Abstract
Objective
While cardiometabolic comorbidities have been well established in adults with polycystic ovary syndrome (PCOS), it remains unclear whether the early signs of cardiovascular morbidity emerge during adolescence. This study aimed to investigate whether metabolic and cardiovascular alterations begin in adolescents with PCOS.
Methods
The study included adolescents with PCOS and age and body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) matched healthy controls who were further grouped based on weight status as overweight/obese and normal weight. All participants underwent metabolic assessment through physical examination, laboratory tests, conventional echocardiography, tissue Doppler imaging, strain echocardiography, and 24‐h Holter monitoring.
Results
The PCOS group (n = 33) consisted of 19 overweight/obese and 14 normal‐weight participants, while the control group (n = 43) included 20 overweight/obese and 23 normal‐weight participants. Compared to controls, the PCOS group exhibited higher waist‐to‐hip ratio, liver enzymes, triglycerides, triglyceride‐glucose index, and lower high‐density lipoprotein cholesterol (P < 0.05). Among BMI subgroups, overweight/obese PCOS group had the highest insulin and homeostasis model assessment of insulin resistance levels compared to all other groups. The PCOS group showed higher left atrial (LA) end‐systolic diameter, LA/aortic ratio, and lower left ventricular global peak longitudinal strain in the apical four‐chamber view than controls (P < 0.05). Among BMI subgroups, the combination of PCOS and overweight/obesity was associated with the most unfavorable metabolic and cardiovascular profiles. The markers of LA diameter, LA/aortic ratio, and left ventricular mass were highest in the overweight/obese PCOS group.
Conclusion
Selected metabolic and echocardiographic markers showed modest differences when PCOS coexisted with obesity, suggesting a potentially higher cardiometabolic risk compared with PCOS or obesity alone. Early identification and targeted management are essential to mitigate long‐term cardiometabolic complications in this high‐risk group.
Keywords: adolescence, cardiovascular, obesity, polycystic ovary syndrome
1. INTRODUCTION
Diagnosing polycystic ovary syndrome (PCOS) during adolescence requires careful interpretation, as normal pubertal physiological changes, such as physiological menstrual irregularity, together with normal pubertal metabolic changes, including transient insulin resistance and fluctuations in cardiometabolic parameters such as lipid profiles and blood pressure (BP), may overlap with diagnostic features of the condition. 1 , 2 Despite this complexity, international PCOS guidelines emphasize the importance of timely and accurate diagnosis to prevent long‐term comorbidities and improve quality of life. 1
PCOS is associated with an increased long‐term risk of both metabolic and non‐metabolic disorders 3 and has been linked to subclinical atherosclerotic cardiovascular (CV) changes, including increased carotid intima‐media thickness, coronary artery calcification, and endothelial dysfunction. 4 Although evidence supports an association between PCOS and cardiometabolic risk, its relationship with overt clinical CV disease remains debated. While some meta‐analyses have reported an increased risk of myocardial infarction and stroke, 5 , 6 others have found no significant difference in fatal or non‐fatal CV events. 7
PCOS has also been associated with an increased risk of diastolic dysfunction, as evidenced by significantly prolonged isovolumetric relaxation times on echocardiography (ECHO). 8 Nevertheless, how the increased CV risk profile in PCOS translates into CV disease and related mortality remains uncertain. A meta‐analysis found no significant association between PCOS and CV disease‐related mortality. 9 However, it is still recommended that women with PCOS be considered at increased risk for CV disease and mortality. 10
Although metabolic and CV alterations in PCOS are known to progress over time and have been well documented in adults, it remains unclear whether these changes begin during adolescence or emerge later in life. Understanding the onset and other associated risk factors of these alterations is crucial for early intervention and prevention strategies. The aim of this study was to evaluate whether the metabolic and subclinical CV alterations observed in adult women with PCOS originate during adolescence.
2. MATERIALS AND METHODS
This prospective cross‐sectional comparative study was conducted between September 2021 and March 2023 at Hacettepe University, Division of Adolescent Medicine, and is reported in accordance with the STROBE Statement guidelines for cross‐sectional studies. The study protocol was approved by the Hacettepe University Non‐Interventional Clinical Research Ethics Committee on January 19, 2021, with project number GO 21/67 and decision number 2021/02‐47. Written informed consent was obtained from all participants and their parents or legal guardians.
