Evaluation of androgen-dependent skin findings of polycystic ovary syndrome (PCOS)

Abstract

Aim

The purpose of the study was to investigate the biochemical and metabolic abnormalities related to the cutaneous characteristics of PCOS.

Material–Methods

Patients diagnosed with PCOS were included in the study. Demographic data and accompanying androgen-dependent skin findings (acne, seborrhea, androgenic alopecia, acanthosis nigricans, skin tag, and hirsutism) were recorded. The free testosterone, total testosterone, dehydroepiandrosterone sulfate, androstenedione,17-Hidroksi progesterone, sex hormone binding globulin, prolactin, fasting glucose, fasting insulin, HbA1C, HDL, and triglycerides, follicle-stimulating hormone, luteinized hormone, free androgen index, and HOMA-IR levels of the patients were measured. The hormonal values of the patients with PCOS with and without skin findings were compared.

Results

The HOMA-IR values of the acanthosis nigricans (+) PCOS group were significantly higher than the acanthosis nigricans (–) PCOS group (p < .001). The DHEA-SO4, FAI, and FI values of patients with hirsutism (HR) (+) PCOS were found to be statistically higher than patients with HR (–) PCOS (p = .006, p = .015, p = .004).

Conclusion

PCOS is among the most common endocrine disorders of women of reproductive age and was associated with some hormonal, metabolic, and skin findings. Certain androgenic and metabolic variables developing in PCOS might correlate with cutaneous symptoms.

摘要

目的

探讨与多囊卵巢综合征皮肤特征相关的生化代谢异常。

材料-方法

研究对象为多囊卵巢综合征患者。记录人口统计学数据和雄激素依赖性皮肤表现(痤疮、脂溢症、雄激素性脱发、黑棘皮病、皮肤标签和多毛症)。测定患者的游离睾酮、总睾酮、脱氢表雄酮硫酸盐、雄烯二酮、17-羟孕酮、性激素结合球蛋白、催乳素、空腹血糖、空腹胰岛素、糖化血红蛋白、高密度脂蛋白、甘油三酯、卵泡刺激素、黄体生成素、游离雄激素指数和HOMA-IR水平。比较有皮肤表现和无皮肤表现的PCOS患者的激素水平。

结果

黑棘皮病(+)PCOS组的HOMA-IR值显著高于黑棘皮病(-)PCOS组(P<0.001)。多毛症(+)PCOS患者的DHEA-SO4、FAI和FI值显著高于多毛症(-)PCOS患者(p=0.006, p=0.015, p=0.004)。

结论

多囊卵巢综合征是育龄女性最常见的内分泌紊乱之一, 与某些激素、代谢和皮肤表现有关。多囊卵巢综合征的某些雄激素和代谢指标可能与皮肤症状相关。

Keywords:

• Acne
• hirsutism
• seborrhea
• polycystic ovary syndrome

关键词:

• 痤疮
• 多毛症
• 脂溢症
• 多囊卵巢综合征

Introduction

Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in women characterized by hyperandrogenism, ovulatory dysfunction, and polycystic ovary morphological characteristics, and affects approximately 2%–7% of women of reproductive age [1]. After other androgen excess disorders were excluded, PCOS diagnosis is made when two of the criteria (1-oligo/anovulation, 2-hyperandrogenism (clinical or biochemical findings), and 3-polycystic ovaries (determined by ultrasound) are present [2]. Also, National Institutes of Health (NIH) consensus panel recommended the use of the following phenotype classification: phenotype A: hyperandrogenism (clinical or biochemical presence) + oligo/anovulation + polycystic ovarian morphology; phenotype B: hyperandrogenism + oligo/anovulation; phenotype C: hyperandrogenism + polycystic ovarian morphology; and phenotype D: oligo/anovulation + polycystic ovarian morphology [3].

