The impact of androgen levels on serum metabolic profiles in patients with polycystic ovary syndrome

Abstract

Objective

This study aimed to investigate the impact of serum androgen levels on metabolic profiles in patients with polycystic ovary syndrome (PCOS).

Methods

We included 216 patients with PCOS and 216 healthy individuals selected as the control group. According to the measured serum androgen levels, patients with PCOS were divided into the hyperandrogenism group and non-hyperandrogenism group. Clinical metabolic indicators were assessed and compared between the two groups. Additionally, we assessed the correlation between androgen levels and clinical metabolic indicators.

Results

The body mass index, waist-to-hip ratio, mF-G score, and acne score, as well as T, LH, LSH/FSH, FPG, Cr, UA, TG, TC, and LDL-C levels were significantly higher in the PCOS group than in the control group. The incidence of hyperandrogenism and clinical hyperandrogenism in the PCOS group was significantly higher than that in the control group. Regarding clinical hyperandrogenism, hirsutism, acne, and acanthosis nigricans were significantly more common in the PCOS group than in the control group. Serum androgen levels were significantly correlated with the mF-G score, acne score, FSH, glucose concentration at 30 min, glucose concentration at 60 min, glucose concentration at 120 min, FINS, N120, HOMA-IR, HbA1c, AUCG, UA, TG, and hHDL-Clevels.

Conclusion

Elevated serum androgen levels are commonly observed in patients with PCOS and are associated with multiple metabolic abnormalities. Therefore, it is recommended to regularly monitor glucose and lipid metabolism-related indicators in patients with PCOS who have elevated androgen levels.

Keywords:

• Polycystic ovary syndrome
• hyperandrogenemia
• insulin resistance
• abnormal blood lipids

Polycystic ovary syndrome (PCOS) is an endocrine disorder that causes hormonal imbalances and metabolic abnormalities in adolescent and reproductive-aged women. It is a complex and multifactorial disease with various clinical manifestations. In clinical practice, PCOS is mainly characterized by hyperandrogenism or clinical hyperandrogenism, oligo-ovulation or anovulation, and polycystic ovarian morphology. Additionally, it is closely related to obesity, insulin resistance (IR), metabolic disorders, metabolic syndrome, and cardiovascular diseases [1–3]. PCOS can significantly impact the quality of life and psychological well-being of reproductive-aged women [4].

High levels of androgens are the main endocrine and metabolic characteristics of PCOS; accordingly, multiple diagnostic criteria and expert consensus worldwide emphasize the importance of high androgen levels in the diagnostic criteria for PCOS [5–7]. The phenomenon of high androgen levels in patients with PCOS can be divided into two aspects: hyperandrogenism (HA) and clinical HA. Clinical HA in the skin mainly manifests as hirsutism, acne, androgenic alopecia, and acanthosis nigricans. Patients with PCOS exhibit various metabolic disturbances; further, androgen levels are closely related to IR, which is a significant factor that affects glucose and lipid metabolism.

This study aimed to evaluate the basic types of metabolic disturbances in patients with PCOS by comparing clinical data between individuals with and without PCOS; further, we assessed the correlation of androgen levels with metabolic indicators in patients with PCOS, in order to provide new ideas for clinical management of PCOS patients.

Materials and methods

Study participants

We include 216 patients diagnosed with PCOS (using Rotterdam criteria) in the gynecology outpatient department of the First People’s Hospital of Yunnan Province between January 2021 and December 2022. The age range of the patients was between 18 and 37 years. Additionally, we recruited a control group of 216 healthy females, aged 18 to 38 years old, who underwent physical examinations during the same period. This study was approved by the Medical Ethics Committee, and written informed consent was obtained from all enrolled participants.

Study methods

General clinical data

We collected general information regarding the participants, including age, height, weight, waist circumference (WC), hip circumference (HC), and body mass index (BMI) calculated as weight (kg) divided by height squared (m²), and waist-to-hip ratio (WHR) calculated as WC (cm) divided by HC (cm). All participants underwent a comprehensive physical examination to assess the distribution of pubic and body hair, as well as the presence of acne, androgenic alopecia, and acanthosis nigricans. Hirsutism was evaluated using the modified Ferriman-Gallwey (mFG) score, which assesses hair growth in nine androgen-sensitive body areas. A total mFG score ≥5 was defined as hirsutism [8]. The severity of acne was graded using the Pillsbury grading method. Androgenetic alopecia was graded using the Ludwig grading method.