Adolescent girls aged 12–18 years who were at least 2 years post‐menarche were recruited consecutively among eligible adolescents who presented to the outpatient clinic during the study period. Participants were clinically assessed and enrolled by three pediatric adolescent medicine specialists. Eligible patients were invited to participate, and those who provided informed consent were enrolled. Participants were classified as having PCOS (study group) or as controls.
PCOS was diagnosed according to the 2015 Pediatric Endocrine Society criteria, requiring both ovulatory dysfunction and clinical and/or biochemical hyperandrogenism, after exclusion of other causes. 11 Ovulatory dysfunction was defined as persistent cycles <20 or >45 days, ≥2 years post‐menarche. Biochemical hyperandrogenism was defined as total testosterone >55 ng/dL; clinical hyperandrogenism as a modified Ferriman‐Gallwey score ≥8 or moderate‐to‐severe inflammatory acne unresponsive to topical treatment. 11 The control group was prospectively recruited from healthy adolescents attending routine well‐child visits at the same outpatient clinic. Controls had regular menstrual cycles, no clinical signs of hyperandrogenism, and did not meet any exclusion criteria. Controls were frequency‐matched to the PCOS group by age (±1 year) and body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) category. BMI categories were defined using CDC age‐ and sex‐specific percentiles (normal weight: 5th–85th percentile; overweight/obese: ≥85th percentile). 12 Each group was further stratified by BMI into overweight/obese and normal‐weight subgroups. Group assignment and BMI stratification were determined prior to cardiac and metabolic assessments; thus, these outcome measures were not available during diagnostic evaluation. Exclusion criteria for both groups included refusal to participate, smoking, pregnancy, use of oral contraceptives, and chronic conditions such as structural heart disease, arrhythmia, CV disease, hypertension, diabetes, thyroid dysfunction, Cushing's syndrome, androgen‐secreting tumors, late‐onset congenital adrenal hyperplasia, or hyperprolactinemia. Participants found to have structural heart disease during the study were also excluded. All participants were of Turkish ethnicity, resulting in an ethnically homogeneous study population; therefore, ethnicity was not included as a covariate in the analyses.
A detailed medical and menstrual history (age at menarche, cycle length, and duration) was obtained, and a standardized physical examination was performed for each participant by the same investigator. Data were recorded using a piloted form and entered into an electronic database, with cross‐checking performed to ensure accuracy. Anthropometric measurements were performed in the morning after an overnight fast under standardized clinical conditions. Weight and height were measured by the same investigator, using a calibrated digital scale and a wall‐mounted stadiometer, with participants wearing light clothing and no shoes. Each measurement was obtained twice, and the average value was used for analysis. Waist circumference was measured at the level of the umbilicus and hip circumference at the widest part of the buttocks using a non‐stretchable tape, with participants standing and breathing normally. Measurements were performed twice and averaged. Waist‐to‐hip ratio was then calculated. BP was measured using a calibrated automated sphygmomanometer and recorded as the average of three seated measurements after 10 min of rest. Hirsutism was assessed by the same investigator, using the modified Ferriman–Gallwey scoring system and a score of ≥8 was considered indicative of hirsutism. 13 , 14 To ensure consistency, all assessments were performed by a single trained examiner, eliminating interobserver variability.
Adolescents diagnosed with PCOS and control participants underwent blood sampling and cardiac evaluations on the same day, provided an appropriate fasting period had been observed. If participants had not fasted or if cardiac assessments could not be scheduled on the same day, a separate appointment was arranged, and blood sampling, ECHO, and Holter monitoring were performed on that date before the initiation of treatment. Blood samples were collected from all participants to assess hyperandrogenism and metabolic comorbidities. Hyperandrogenism was evaluated by measuring total testosterone, dehydroepiandrosterone sulfate (DHEAS), and sex hormone‐binding globulin (SHBG). For the differential diagnosis of PCOS, follicle‐stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), prolactin, thyroid‐stimulating hormone (TSH), free T4 (fT4), and 17‐hydroxyprogesterone were analyzed. Metabolic comorbidities were assessed by measuring alanine aminotransferase (ALT), aspartate aminotransferase (AST), triglycerides, high‐density lipoprotein cholesterol (HDL‐C), low‐density lipoprotein cholesterol (LDL‐C), total cholesterol, fasting glucose, hemoglobin A1c (HbA1c), and fasting insulin.