Hyperandrogenemia is a key characteristic of PCOS and causes or contributes to the clinical phenotype of such patients as the main source of hyperandrogenemia. Some of the hyperandrogenic findings in PCOS also give findings on the skin some of which are hirsutism (HR), acne, seborrhea, androgenic alopecia (AGA), and acanthosis nigricans (AN), skin tag (ST). Although studies in the literature report that hyperandrogenemia is correlated with HR, acne, and other skin findings, studies are reporting the opposite [4, 5].

Androgens have a very important place in the development of skin findings in PCOS. Negative and positive relationships were shown between skin findings and androgens in the literature.

In this prospective study, the purpose was to investigate the relationships between skin characteristics and biochemical changes in women with PCOS.

Material and methods

The study was carried out as cross-sectional research between November 2020 and February 2021 at Bagcilar Research and Training Hospital of the University of Health Sciences. The study was conducted with females older than the ages of 18 and diagnosed with PCOS (phenotype A). The subjects who were under the age of 18 and did not give consent, menopausal and pregnant women, those having known diseases that lead to hormonal changings, who used drugs that interfere with endocrine or lipid metabolism, and who used drugs that can cause hyperprolactinemia, within the last three months were excluded from the study.

The study was approved by the Local Ethics Committee of the University of Health Sciences (approval number: E-48670771-514.10). Written informed consent was obtained from each subject at the gynecology outpatient clinics before starting the work. It was performed in accordance with the Declaration of Helsinki 2013. After a detailed medical interview, gynecological and ultra-sonographic examinations of the subjects were performed by an experienced gynecologist. Accompanying androgen-dependent dermatological conditions including acne, seborrhea, AGA, AN, ST, and HR were diagnosed clinically by a dermatologist who was experienced with the diseases. The severity of HR was visually determined by the modified Ferriman Gallwey (mFGS) system [2]. A total of eight points were considered HR in this scoring system [2]. Diagnosis of PCOS was made based on the Revised Criteria of Rotterdam (2003) [6].

The patients who were diagnosed with PCOS were divided into two groups as those with and without skin findings (acne, seborrhea, AGA, AN, ST, and HR). The free testosterone (FT), total testosterone (TT), dehydroepiandrosterone sulfate (DHEASO₄), androstenedione, 17-hydroxy progesterone (17-OH-PG), SHBG, prolactin, fasting glucose (FG), fasting insulin (FI), glycated hemoglobin (HbA1C), HDL, and triglycerides (TG) values were compared in both groups. The free androgen index (FAI) was calculated as testosterone (nmol/L)/SHBG (nmol/L) × 100 [1]. Homeostatic model assessment of insulin resistance (HOMA-IR) = [fasting insulin (lU/mL) × fasting plasma glucose (mmol/L)]/22.5 [1].

Fasting blood samples for all tests including FT and TT, DHEASO₄, androstenedione, 17-OH-PG, SHBG, prolactin, FG, FI, HbA1C, HDL, and TG were taken during the follicular phase (cycle days 2–8).

Statistical analysis

Continuous variables were given as mean ± standard deviation and/or median (min–max), and categorical data were given in numbers and percentages. The normality analyzes of the continuous variables were made by using the Kolmogorov–Smirnov Goodness of Fit Test. The T Test was used for independent groups in the analyzes between the two groups that were suitable for normal distribution, and the Mann–Whitney U Test was used for the analysis of the variables that were not suitable for normal distribution. The categorical data were compared with chi-square test and Fisher’s exact test. Analyzes were performed with IBM SPSS version 26.0 (IBM Corporation, Armonk, NY, USA) and statistical significance level was accepted as p < .05.

Results

Between November 2020 and February 2021, a total of 4150 patients applied to the gynecology clinic of our hospital. Two hundred thirty-two of these patients were compatible with the diagnosis of PCOS. Twenty-four patients were excluded from the study because they did not meet the inclusion criteria. As a result, a total of 208 females with PCOS completed the work. The mean age was 24.15 ± 7.10 years. The mean BMI was 25.84 ± 6.00 kg/m² with about half of the patients being normal or underweight (105 patients, 50.48%). The most common cutaneous manifestation of PCOS among the patients in this study were HR (93.26%) followed by AGA (91.80%), then seborrhea (89.40%), acne (85.60%), AN (56.30%), and skin tag (33.20%) (Table 1).