Examination of biochemical indicators

Fasting venous blood samples were collected from all participants in the morning on days 2–5 of their menstrual cycle for laboratory analysis. The levels of six sex hormones, renal function markers, and lipid metabolism-related biomarkers in serum samples were measured using an automated chemiluminescence analyzer and the corresponding test kits.

Regarding glucose metabolism-related indicators, only fasting plasma glucose (FPG) levels were measured in participants without glucose abnormalities in the control group. Contrastingly, participants in the PCOS group underwent an oral glucose tolerance test and insulin release test, with blood samples being collected via venipuncture at 30, 60, 120, and 180 min after glucose administration to measure blood glucose and insulin levels. The Homeostatic Model Assessment of IR (HOMA-IR) value was calculated as FPG (mmol/L)× fasting insulin (FINS) (mIU/L)/22.5. The area under the curve of glucose (AUCG) value was calculated as 0.5×(FPG + G180)+G30 + G60 + G120, while the area under the curve of insulin (AUCI) value was calculated as 0.5×(FINS + N180)+N30 + N60 + N120.

Statistical analysis

Statistical analyses were performed using SPSS 25.0 software. Continuous variables are expressed as mean ± standard deviation (SD) and were analyzed using independent sample t-tests. Categorical variables are presented as counts and percentages and were analyzed using chi-square tests. Bivariate Pearson’s correlation was used for correlation analysis. Prior to analysis, normality and homogeneity of the variances were tested; moreover, data that met the criteria were analyzed using repeated measures analysis of variance (ANOVA) in order to determine statistical significance, with a significance threshold of p < 0.05.

Results

Comparison of general clinical data and biochemical indicators between the PCOS and control groups

There was no significant between-group difference in age (p > 0.05), indicating comparability. Compared with the control group, the PCOS group had significantly higher BMI, WHR, modified Ferriman–Gallwey score (mF-G) score, and acne score (p < 0.05). Regarding sex hormone levels, there were significant between-group differences in testosterone (T), luteinizing hormone (LH), LSH/follicle stimulating hormone (FSH), and progesterone (P) levels (p < 0.05). Specifically, T, LH, and LH/FSH levels were significantly higher in the PCOS group than in the control group, while P levels were significantly lower in the PCOS group than in the control group. Regarding metabolic indicators, FPG, creatinine (Cr), uric acid (UA), triglyceride (TG), total cholesterol (TC), and low density lipoprotein-cholesterol (LDL-C) levels were significantly higher in the PCOS group than in the control group (p < 0.05). Contrastingly, HDL-C levels were non-significantly lower in the PCOS group than in the control group (p > 0.05). The between-group comparison of general clinical data and biochemical indicators is presented in Table 1.

Table 1. Comparison of general clinical data and biochemical indicators between the PCOS group and the control group.

	PCOS (n = 216)	Control (n = 216)	t	p
Age (years)	25.70 ± 4.92	26.07 ± 4.50	−0.816	0.415
BMI(Kg/m^2)	24.63 ± 5.34	20.96 ± 1.98	9.471	0.000**
WHR	0.84 ± 0.05	0.77 ± 0.04	14.210	0.000**
mF-G(points)	6.69 ± 5.17	3.05 ± 2.80	9.104	0.000**
Acne (points)	1.18 ± 0.99	0.58 ± 0.76	7.090	0.000**
T(ng/mL)	1.05 ± 0.58	0.51 ± 0.24	12.672	0.000**
LH(mIU/mL)	13.00 ± 5.51	5.97 ± 1.45	18.147	0.000**
FSH(mIU/mL)	6.23 ± 1.74	6.31 ± 1.26	−0.520	0.604
LH/FSH	2.14 ± 0.92	0.96 ± 0.22	18.356	0.000**
E2(pg/mL)	44.48 ± 13.90	46.26 ± 17.17	−1.184	0.237
PRL(ng/mL)	13.96 ± 6.55	12.94 ± 6.52	1.620	0.106
P(ng/mL)	0.56 ± 0.26	0.66 ± 0.46	−2.936	0.004**
FPG(mmol/L)	5.00 ± 0.58	4.82 ± 0.46	3.625	0.000**
BUN(mmol/L)	4.61 ± 1.28	4.63 ± 1.28	−0.177	0.859
Cr(umol/L)	57.93 ± 7.62	54.71 ± 8.30	4.197	0.000**
UA(umol/L)	378.75 ± 90.39	307.19 ± 59.57	9.714	0.000**
TG(mmol/L)	1.54 ± 1.15	1.00 ± 0.37	6.530	0.000**
TC(mmol/L)	4.58 ± 0.86	4.42 ± 0.64	2.281	0.023*
HDL-C(mmol/L)	1.29 ± 0.30	1.30 ± 0.20	−0.088	0.930
LDL-C(mmol/L)	2.78 ± 0.75	2.56 ± 0.53	3.500	0.001**