Blood samples were collected in the early follicular phase (within the first 7 days of the menstrual cycle) or at any time if the participant had amenorrhea for ≥3 months. Samples were obtained at 9:00 a.m. after 12 h of fasting. Biochemical and hormonal parameters were assessed using standardized automated assays in the hospital central laboratory. Detailed methodologies are provided in the Supporting Information.
Insulin resistance was evaluated using the homeostasis model assessment of insulin resistance (HOMA‐IR) and the triglyceride‐glucose (TyG) index. A HOMA‐IR threshold of 3.82 15 and a TyG index threshold of 8.45 16 were considered indicative of insulin resistance. HOMA‐IR was calculated using the formula: HOMA‐IR = fasting insulin (μU/mL) × fasting glucose (mg/dL)/405. 15 TyG index was calculated using the following formula: TyG index = ln (fasting glucose [mg/dL] × triglycerides [mg/dL]/2). 16
The diagnosis of metabolic syndrome was made according to the age‐specific diagnostic criteria outlined in the 2007 International Diabetes Federation consensus report for children and adolescents. In addition to central obesity, the presence of any two of the following criteria—elevated triglycerides, low HDL‐C, elevated BP, or elevated blood glucose—was required for diagnosis. 17
Left ventricular (LV) function of the PCOS and control groups was evaluated using conventional ECHO, tissue Doppler imaging (TDI), and strain ECHO. All participants underwent 24‐h Holter monitoring. All cardiac assessments were performed and analyzed by the same pediatric cardiology specialist, who was blinded to the participants' clinical diagnosis (PCOS vs control) and BMI category. Details of the conventional ECHO, TDI, strain ECHO, and Holter monitoring protocols are provided in the Supporting Information.
Clinical and laboratory data were recorded on standardized case report forms and entered into a secure electronic database with double‐entry verification and random cross‐checking. The database was password‐protected and accessible only to authorized research personnel.
2.1. Statistical analysis
To determine the required sample size, an a priori power analysis was conducted using G*Power (version 3.1). The calculation was based on previously reported differences in myocardial performance index (MPI) between adolescents with PCOS and healthy controls. 18 Based on these data, a moderate effect size was assumed for the primary outcome. Using a two‐tailed independent samples t‐test, with a significance level of α = 0.05 and a desired statistical power of 80% (1 − β = 0.80), the required sample size was calculated as 40 participants per group. While the target number of participants in the control groups was achieved, the planned sample size for the PCOS groups could not be fully reached due to recruitment limitations, which were partly related to the COVID‐19 pandemic period, during which some patients were reluctant to visit the hospital and pandemic‐related restrictions were still in place.
Analyses were performed in SPSS version 20.0, with significance set at P less than 0.05. Categorical variables are expressed as frequencies and percentages; continuous variables (e.g., age) as mean ± standard deviation (SD), median, and range. Normality was assessed using the Shapiro–Wilk test and visual inspection of histograms. Homogeneity of variances was evaluated using Levene's test prior to parametric analyses. If any group violated the normality assumption, non‐parametric tests were used. Between‐group comparisons (PCOS vs control) employed the Student's t‐test or Mann–Whitney U test, as appropriate. For comparisons among four subgroups (overweight/obese PCOS, normal‐weight PCOS, overweight/obese control, normal‐weight control), one‐way analysis of variance (ANOVA) or Kruskal–Wallis test was used. Post hoc comparisons following Kruskal–Wallis were performed using Bonferroni‐adjusted P values, thereby controlling for multiple comparisons. Post hoc comparisons after ANOVA were performed using the Student–Newman–Keuls test, which controls type I error in a stepwise manner. Two‐way ANOVA examined group (PCOS/control) and BMI (normal/overweight‐obese) effects, testing main and interaction effects. Two‐way ANOVA was selected to simultaneously evaluate the independent and interaction effects of group and BMI category, allowing assessment of potential effect modification while maintaining statistical efficiency compared with conducting multiple separate one‐way analyses. Effect sizes were reported using partial eta squared (η 2): <0.01 (negligible), 0.01–0.059 (small), 0.06–0.139 (medium), ≥0.14 (large). A minimal proportion of data was missing for some variables; analyses were performed using complete‐case analysis without imputation.