Table 1. Descriptive statistics of demographics and clinical findings of the study population.

Clinical Parameters	Mean ± SD		N (208)	%
Age(years)^a	24.15 ± 7.1	18 ≤ Age ≤ 35	186	89.40
		≥36	22	10.60
Waist circumference (cm)^a	92 ± 10	<88	123	59.1
		≥88	85	40.9
Body mass index (kg/m^2)^a	25.84 ± 6.0	30 < BMI	165	79.30
		≥30 BMI	43	20.70
Marital status^b		Married	154	74.0
		Single	54	26.0
Systolic blood pressure(mm/Hg)^b		<130	95	45.70
		≥130	113	54.30
Diastolic blood pressure (mm/Hg)^b		<85	185	88.90
		≥85	23	11.10
Acne^b		Presence	178	85.60
Seborrhea^b		Presence	186	89.40
Hirsutism (Total)^b		Presence	194	93.26
Hirsutism (MFG scores)^b		MFG1 = 8–15	17	8.20
		MFG2 = 16–25	94	45.20
		MFG 3= ≥26	83	39.90
Androgenetic alopecia (AGA)^b		Presence	191	91.80
Acanthosis Nigricans (AN)^b		Presence	117	56.30
Skin tag (ST)^b		Presence	69	33.20

mFGS: modified Ferriman Gallwey, SD: standard deviation.

^a Values are mean ± standard deviation.

^b alculated as n and %.

Hormone values were compared in patient groups that were positive and negative for each skin findings. The DHEA-SO4 values, TT values, 17OHPG, LH values of the Acne (–) PCOS group were significantly higher than those of the Acne (+) PCOS group (p = .044, p = .002, p = .020, p = .019 respectively). The 17OHPG and LH values of the patients with seborrhea (–) PCOS were statistically and significantly higher than the seborrhea (+) PCOS group (p = .043, p = .008, respectively). The FT values, FI values, HbA1c values, and HOMA-IR values of the AGA (+) PCOS patients were significantly higher than those of the patients with AGA (–) PCOS (p: .025, p: .001, p: .017, p: .001 respectively). The HDL values were significantly lower in the AGA (+) PCOS group than in the AGA (–) PCOS group (p: .022). The DHEA-SO₄ values, FAI values, and FI values of the AN (+) PCOS group were significantly higher than those of AN (–) group (p: .006, p: 0.015, p: .004 respectively). The FT values were significantly higher in AN (–) group than in AN (+) group (p: 0.001). The FT values, FI values, HbA1c values, and HOMA-IR values were higher and HDL values were lower in ST (+) group than in ST (–) group (p: .025, p: .001, p: .017, p: .001, p: .022 respectively). DHEA-SO₄ values, FAI values, and FI values were hıgher and FT values were lower in hirsutism (+) PCOS group than in hirsutism (–) PCOS group (p: .006, p: .015, p: .004, p: .001 respectively). All the comparisons of hormone values between groups are given in Table 2. FI, HOMA-IR values of patients with AN (+) PCOS were found to be statistically higher than patients with AN (–) PCOS (p < .001, p < .001). The FT and FI, HbA1c, HDL, HOMA-IR values of the ST (+) PCOS group were statistically higher than those of the patients with ST (–) PCOS (p = .025, p < .001, p = .017, p = .022, p < .001, respectively). The DHEA-SO₄, FAI, FI values of patients with HR (+) PCOS were statistically higher than those of patients with HR (–) PCOS (p = .006, p = .015, p = .004).

Table 2. Comparison of hormonal values of patients with PCOS with and without skin findings.