ps: *p < 0.05; **p < 0.01.

Between-group comparison of the incidence of HA and clinical HA

Based on the laboratory reference values, T levels above 0.75 ng/mL were defined as biochemical HA. Hirsutism, acne, androgenic alopecia, and acanthosis nigricans were all considered clinical manifestations of elevated androgens. If a patient had one or more of these symptoms, they were considered to have clinical HA. As shown in Table 2, there were significant between-group differences in the incidence of HA and clinical HA (p < 0.05). The incidence of hirsutism, acne, and acanthosis nigricans was significantly higher in the PCOS group than in the control group (59.3% vs. 27.3%, 75.9% vs. 44.9%, 13.0% vs. 6.0%, respectively); however, there was no significant between-group difference in the incidence of androgenic alopecia (5.1% vs. 2.8%; p > 0.05).

Table 2. Comparison of the occurrence of hyperandrogenism and physical signs between the PCOS group and the control group.

	PCOS(n = 216)	Control(n = 216)	X2	p
Hyperandrogenism			72.202	0.000**
Yes	133(61.6%)	46(21.3%)
No	83(38.4%)	170(78.7%)
KaohsiungPhysical Signs			47.457	0.000**
Yes	193(89.4%)	131(60.6%)
No	23(10.6%)	85(39.4%)
Hirsutism			44.893	0.000**
Yes	128(59.3%)	59(27.3%)
No	88(40.7%)	157(72.7%)
Acne			43.451	0.000**
Yes	164(75.9%)	97(44.9%)
No	52(24.1%)	119(55.1%)
Androgenic alopecia			1.531	0.216
Yes	11(5.1%)	6(2.8%)
No	205(94.9%)	210(97.2%)
Echinosis nigra			6.063	0.014*
Yes	28(13.0%)	13(6.0%)
No	188(87.0%)	203(94.0%)

ps: *p < 0.05; **p < 0.01.

Comparison of patients in the PCOS group with and without HA and clinical HA

Patients with PCOS were divided into HA and non-HA groups based on their serum T levels, followed by between-group comparisons of the general clinical data and sex hormone levels. The HA group had significantly higher WHR, mF-G scores, and acne scores than the non-HA group (p < 0.05). Regarding sex hormone levels, T and FSH levels were significantly higher and lower, respectively, in the HA group than in the non-HA group. Additionally, the patients were divided into the clinical HA group and non-clinical HA group based on the presence of clinical HA. The clinical HA group had significantly higher mF-G and acne scores than the non-clinical HA group (p < 0.05). See Table 3 for details.

Table 3. Comparison of general clinical data and sex hormone indicators between the PCOS hyperandrogenism group and the non hyperandrogenism group, as well as the hyperandrogenicity sign group and the non hyperandrogenicity sign group.