3. RESULTS
A total of 86 adolescents were included: 43 with PCOS and 43 controls. Of the PCOS group, nine patients were excluded at baseline and one during follow‐up, leaving 33 patients who completed the study (see Figure 1). Mean age was 15.7 ± 0.8 years in the PCOS group and 15.4 ± 1.3 years in controls (P = 0.215). Mean BMI was 25.9 ± 6.4 versus 24.5 ± 5.2, respectively (P = 0.297). Among PCOS participants, 19 (57.6%) were overweight/obese and 14 (42.4%) normal‐weight; in controls, 20 (46.5%) were overweight/obese and 23 (53.5%) normal‐weight.
FIGURE 1.
Flow chart illustrating the inclusion and exclusion of the polycystic ovary syndrome (PCOS) group.
No significant differences were observed in BMI, waist and hip circumferences, waist‐to‐hip ratio, and systolic and diastolic BP between overweight/obese PCOS and overweight/obese controls. However, the waist‐to‐hip ratio was higher in normal‐weight PCOS compared to normal‐weight controls (P < 0.001).
ALT levels were higher in the overweight/obese PCOS group than in both normal‐weight PCOS and control groups (P < 0.01, partial η 2 = 0.207), indicating a large effect size. AST levels were also elevated compared to the normal‐weight control group (P < 0.05, partial η 2 = 0.156), likewise demonstrating a large effect. Triglycerides were highest (P < 0.05, partial η 2 = 0.154) and HDL‐C lowest (P < 0.05, partial η 2 = 0.018) in the overweight/obese PCOS subgroup compared to normal‐weight controls, indicating a large effect size for triglycerides and a small effect size for HDL‐C.
Fasting glucose levels did not differ significantly (P > 0.05). However, insulin (P < 0.001, partial η 2 = 0.247) and HOMA‐IR (P < 0.001, partial η 2 = 0.251) were significantly higher in the overweight/obese PCOS group than in all other groups, both demonstrating large effect sizes. The TyG index was also elevated in this group compared to the normal‐weight controls (P = 0.016, partial η 2 = 0.137), indicating a moderate‐to‐large effect size. Metabolic syndrome was present in five overweight/obese PCOS and four overweight/obese control participants. Table 1 summarizes metabolic and laboratory findings across all subgroups.
TABLE 1.
Comparison of metabolic syndrome parameters and laboratory findings among overweight/obese and normal‐weight PCOS and control groups.
| PCOS | Control | P | |||
|---|---|---|---|---|---|
| Overweight/obesea (n = 19) | Normal‐weightb (n = 14) | Overweight/obesec (n = 20) | Normal‐weightd (n = 23) | ||
| BMI | 29.9 ± 5.3 | 20.5 ± 2.1 | 29.0 ± 4.0 | 20.6 ± 2.0 |
F (3.72) = 37.872 P < 0.001a‐b,a‐d,b‐c,c‐d |
| Waist circumference (cm) | 90.2 ± 15.0 | 69.4 ± 3.7 | 85.2 ± 10.5 | 66.5 ± 4.8 |