	Acne			Seborrhea			Androgenetic alopecia (AGA)			Acanthosis Nigricans (AN)			Skin tag (ST)			Hirsutism
	No (n = 29)	Yes (n = 178)	p	No (n = 22)	Yes (n = 186)	p	No (n = 139)	Yes (n = 69)	p	No (n = 17)	Yes (n = 191)	p	No (n = 139)	Yes (n = 69)	p	No (n = 17)	Yes (n = 191)	p
DHEA-SO4 (nmol/L)^a	384.63 ± 146.48	331.86 ± 127.46	.044^b	353.57 ± 148.72	337.49 ± 128.96	.587^b	346.46 ± 132.78	324.55 ± 126.67	.257^b	256.79 ± 96.69	346.53 ± 131.2	.006^b	346.46 ± 132.78	324.55 ± 126.67	.257^b	256.79 ± 96.69	346.53 ± 131.2	0.006^b
Androstenedione (ng/mL)^c	1.35 (0.11–2.33)	1.21 (0.01–9.92)	.355^d	1.4 (0.11–2.33)	1.21 (0.01–9.92)	.389^d	1.2 (0.01–3.95)	1.26 (0.1–9.92)	.318^d	1.54 (0.44–9.92)	1.2 (0.01–3.95)	.062^d	1.2 (0.01–3.95)	1.26 (0.1–9.92)	.318^d	1.54 (0.44–9.92)	1.2 (0.01–3.95)	.062^d
TT (ng/mL)^c	0.86 (0.23–1.69)	0.51 (0.01–2.09)	.002^d	0.81 (0.23–1.69)	0.53 (0.01–2.09)	.104^d	0.53 (0.07–2.09)	0.55 (0.01–1.76)	.839^d	0.64 (0.05–1.96)	0.54 (0.01–2.09)	.326^d	0.53 (0.07–2.09)	0.55 (0.01–1.76)	.839^d	0.64 (0.05–1.96)	0.54 (0.01–2.09)	.326^d
FT (pg/mL)^a	1.88 ± 1.07	1.68 ± 0.78	.238^b	1.98 ± 1.17	1.68 ± 0.77	.264^b	1.63 ± 0.81	1.87 ± 0.84	.025^d	2.45 ± 0.76	1.65 ± 0.79	<.001^b	1.63 ± 0.81	1.87 ± 0.84	.025^d	2.45 ± 0.76	1.65 ± 0.79	<.001^b
SHBG (ng/mL)^c	38 (12.5–210)	38.6 (1.35–201)	.903^d	36.35 (12.5–138.7)	38.6 (1.35–210)	.890^d	38.9 (14.2–201)	33.6 (1.35–210)	.100^d	33.6 (17.8–154.3)	38.6 (1.35–210)	.560^d	38.9 (14.2–201)	33.6 (1.35–210)	.100^d	33.6 (17.8–154.3)	38.6 (1.35–210)	.560^d
17OHPG (ng/mL)^c	0.46 (0.21–3.75)	0.38 (0.04–2.56)	.020^d	0.47 (0.21–3.75)	0.38 (0.04–2.56)	.043^d	0.39 (0.04–2.14)	0.41 (0.17–3.75)	.398^d	0.46 (0.1–2.05)	0.39 (0.04–3.75)	.987^d	0.39 (0.04–2.14)	0.41 (0.17–3.75)	.398^d	0.46 (0.1–2.05)	0.39 (0.04–3.75)	.987^d
Prolactin(ng/mL)^c	18 (5.03–31.75)	18.02 (4.7–40.58)	.639^d	18.15 (5.03–31.75)	18.02 (4.7–40.58)	.675^d	18.29 (4.7–40.58)	17.8 (5.03–39.8)	.853^d	19.8 (5.03–39.8)	18 (4.7–40.58)	.754^d	18.29 (4.7–40.58)	17.8 (5.03–39.8)	.853^d	19.8 (5.03–39.8)	18 (4.7–40.58)	.754^d