	Hyperandrogenism (n = 133)	No Hyperandrogenism (n = 83)	p	Kaohsiung Physical Signs (n = 193)	No Kaohsiung Physical Signs (n = 23)	p
Age(years)	25.53 ± 4.82	25.96 ± 5.11	0.534	25.45 ± 4.93	27.78 ± 4.46	0.031*
Height(cm)	159.21 ± 5.00	160.04 ± 6.91	0.310	159.61 ± 5.93	158.83 ± 4.64	0.541
BMI(Kg/m2)	25.11 ± 5.44	23.85 ± 5.11	0.092	24.86 ± 5.48	22.66 ± 3.52	0.061
WHR	0.84 ± 0.05	0.83 ± 0.05	0.036*	0.84 ± 0.05	0.83 ± 0.04	0.342
mF-G(points)	6.47 ± 4.52	4.82 ± 4.28	0.008**	6.06 ± 4.53	3.91 ± 3.68	0.030*
Acne (points)	1.16 ± 0.94	1.00 ± 0.73	0.169	1.17 ± 0.86	0.52 ± 0.67	0.001**
T(ng/mL)	1.34 ± 0.55	0.59 ± 0.22	0.000**	1.07 ± 0.60	0.85 ± 0.38	0.019*
LH(mIU/mL)	12.92 ± 5.58	13.14 ± 5.42	0.768	13.10 ± 5.56	12.17 ± 5.11	0.444
FSH(mIU/mL)	6.00 ± 1.48	6.60 ± 2.05	0.015*	6.21 ± 1.70	6.43 ± 2.12	0.571
LH/FSH	2.18 ± 0.85	2.08 ± 1.03	0.466	2.17 ± 0.94	1.93 ± 0.66	0.248
E2(pg/mL)	45.57 ± 14.76	42.74 ± 12.30	0.129	44.26 ± 14.05	46.37 ± 12.77	0.493
PRL(ng/mL)	13.73 ± 5.73	14.32 ± 7.70	0.519	13.95 ± 6.44	14.01 ± 7.50	0.971
P(ng/mL)	0.55 ± 0.24	0.56 ± 0.29	0.768	0.55 ± 0.25	0.58 ± 0.36	0.621

ps: *p < 0.05; **p < 0.01.

Comparison of metabolic indicators according to the presence of HA and clinical HA

Regarding metabolic indicators, the HA group had significantly higher levels of glucose concentration at 120 min (G120), FINS, insulin concentration at 30 min (N30), insulin concentration at 120 min (N120), HOMA-IR, AUCI, UA, TG, and TC, as well as significantly lower HDL-C levels, than the non-HA group (p < 0.05). Moreover, the clinical HA group had significantly higher levels of glucose concentration at 30 min (G30), FINS, N30, insulin concentration at 180 min (N180), HOMA-IR, and AUCI than the non-clinical HA group (p < 0.05). For detailed comparison of the clinical metabolic indicators, please see Table 4.

Table 4. Comparison of metabolic related clinical indicators between the PCOS hyperandrogenism group and the non hyperandrogenism group, as well as the hyperandrogenicity sign group and the non hyperandrogenicity sign group.