F (3.72) = 28.010 P < 0.001a‐b,a‐d,b‐c,c‐d |
| Hip circumference (cm) | 109.9 ± 10.5 | 91.6 ± 3.8 | 107.3 ± 8.4 | 93.1 ± 5.6 |
F (3.72) = 28.339 P < 0.001a‐b,a‐d,b‐c,c‐d |
| Waist‐to‐hip ratio | 0.82 ± 0.07 | 0.76 ± 0.03 | 0.79 ± 0.05 | 0.71 ± 0.03 |
F (3.72) = 17.455 P < 0.001a‐b,a‐d,b‐c,b‐d,c‐d |
| SBP (mmHg) | 123.4 ± 6.5 | 110.0 ± 8.1 | 120.3 ± 6.2 | 113.0 ± 7.2 |
F (3.72) = 14.141 P < 0.001a‐b,a‐d,b‐c,c‐d |
| DBP (mmHg) | 81.8 ± 4.8 | 72.9 ± 6.1 | 78.8 ± 4.8 | 72.4 ± 5.2 |
F (3.72) = 15.113 P < 0.001a‐b,a‐d,b‐c,c‐d |
| ALT* (U/L) | 18 (9–56) | 12.5 (7–18) | 13 (7–26) | 12 (2–20) | P < 0.01 a‐b,a‐d |
| AST* (U/L) | 19 (14–35) | 15.5 (11–24) | 16 (9–25) | 15 (10–21) | P < 0.05a‐d |
| Triglyceride* (mg/dL) | 96 (50–178) | 73 (40–136) | 65.5 (46–163) | 63 (39–122) | P < 0.05a‐d |
| HDL‐C* (mg/dL) | 42.5 (36–125) | 52.5 (42–88) | 52.5 (36–69) | 53 (41–69) | P < 0.05a‐d |
| LDL‐C (mg/dL) | 97.6 ± 23.1 | 97.4 ± 24.9 | 99.6 ± 18 | 96.5 ± 19 |
F (3.71) = 0.081 P = 0.970 |
| Total cholesterol (mg/dL) | 161.1 ± 22.9 | 164.0 ± 32.8 | 160 ± 23.9 | 156.3 ± 23.8 |
F (3.71) = 0.290 P = 0.832 |
| Glucose (mg/dL) | 83.4 ± 7.2 | 82.3 ± 6.2 | 85.4 ± 6.4 | 82.6 ± 7.6 |
F (3.70) = 0.756 P = 0.522 |
| HbA1c (%) | 5.4 ± 0.3 | 5.3 ± 0.3 | 5.4 ± 0.3 | 5.3 ± 0.2 |
F (3.69) = 0.929 P = 0.432 |
| Insulin (μIU/mL) | 20.5 ± 16.1 | 8.2 ± 2.0 | 13.5 ± 5.1 | 8.9 ± 3.2 |
F (3.69) = 7.530 P < 0.001a‐b,a‐c,a‐d |
| HOMA‐IR | 4.2 ± 3.2 | 1.7 ± 0.4 | 2.9 ± 1.3 | 1.8 ± 0.7 |
F (3.69) = 7.723 P < 0.001a‐b,a‐c,a‐d |
| TyG index | 8.29 ± 0.42 | 8.02 ± 0.35 | 8.04 ± 0.40 | 7.90 ± 0.33 |
F (3.70) = 3.7020 P = 0.016a‐d |
| Total testosterone* (ng/dL) | 50.9 (24.5–108.1) | 45.0 (26.8–95.9) | 23.0 (12.5–41.3) | 24.4 (7.0–40.9) | P < 0.001a‐c,a‐d,b‐c,b‐d |
| SHBG (nmol/L) | 19.4 ± 10.5 | 32.4 ± 11.8 | 34.1 ± 18.5 | 53.2 ± 18.9 |
F (3.55) = 9.950 P < 0.001a‐d,b‐d,c‐d |
| DHEAS (μg/dL) | 281.9 ± 92.2 | 275.1 ± 124.8 | 247.4 ± 93.6 | 231.2 ± 125.2 |
F (3.72) = 0.928 P = 0.432 |
| FSH (mIU/mL) | 5.0 ± 1.9 | 4.9 ± 1.5 | 3.6 ± 1.6 | 3.4 ± 1.8 |
F (3.71) = 4.348 P < 0.01a‐c,a‐d,b‐c,b‐d |
| LH* (mIU/mL) | 9.1 (2.5–19.8) | 10.7 (0.8–21.1) | 3.7 (1.6–8.2) | 3.6 (1.2–11.9) | P < 0.001a‐c,a‐d,b‐c,b‐d |
| LH/FSH ratio* | 1.8 (0.6–5.5) | 2.0 (0.4–5.8) | 1.0 (0.4–2.9) | 1.2 (2–7.4) | P = 0.092 |
| E2* (pg/mL) | 51.3 (33–194.1) | 63 (32–333) | 65.0 (11.8–212.8) | 60.3 (24.5–255) | P = 0.779 |
| Prolactin (ng/mL) | 14.3 ± 7.0 | 12.4 ± 3.2 | 10.6 ± 4.3 | 11.1 ± 4.7 |
F (3.71) = 2.114 P = 0.106 |
| TSH* (μIU/mL) | 1.8 (0.5–5.7) | 1.8 (0.8–3.7) | 2.1 (1.1–6.8) | 2.1 (0.6–5.6) | P = 0.711 |