LH (IU/mL)^c	8.26 (3.58–40.59)	6.32 (0.16–45.7)	.019^d	8.89 (3.6–40.59)	6.32 (0.16–45.7)	.008^d	6.56 (0.16–35.12)	5.85 (0.53–45.7)	.861^d	6.78 (1.67–24.06)	6.4 (0.16–45.7)	.656^d	6.56 (0.16–35.12)	5.85 (0.53–45.7)	.861^d	6.78 (1.67–24.06)	6.4 (0.16–45.7)	.656^d
FSH (IU/mL)^c	6.48 (3.44–11.91)	6.63 (1.68–29.91)	.820^d	6.65 (3.79–11.7)	6.63 (1.68–29.91)	.903^d	6.66 (2.2–29.91)	6.51 (1.68–14.2)	.532^d	6.92 (2.72–14.7)	6.6 (1.68–29.91)	.182^d	6.66 (2.2–29.91)	6.51 (1.68–14.2)	.532^d	6.92 (2.72–14.7)	6.6 (1.68–29.91)	.182^d
FAI^c	3.6 (0.9–18.6)	2.9 (0.15–22.3)	.217^d	2.48 (0.9–18.6)	2.96 (0.15–22.3)	.636^d	2.9 (0.18–20.8)	2.9 (0.15–22.3)	.499^d	2.11 (0.15–7.5)	2.98 (0.18–22.3)	.015^d	2.9 (0.18–20.8)	2.9 (0.15–22.3)	.499^d	2.11 (0.15–7.5)	2.98 (0.18–22.3)	.015^d
FG (mg/dL)^c	98 (76–120)	98 (74–122)	.689^d	97 (76–120)	98 (74–122)	.661^d	97.12 ± 10.75	98.76 ± 10.54	.300^b	97 (76–116)	98 (74–122)	.819^d	97.12 ± 10.75	98.76 ± 10.54	.300^b	97 (76–116)	98 (74–122)	.819^d
FI (mUI/mL)^c	11 (3.96–23.3)	10.2 (3.04–35.1)	.457^d	10.1 (3.96–23.3)	10.38 (3.04–35.1)	.837^d	9.84 ± 3.06	13.76 ± 6.46	<.001^b	8.12 (3.04–14.55)	10.5 (3.96–35.1)	.004^d	9.84 ± 3.06	13.76 ± 6.46	<.001^b	8.12 (3.04–14.55)	10.5 (3.96–35.1)	.004^d
HbA1C (%)^c	5.4 (4.4–6.8)	5.5 (4–11.36)	.541^d	5.4 (4.4–6.3)	5.5 (4–11.36)	.283^d	5.4 (4–11.36)	5.6 (4.4–8)	.017^d	5.4 (4.4–6.7)	5.5 (4–11.36)	.787^d	5.4 (4–11.36)	5.6 (4.4–8)	.017^d	5.4 (4.4–6.7)	5.5 (4–11.36)	.787^d
HDL (mg/dL)^c	46 (36–86)	46.9 (29.9–112)	.476^d	45.65 (36–86)	46.8 (29.9–112)	.289^d	47 (33–112)	45 (29.9–96)	.022^d	46 (36–86)	46.8 (29.9–112)	.574^d	47 (33–112)	45 (29.9–96)	.022^d	46 (36–86)	46.8 (29.9–112)	.574^d
TG (mg/dL)^c	150 (66–301)	152 (33–365)	.979^d	140 (66–301)	152 (33–365)	.961^d	152(33–289)	152 (65–365)	.353^d	167 (65–278)	152 (33–365)	.054^d	152(33–289)	152 (65–365)	.353^d	167 (65–278)	152 (33–365)	.054^d
HOMA-IR^c	2.67 (0.83–6.04)	2.46 (0.81–12.67)	.755^d	2.35 (0.83–6.04)	2.51 (0.81–12.67)	.471^d	2.3 (0.81–12.67)	3.2 (0.83–8.49)	<.001^d	2.22 (1.01–3.61)	2.52 (0.81–12.67)	.310^d	2.3 (0.81–12.67)	3.2 (0.83–8.49)	<.001^d	2.22 (1.01–3.61)	2.52 (0.81–12.67)	.310^d