	Hyperandrogemia (n = 133)	No Hyperandrogemia (n = 83)	p	Kaohsiung syndrome (n = 193)	No Kaohsiung syndrome (n = 23)	p
FPG(mmol/L)	5.02 ± 0.38	4.93 ± 0.36	0.103	5.00 ± 0.39	4.92 ± 0.26	0.374
G30(mmol/L)	8.41 ± 1.57	8.14 ± 1.39	0.190	8.39 ± 1.48	7.60 ± 1.61	0.018*
G60(mmol/L)	7.73 ± 2.11	7.42 ± 2.12	0.285	7.69 ± 2.11	6.92 ± 2.10	0.098
G120(mmol/L)	6.53 ± 1.54	6.02 ± 1.40	0.016*	6.35 ± 1.48	6.21 ± 1.71	0.675
G180(mmol/L)	4.97 ± 1.25	4.79 ± 1.00	0.255	4.88 ± 1.15	5.09 ± 1.23	0.420
FNS(mIU/L)	20.81 ± 13.77	15.24 ± 10.12	0.001**	19.25 ± 13.12	13.77 ± 7.81	0.006*
N30(mIU/L)	161.31 ± 107.02	127.55 ± 69.35	0.005**	151.99 ± 99.04	117.69 ± 50.82	0.010*
N60(mIU/L)	161.69 ± 117.32	136.30 ± 100.38	0.104	154.95 ± 115.39	126.65 ± 68.39	0.094
N120(mIU/L)	146.08 ± 120.55	107.91 ± 80.31	0.006**	134.31 ± 112.64	107.12 ± 56.50	0.063
N180(mIU/L)	66.72 ± 66.73	58.05 ± 66.60	0.354	65.53 ± 69.69	45.37 ± 25.63	0.008**
HOMA-IR	4.72 ± 3.30	3.39 ± 2.37	0.001**	4.35 ± 3.13	3.01 ± 1.73	0.003**
HbA1c(%)	5.40 ± 0.38	5.32 ± 0.37	0.141	5.36 ± 0.36	5.40 ± 0.57	0.697
AUCG	27.67 ± 5.03	26.44 ± 4.55	0.070	27.37 ± 4.78	25.74 ± 5.48	0.129
AUCI	512.84 ± 335.37	408.40 ± 240.85	0.008**	483.64 ± 319.01	381.02 ± 135.57	0.007**
BUN(mmol/L)	4.56 ± 1.25	4.70 ± 1.34	0.435	4.62 ± 1.28	4.55 ± 1.33	0.821
Cr(umol/L)	58.60 ± 7.76	56.84 ± 7.31	0.099	58.20 ± 7.70	55.61 ± 6.66	0.123
UA(umol/L)	393.83 ± 92.11	354.59 ± 82.50	0.002**	381.05 ± 91.17	359.48 ± 82.95	0.280
TG(mmol/L)	1.76 ± 1.30	1.23 ± 0.74	0.000**	1.58 ± 1.15	1.35 ± 1.11	0.349
TC(mmol/L)	4.77 ± 0.84	4.28 ± 0.80	0.000**	4.57 ± 0.87	4.68 ± 0.74	0.559
HDL-C(mmol/L)	1.24 ± 0.25	1.33 ± 0.34	0.040*	1.28 ± 0.30	1.26 ± 0.20	0.753
LDL-C(mmol/L)	2.80 ± 0.70	2.74 ± 0.83	0.602	2.77 ± 0.76	2.86 ± 0.66	0.551

ps: *p < 0.05; **p < 0.01.

Correlation analysis between serum testosterone and various clinical indicators in the PCOS group

The correlation analysis results are shown in Table 3. Serum testosterone levels showed a significant positive correlation with the Mf-G and acne scores (p < 0.05), with correlation coefficients of 0.537 and 0.265, respectively. Serum testosterone levels were negatively correlated with FSH levels (p < 0.05), with a correlation coefficient of −0.228. Serum testosterone levels showed a significant positive correlation with the glucose metabolism-related indicators of G30, G60, G120, FINS, N120, HOMA-IR, glycated hemoglobin (HbA1c), and AUCG (p < 0.05). Moreover, serum testosterone levels were positively correlated with the renal function and purine metabolism-related indicator UA, with a correlation coefficient of 0.201. Serum testosterone levels showed a significant positive and negative correlation with the blood lipid indicators TG (p < 0.05; correlation coefficient: 0.231) and HDL-C (p < 0.05; correlation coefficient: −0.226), respectively. Table 5 shows the results of correlation analysis between serum testosterone and various clinical indicators in the PCOS group.

Table 5. Correlation analyses between T and various clinical indicators in the total PCOS group, hyperandrogenic PCOS group, and non hyperandrogenic PCOS group.