| fT4* (ng/dL) | 11.2 (1.2–13.7) | 10.5 (8.6–13.4) | 10.9 (1.3–13.1) | 11.0 (8.6–15.5) | P = 0.619 |
| 17‐OHP (ng/dL) | 104.2 ± 48.5 | 105.7 ± 49.0 | 85.6 ± 49.1 | 96.4 ± 42.9 |
F (3.69) = 0.692 P = 0.560 |
Note: BMI, calculated as weight in kilograms divided by the square of height in meters. Lowercase superscript letters (a–d) indicate the subgroups (a: overweight/obese PCOS; b: normal‐weight PCOS; c: overweight/obese control; d: normal‐weight control). Significant pairwise differences between groups are indicated accordingly. Data are presented as mean ± SD unless otherwise specified. Variables marked with * are presented as median (min–max)
Abbreviations: 17‐OHP, 17‐hydroxyprogesterone; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; DBP, diastolic blood pressure; DHEAS, dehydroepiandrosterone sulfate; E2, estradiol; FSH, follicle‐stimulating hormone; fT4, free T4; HbA1c, hemoglobin A1c; HDL‐C, high‐density lipoprotein cholesterol; HOMA‐IR, homeostasis model assessment of insulin resistance; LDL‐C, low‐density lipoprotein cholesterol; LH, luteinizing hormone; PCOS, polycystic ovary syndrome; SBP, systolic blood pressure; SD, standard deviation; SHBG, sex hormone‐binding globulin; TSH, thyroid‐stimulating hormone; TyG, triglyceride/glucose.
When the overall PCOS and control groups were compared (without BMI stratification), no significant differences were observed in waist or hip circumferences or BP (P > 0.05). However, the PCOS group had a higher waist‐to‐hip ratio (0.79 ± 0.06 vs 0.75 ± 0.06, P < 0.01). Median ALT, AST, and triglyceride levels were also significantly higher in the PCOS group than in controls (P < 0.05 for all). HDL‐C, LDL‐C, total cholesterol, fasting glucose, HbA1c, insulin, and HOMA‐IR did not differ significantly (P > 0.05), but the TyG index was significantly higher in the PCOS group compared to controls (8.17 ± 0.41 vs 7.97 ± 0.37, P < 0.05). See Table S1 for details.
Conventional M‐mode ECHO parameters, including interventricular septal thickness in systole (IVSTs), LV end‐diastolic diameter (LVEDD), posterior wall thickness in systole (PWTs), LV mass, left atrial (LA) end‐systolic diameter, and LA/Ao ratio, differed between overweight/obese and normal‐weight subgroups but did not differ according to PCOS status, indicating that these findings were attributable to adiposity rather than PCOS. Strain ECHO and 24‐h Holter parameters showed no significant differences among subgroups (P > 0.05 for all). Detailed comparisons are presented in Tables 2 and 3.
TABLE 2.
Comparison of conventional and M‐mode echocardiographic findings among overweight/obese and normal‐weight PCOS and control groups.