^a Data were presented as means ± SD.

^b The t test in independent groups.

^c Data were presented as median (min–max).

^d Mann–Whitney U Test.

p < .05 was considered statistically significant.

DHEA, dehydroepiandrosterone; FAI, free androgen index; FI, fasting insulin; FSH, follicle stimulating; FT, free testosterone; HbA1C: Hemoglobin A1 C; HDL, high-density lipoprotein cholesterol; HOMA-IR, Homeostatic model assessment of insulin resistance; LH, luteinizing hormone; SHBG, sex hormone-binding globulin; TG, triglycerides; TT, total testosterone; 17OHPG, 17-hydroxy progesterone.

Discussion

Cutaneous characteristics e.g. acne, HR, AGA, and seborrhea are important for the early diagnosis of PCOS. Although it is already known that excessive androgen and insulin resistance play important roles in the development of cutaneous characteristics, the etiology is still unknown [7].

Acne might be a marker of hyperandrogenism, and persistent, severe, or late-onset acne may suggest PCOS in women. Although the prevalence of acne in women with PCOS was reported to be highly variable, ranging from 15% to 95% [8, 9], acne rate was found to be 85.6% in the present study. Hormonal changes might play roles in the pathophysiology of acne, as well as infectious factors such as propionibacterium acnes. Therefore, the prevalence may be variable. Although there are studies in the literature reporting a positive correlation between acne and high testosterone levels and a negative correlation between SHBG levels and acne [10], there are also some other studies conducted with women with PCOS suggesting that acne local androgen levels near the pilosebaceous unit might not be directly proportional to serum androgen levels [8, 11]. Similarly, unlike what was expected, the DHEA-SO4 values of Acne (–) PCOS group, TT values, 17OHPG, LH values were found to be higher than those of Acne (+) PCOS group. This finding in our study makes us think that local androgen levels may be more important for acne development.

HR is defined as the presence of male terminal or coarse hair in females. PCOS accounts for 70%–80% of the causes of HR [1]. Also, HR is the most common skin manifestation of PCOS, and the prevalence of HR in women with PCOS ranges between 50% and 70% [1]. In the present study, the incidence of HR was found to be 93.2% and the most common cutaneous finding in patients with PCOS was HR. The HR detected in PCOS was generally associated with high levels of androgen and insulin resistance and low SHBG levels in the literature [9]. Falsetti et al. also reported abnormalities in glucose metabolism in patients with HR in PCOS [12]. In the present study, the DHEA-SO4, FAI, and FI values of patients with HR (+) PCOS were found to be statistically higher than those of patients with HR (–) PCOS.

In the physiology of HR, the stimulation of hair growth from the follicle depends not only on the circulating androgen concentration but also on the peripheral metabolism of androgens and the end-organ sensitivity of circulating androgens [13, 14]. Local factors are also effective in this mechanism. Such as l-ornithine decarboxylase enzyme activity is an important contributor to hair follicle development as a local factor that catalayses the synthesis of polyamines implicated in cell migration, androgen receptor concentration, and 17-β hydroxysteroid dehydrogenase and 5 alpha reductase activities [15]. For this reason, the severity of HR may not correlate well with androgen levels because the androgen-dependent hair follicle response to androgen excess differs at significant levels in women. The laboratory results of women with HR might also change for this reason.

Seborrhea was also described as a cutaneous characteristic in PCOS. However, the prevalence of seborrhea is still unknown in PCOS [1]. In the present study, the rate of seborrhea was found to be 89.4%. The factors that affect seborrhea are androgen levels in women, genetic predisposition of the patient, the climate of the place where the patient lives, and psychological factors. Seborrhea may also be associated with seborrheic dermatitis, which is not associated with PCOS, AGA, and acne vulgaris [16].