	Total PCOS(n = 216)		Hyperandrogenism(n = 133)		No Hyperandrogenism (n = 83)
	r	p	r	p	r	p
Age (years)	−0.007	0.914	0.019	0.825	0.054	0.626
Height(cm)	0.043	0.528	0.141	0.105	0.090	0.418
BMI(Kg/m2)	0.099	0.145	0.027	0.754	0.083	0.458
WHR	0.119	0.080	0.019	0.832	0.137	0.217
mF-G(points)	0.537	0.000**	0.502	0.000**	0.421	0.000**
Acne (points)	0.265	0.000**	0.173	0.046*	0.259	0.018*
LH(mIU/mL)	−0.118	0.084	−0.168	0.053	−0.039	0.725
FSH(mIU/mL)	−0.228	0.001**	−0.239	0.006**	−0.034	0.760
LH/FSH	−0.008	0.904	−0.056	0.522	−0.058	0.606
E2(pg/mL)	0.053	0.443	−0.037	0.672	0.105	0.345
PRL(ng/mL)	−0.048	0.486	−0.039	0.659	−0.002	0.982
P(ng/mL)	−0.020	0.773	−0.020	0.815	0.026	0.817
FPG(mmol/L)	0.107	0.117	0.062	0.477	0.003	0.980
G30(mmol/L)	0.178	0.009**	0.151	0.083	0.223	0.043
G60(mmol/L)	0.206	0.002**	0.211	0.015*	0.259	0.018
G120(mmol/L)	0.262	0.000**	0.244	0.005**	0.105	0.346
G180(mmol/L)	0.088	0.199	0.077	0.376	−0.074	0.504
FNS(mIU/L)	0.161	0.018*	0.010	0.906	0.194	0.079
N30(mIU/L)	0.123	0.071	0.002	0.984	0.139	0.209
N60(mIU/L)	0.074	0.278	−0.027	0.755	0.167	0.131
N120(mIU/L)	0.138	0.043*	0.017	0.846	0.185	0.095
N180(mIU/L)	0.054	0.433	0.001	0.995	0.095	0.392
HOMA-IR	0.176	0.010**	0.034	0.696	0.184	0.097
HbA1c(%)	0.168	0.013*	0.141	0.106	0.158	0.153
AUCG	0.240	0.000**	0.223	0.010**	0.213	0.053
AUCI	0.124	0.070	−0.003	0.976	0.189	0.088
BUN(mmol/L)	−0.081	0.236	−0.075	0.393	−0.034	0.760
Cr(umol/L)	0.080	0.242	0.023	0.791	−0.030	0.790
UA(umol/L)	0.201	0.003**	0.100	0.250	0.069	0.536
TG(mmol/L)	0.231	0.001**	0.144	0.099	−0.047	0.672
TC(mmol/L)	0.035	0.609	−0.224	0.010**	−0.101	0.364
HDL-C(mmol/L)	−0.226	0.001**	−0.242	0.005	−0.089	0.424
LDL-C(mmol/L)	−0.024	0.726	−0.081	0.353	−0.019	0.864

ps: *p < 0.05; **p < 0.01.

Discussion

PCOS is an endocrine and metabolic disorder that can occur due to both genetic and non-genetic factors. In this study, we compared general information and laboratory indicators between the PCOS and control groups. The PCOS group had significantly higher levels of BMI, WHR, mF-G score, acne score, T, LH, LH/FSH, FPG, Cr, UA, TG, TC, and LDL-C, as well as significantly lower P levels, than the control group.

Compared with the control group, the PCOS group had a higher BMI, with a mean of (24.63 ± 5.34), and predominantly exhibited abdominal obesity. Obesity may be associated with various clinical features of PCOS, affecting its metabolic and reproductive manifestations [9]. A clinical epidemiological survey revealed that patients with PCOS who have a high BMI are more susceptible to clinical or biochemical HA. The primary characteristics of hormonal disruption in PCOS are elevated levels of T, LH, and LH/FSH ratio, while low P levels in patients with PCOS indicate oligo-ovulation or anovulation. PCOS is mainly characterized by abnormal levels of androgens, often resulting in an increase in androgen levels, while E₂ levels are not specific. E₂ appears as normal or slightly elevated, and some patients may experience a decrease in E₂, which affects their normal fertility. Additionally, between-group comparisons of metabolic indicators, including FPG, Cr, UA, TG, TC, and LDL-C, indicates that PCOS is primarily characterized by abnormal glucose metabolism, lipid metabolism, and purine metabolism (uric acid).

HA is characterized by elevated serum levels of biologically active androgens, leading to hirsutism, acne, androgenic alopecia, and acanthosis nigricans [10]. In PCOS, the presence of excessive androgen levels can be classified as HA and clinical HA, which may coexist or exist independently. In this study, the incidence of HA in PCOS (61.6%) was lower than that of clinical HA (89.4%). This indicates that some patients with PCOS may present with clinical HA without an elevation in serum testosterone levels, potentially suggesting a lack of a clear correlation between HA and clinical HA [8,11]. The incidence of hirsutism and acne vulgaris in patients with PCOS is higher than in the normal population. Specifically, 59.3% of patients with PCOS present with hirsutism, typically exhibiting mild to moderate severity, while 75.9% of patients with PCOS present acne vulgaris. However, the incidence of androgenic alopecia and acanthosis nigricans among patients with PCOS is relatively low, at only 5.1% and 13.0%, respectively, which is consistent with the interpretation of the 2018 guideline for the assessment and management of PCOS [12].

