, the potential influence of ethnicity on response to lifestyle interventions or medical treatment in PCOS is an area of interest for future research.
Overall, ethnicity plays an important role in the phenotypic presentation of PCOS and its individual components. For example, independent of PCOS, mFG cut-off scores for defining hirsutism and the severity of hirsutism vary by ethnicity. Scores of ≥4 and ≥6 define hirsutism in White and Black women and Han Chinese women, respectively. Accordingly, in 2018, the International PCOS Network suggested that ethnic variation should be considered in the management of PCOS (
). However, the studies included in our review did not compare phenotypic variations in PCOS using ethnicity-specific recommendations. Future studies on the impact of ethnicity are needed to inform the guidelines (Table 5).
Table 5Areas for future research regarding impact of ethnicity on PCOS
•
Development of age- and ethnicity-specific visual tools and cut-off values for assessment and definition of clinical hyperandrogenism, including hirsutism and acne
•
Comparative analysis of biochemical hyperandrogenism in multiethnic cohorts with high-quality assays
•
Assessment of phenotypic variation including prevalence of prediabetes/diabetes and risk of cardiovascular disease in multiethnic longitudinal studies of unselected populations
•
Evaluation of the impact of ethnicity on fertility, obstetric and neonatal outcomes in comparative studies of women with PCOS
•
Determination of the role of ethnicity on perception of PCOS and emotional well-being and quality of life
•
Assessment of the role of ethnicity on long-term medical management of PCOS
There are some limitations to the current systematic review. First, the review included studies that were performed in clinics. The clinical presentation and ethnic characteristics of PCOS may vary in unselected populations (
). Therefore these findings on clinical referral populations might not be able to be extrapolated to unselected populations. Second, in all studies, ethnicity was assessed by self-reported data. However, self-reported ethnicity might be subjective and biased. Using genetic ancestry instead of self-reported ethnicity may provide a better ethnic determination (
). Finally, although in some studies the participants lived in the same environment, in others study participants were recruited from different countries. Considering the effect of environmental factors on PCOS, the disparities between different ethnic groups might not be attributable to ethnic factors per se.
Conclusion
Based on the limited data available, the clinical presentation of PCOS shows a wide variation among different ethnic populations. Larger multiethnic comparative studies specifically assessing the role of ethnicity in the diagnosis and management of PCOS are needed for developing ethnicity-specific guidelines.
REFERENCES
Afifi L.
Saeed L.
Pasch L.A.
Huddleston H.G.
Cedars M.I.
Zane L.T.
Shinkai K.
Association of ethnicity, fitzpatrick skin type, and hirsutism: A retrospective cross-sectional study of women with polycystic ovarian syndrome.
International journal of women's dermatology.2017; 3: 37-43
Positions statement: Criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: An androgen excess society guideline.
The Journal of clinical endocrinology and metabolism.2006; 91: 4237-4245
Racial and ethnic differences in the prevalence of metabolic syndrome and its components of metabolic syndrome in women with polycystic ovary syndrome: A regional cross-sectional study.
American journal of obstetrics and gynecology.2017; 217 (189.e1-189.e8)
Androgen excess- polycystic ovary syndrome society: Position statement on depression, anxiety, quality of life, and eating disorders in polycystic ovary syndrome.
The impact of the weight status on subjective symptomatology of the polycystic ovary syndrome: A cross-cultural comparison between brazilian and austrian women.
Anthropologischer Anzeiger; Bericht uber die biologisch-anthropologische Literatur.2003; 61: 297-310
Black women with polycystic ovary syndrome (pcos) have increased risk for metabolic syndrome and cardiovascular disease compared with white women with pcos [corrected].
Prevalence and predictors of risk for type 2 diabetes mellitus and impaired glucose tolerance in polycystic ovary syndrome: A prospective, controlled study in 254 affected women.
The Journal of clinical endocrinology and metabolism.1999; 84: 165-169
Consensus statement for diagnosis of obesity, abdominal obesity and the metabolic syndrome for asian indians and recommendations for physical activity, medical and surgical management.
The Journal of the Association of Physicians of India.2009; 57: 163-170
Genetic polymorphisms of organic cation transporter 1 (oct1) and responses to metformin therapy in individuals with type 2 diabetes: A systematic review.
Rocca, C.H., Harper, C.C., 2012. Do racial and ethnic differences in contraceptive attitudes and knowledge explain disparities in method use? Perspectives on sexual and reproductive health. 44, 150-8
Suleyman Nahit Sendur completed his training in internal medicine and endocrinology at Hacettepe University, Turkey. He was appointed as an Assistant Professor at the same institution in 2020. His research and clinical interests are in pituitary, adrenal and gonadal disorders.
Key message
Ethnicity affects the clinical presentation of PCOS and needs to be taken into consideration in evaluating and managing the syndrome.
Article info
Publication history
Published online: December 16, 2020
Accepted:
December 10,
2020
Received in revised form:
December 6,
2020
Received:
August 20,
2020
Declaration: The authors report no financial or commercial conflicts of interest.
30642-8&id=gr1.jpg){kind=link}
30642-8&id=gr2.jpg){kind=link}