AGA is characterized by widespread, non-scarring hair loss in the frontal, central, and parietal regions of the scalp as the most common cause of alopecia in women. So far, there have been very few studies examining the relationship between PCOS and AGA. Some studies reported the prevalence of AGA in women with PCOS to be in the range of 67%–77.8% [9, 17]. The rate of AGA in women with PCOS in the study was found to be 91.8%. The pathophysiology of AGA is still not fully understood and multifactorial genetics may be responsible. Although PCOS women with AGA were found to have elevated androgen levels [17], studies are reporting the contrary [18]. Previous studies reported negative correlations among AGA and FT, LDL, and insulin [16]. Feng et al. reported that there were no differences in biochemical hyperandrogenemia between PCOS women with and without AGA, but FAI was higher in the PCOS group at significant levels without TG and insulin AGA [1]. In the present study, the FT values of only AGA (–) PCOS groups were found to be statistically higher than AGA (+) PCOS groups. As a result of these differences, it is possible to argue that AGA is not a marker for hyperandrogenemia, as both Feng et al. [1] and Özdemir et al. [16] reported in their studies.

AN is characterized by locally developing, velvety, brownish, thickened plaques in areas e.g. the neck, armpits, and groin. The prevalence of AN was reported to be between 22.5% and 44.16% in various studies [7, 19, 20]. AN is most commonly detected in diseases e.g. obesity, PCOS, and diabetes, and the incidence of AN in obese women was correlated positively with the severity of obesity [1]. AN may also be associated with elevated free testosterone levels and metabolic dysfunction, multiple genetic variants, reactions to certain drugs, and malignancy [1]. In the present study, the AN rate seen in patients with PCOS was found to be 56.3%. Similarly, in the present study, the FI, HOMA-IR values of patients with AN (+) PCOS were statistically higher than patients with AN (–) PCOS (p < .001, p < .001). A recent study showed that the TT level has a significant effect on insulin secretion and insulin resistance and its effect is more significant in patient with AN (–) than in patients with AN (35012908). In the present study despite the IR is higher in patients with AN (+), TT levels were higher in patients with AN (–) than the patients with AN (+) but it is not statistically significant (respectively p: .31 and p: .32). Although TT was higher, SHBG was lower in patients with AN (–), which may explain the higher FT levels in patients with AN (–).

The reason for the high AN rate in the study might be the differences in the number of body regions including differences in demographic data, and differences in obesity prevalence.

ST, which is also known as acrochordon, is usually benign cutaneous growths that can be recognized as soft tumors of the skin. Tamega et al. reported that there is an association between the presence of more than five STs in dermatological patients and an increase of 1.4 units in the HOMA-IR Index, there is a significant association between BMI and hypertriglyceridemia, and that STs might be an indicator of insulin resistance [21]. In the present study, it was found that 33.2% of the patients who had PCOS were found to have ST. The HbA1c and HOMA-IR values of the ST (+) PCOS group were statistically higher than those of patients with ST (–) PCOS. All these results confirm that ST might be associated with abnormalities in insulin pathogenesis and might be a significant cutaneous marker in PCOS.

Although other studies compared PCOS and cutaneous findings, to the best of our knowledge, the present study is among the few studies that were conducted with the largest number of cutaneous findings using very extensive laboratory tests. Also, all skin lesions were performed by the same dermatologist, and the number of patients who had PCOS and who were included in the study was higher than in other studies. However, the study also had some limitations. First, the appearance of some cutaneous findings and hormonal laboratory results over time may not match our evaluation time because it was a cross-sectional study, and this may have limited the optimal evaluation of cutaneous findings. Second, the measurement of hormonal values might have caused laboratory-based differences. Third, this study did not have a control group with non-PCOS. Finally, the study reflects a very small part of the population because the study was conducted in one single center. For this reason, the results must be supported by multicenter studies with more patient numbers for more precise results.

Conclusion

In conclusion, PCOS is among the most common endocrine disorders in women of reproductive age and can be associated with numerous long-term health issues and significant psychological effects. The cutaneous manifestations of PCOS play important roles in the diagnosis constituting a significant portion of the symptoms experienced by women with PCOS.

Disclosure statement

The authors report there are no competing interests to declare

Funding

The author(s) reported there is no funding associated with the work featured in this article.