HA and IR are the most prominent characteristics of metabolic syndrome in PCOS, with both of them being involved in the development and progression of PCOS. PCOS is characterized by abnormal glucose metabolism, with 50%–70% of patients with PCOS presenting with IR; moreover, the incidence of IR in obese patients with PCOS is as high as 95% [13]. Androgen excess may lead to impaired glucose tolerance and pancreatic β-cell dysfunction in women with PCOS [14,15]. Among patients with PCOS, comparison of glucose metabolism-related indicators between the HA and non-HA groups revealed significantly increased values of G120, FINS, N30, N120, HOMA-IR, and AUCI in the HA group. Moreover, there was a significant positive correlation (p < 0.05) between testosterone levels and glucose metabolism-related indicators, including G30, G60, G120, FINS, N120, HOMA-IR, HbA1c, and AUCG. Androgen excess may cause dysfunction of pancreatic β-cells in women with PCOS through both direct and indirect mechanisms, leading to excessive insulin secretion, IR, and impaired glucose tolerance [16]. HA is associated with insulin clearance; furthermore, excessive fat deposition in patients with PCOS may result in hyperinsulinemia due to an increase and decrease in insulin secretion and clearance, respectively [17].

The incidence of dyslipidaemia among patients with PCOS reaches up to 60%, with increasing evidence suggesting that dyslipidaemia in PCOS is related to androgen excess and IR [18–20]. In our study, patients with PCOS were divided into HA and non-HA groups, followed by a between-group comparison of lipid metabolism-related indicators. The HA group exhibited significantly higher TG and TC levels than the non-HA group; further, there was a significant positive correlation between testosterone and TG levels. Patients with clinical HA and IR exhibit more pronounced lipid metabolism disorders, suggesting a possible correlation between dyslipidaemia and androgen levels [21]. Patients with PCOS often present with abdominal obesity, which can lead to increased visceral fat breakdown, resulting in elevated free fatty acid (FFA) levels and increased TG synthesis. FFAs can increase lipase activity, which significantly influences the levels of TG, HDL-C, LDL-C, and other lipids [22]. Moreover, androgen excess can lead to increased hepatic lipase activity and increased breakdown of adipocytes in patients with PCOS, resulting in elevated TG and LDL levels, as well as decreased HDL levels. However, since this study was a cross-sectional study, we could not analyze lifestyle and dietary patterns, which limited our ability to systematically evaluate the impact of these factors on glucose and lipid metabolism-related indicators in the patients with PCOS.

In conclusion, PCOS is a complex endocrine disorder that can cause multiple metabolic abnormalities. Our study primarily compared metabolic indicators between patients with PCOS with and without HA and conducted correlation analysis between testosterone and relevant parameters to explore the possible relationship between androgen excess and metabolic parameters. Our findings indicated that the patients with PCOS who have HA exhibited more pronounced glucose and lipid metabolism disorders than patients without HA. Therefore, we believe that regular monitoring of glucose and lipid metabolism-related indicators should be emphasized in patients with PCOS who have elevated androgen levels. Moreover, timely treatment should be provided for patients with PCOS who have combined glucose and lipid metabolism disorders.

Ethics approval and consent to participate

This study was approved by the Ethical Research Committee at the First People’s Hospital of Yunnan Province (approval number: 2018FE117). Written informed consent was obtained from the participants and in the case of minors, informed consent was granted from their legally authorized representatives with the option to withdraw them from the study at any time.

Acknowledgment

We would like to thank Editage (www.editage.cn) for English language editing.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data openly available in a public repository that issues datasets with DOIs

Additional information

Funding

This work was supported by Yunnan Basic Research Program (2018FE117); Yunnan Provincial Center for Obstetrics and Gynecology Research (2022YJZX-FC08; 2023YJZX-FC05); Key Laboratory of Birth Defects and Genetic Diseases Research in Yunnan Province (2022ZDFKT002).