| PCOS | Control | P | |||
|---|---|---|---|---|---|
| Overweight/obesea (n = 19) | Normal‐weightb (n = 14) | Overweight/obesec (n = 20) | Normal‐weightd (n = 23) | ||
| IVSTd* (mm) | 8.4 (5.0–12.7) | 7.7 (6.2–8.8) | 7.7 (5.0–14.2) | 7.7 (4.2–11.1) | P = 0.126 |
| IVSTs (mm) | 11.19 ± 1.76 | 8.99 ± 1.17 | 11.49 ± 2.12 | 9.78 ± 2.46 |
F (3.61) = 5.463 P < 0.01a‐b,a‐d,b‐c,c‐d |
| LVEDD (mm) | 46.55 ± 2.44 | 42.96 ± 2.76 | 47.19 ± 3.37 | 43.24 ± 3.69 |
F (3.64) = 8.169 P < 0.001a‐b,a‐d,b‐c,c‐d |
| LVESD (mm) | 28.04 ± 3.84 | 26.88 ± 3.27 | 28.07 ± 2.39 | 26.33 ± 3.55 |
F (3.64) = 1.262 P = 0.295 |
| PWTd* (mm) | 7.70 (5.60–11.50) | 7.70 (6.20–10.40) | 7.70 (6.20–10.00) | 7.65 (4.60–9.20) | P = 0.060 |
| PWTs* (mm) | 14.6 (11.1–18.8) | 12.9 (7.7–15.7) | 13.5 (11.9–23.0) | 13.0 (10.4–17.7) | P = 0.014a‐b,a‐d |
| EF (%) | 70.07 ± 8.15 | 67.36 ± 7.60 | 70.95 ± 4.67 | 69.05 ± 8.91 |
F (3.64) = 0.672 P = 0.572 |
| SF (%) | 39.87 ± 6.55 | 37.57 ± 5.89 | 40.37 ± 3.92 | 39.05 ± 7.41 |
F (3.64) = 0.626 P = 0.601 |
| LV mass* (g/m2) | 124 (79–200) | 99.5 (66–137) | 119 (93–181) | 95 (54–134) |
F (3.63) = 9.762 P < 0.001a‐b,a‐d,b‐c,c‐d |
| LV mass z‐score | −1.41 ± 1.02 | −1.63 ± 0.72 | −1.66 ± 0.71 | −1.73 ± 1.09 |
F (3.63) = 0.357 P = 0.784 |
| Peak E wave velocity (m/s) | 0.92 ± 0.17 | 0.94 ± 0.14 | 0.99 ± 0.21 | 0.90 ± 0.15 |
F (3.64) = 0.889 P = 0.452 |
| Peak E wave velocity z‐score | −0.09 ± 0.96 | 0.00 ± 0.76 | 0.29 ± 1.20 | −0.21 ± 0.87 |
F (3.64) = 0.913 P = 0.440 |
| Peak A wave velocity (m/s) | 0.58 ± 0.11 | 0.56 ± 0.11 | 0.64 ± 0.12 | 0.58 ± 0.10 |
F (3.62) = 1.752 P = 0.166 |
| Peak A wave velocity z‐score | 1.58 ± 1.28 | 1.34 ± 1.26 | 2.22 ± 1.39 | 1.54 ± 1.15 |
F (3.62) = 1.565 P = 0.207 |
| Peak E/A ratio* | 1.51 (1.35–2.19) | 1.68 (0.14–2.32) | 1.51 (1.27–1.90) | 1.50 (1.30–2.88) | P = 0.652 |
| Peak E/A ratio z‐score* | −1.26 (−1.45 to −0.15) | −1.01 (−1.72 to 0.02) | −1.25 (−1.58 to −0.67) | −1.29 (−1.64 to 0.92) | P = 0.459 |
| LA end‐systolic diameter (mm) | 33.27 ± 3.61 | 29.64 ± 2.79 | 31.76 ± 3.42 | 27.52 ± 3.33 |
F (3.63) = 10.200 P < 0.001a‐b,a‐d,c‐d |
| Aortic root (mm) | 24.47 ± 3.23 | 23.50 ± 2.41 | 24.78 ± 2.53 | 23.00 ± 2.39 |
F (3.64) = 1.821 P = 0.152 |
| Aortic root z‐score | −1.32 ± 0.96 | −0.99 ± 1.02 | −1.24 ± 0.78 | −1.14 ± 0.81 |
F (3.62) = 0.350 P = 0.790 |
| LA/Ao ratio | 1.37 ± 0.18 | 1.27 ± 0.16 | 1.28 ± 0.13 | 1.21 ± 0.17 |
F (3.63) = 2.973 P = 0.038a‐d |
| MAPSE* (mm) | 17.00 (16.00–31.00) | 15.5 (12–19) | 18 (12–23) | 17 (14–22) | P = 0.262 |
| TAPSE (mm) | 25.01 ± 4.12 | 24.97 ± 4.48 | 25.57 ± 3.14 | 23.59 ± 3.20 |
F (3.66) = 1.052 P = 0.376 |
Note: Lowercase superscript letters (a–d) indicate the subgroups (a: overweight/obese PCOS; b: normal‐weight PCOS; c: overweight/obese control; d: normal‐weight control). Significant pairwise differences between groups are indicated accordingly. Data are presented as mean ± SD unless otherwise specified. Variables marked with * are presented as median (min–max).
Abbreviations: Ao, aorta; EF, ejection fraction; IVSTd, interventricular septal thickness in diastole; IVSTs, interventricular septal thickness in systole; LA, left atrial; LV, left ventricular; LVEDD, left ventricular end‐diastolic diameter; LVESD, left ventricular end‐systolic diameter; MAPSE, mitral annular plane systolic excursion; PCOS, polycystic ovary syndrome; PWTd, posterior wall thickness in diastole; PWTs, posterior wall thickness in systole; SD, standard deviation; SF, shortening fraction; TAPSE, tricuspid annular plane systolic excursion.
TABLE 3.
Comparison of tissue Doppler imaging, strain echocardiography, and 24‐h Holter monitoring data among overweight/obese and normal‐weight PCOS and control groups.
| PCOS | Control | P | |||
|---|---|---|---|---|---|
| Overweight/obesea (n = 19) | Normal‐weightb (n = 14) | Overweight/obesec (n = 20) | Normal‐weightd (n = 23) | ||
| Tissue Doppler imaging | |||||
| E' wave (m/s) | 0.12 ± 0.02 | 0.15 ± 0.03 | 0.13 ± 0.03 | 0.13 ± 0.02 |
F (3.67) = 2.438 P = 0.072 |
| A' wave* (m/s) | 0.06 (0.05–0.10) | 0.06 (0.01–0.14) | 0.07 (0.04–0.11) | 0.07 (0.05–0.11) | P = 0.844 |
| S′ wave (m/s) | 12.56 ± 2.54 | 13.47 ± 2.79 | 13.34 ± 2.38 | 13.28 ± 2.34 |
F (3.66) = 0.426 P = 0.735 |
| IVRT (ms) | 55.06 ± 14.86 | 51.57 ± 11.11 | 53.65 ± 8.65 | 52.76 ± 10.44 |
F (3.68) = 0.267 P = 0.849 |
| IVCT (ms) | 52.59 ± 9.57 | 55.07 ± 8.43 | 53.05 ± 9.11 | 52.24 ± 7.40 |
F (3.68) = 0.334 P = 0.801 |
| ET (ms) | 367.13 ± 91.21 | 291.36 ± 42.68 | 330.50 ± 72.40 | 369.14 ± 73.43 |
F (3.67) = 4.023 P = 0.011a‐b,b‐d |
| MPI | 0.31 ± 0.07 | 0.37 ± 0.07 | 0.34 ± 0.09 | 0.30 ± 0.07 |
F (3.67) = 3.139 P = 0.031a‐b,b‐d |
| E/E' ratio* | 6.90 (1.02–9.50) | 6.15 (4.10–9.60) | 7.85 (5.20–9.80) | 6.80 (1.00–9.00) | P = 0.130 |
| Strain echocardiography | |||||
| LVGLS A2C (%) | −17.51 ± 3.14 | −19.86 ± 3.23 | −19.29 ± 3.49 | −19.92 ± 4.60 |
F (3.65) = 1.480 P = 0.228 |
| LVGLS A3C (%) | −17.18 ± 3.52 | −18.44 ± 3.31 | −18.93 ± 3.61 | −17.40 ± 3.79 |
F (3.62) = 0.926 P = 0.433 |
| LVGLS A4C (%) | −17.88 ± 3.91 | −18.45 ± 2.37 | −20.36 ± 3.96 | −20.13 ± 3.65 |
F (3.67) = 2.079 P = 0.111 |
| LVGLS mean (%) | −17.75 ± 2.67 | −18.95 ± 2.10 | −19.69 ± 2.79 | −19.09 ± 3.14 |
F (3.62) = 1.434 P = 0.242 |
| 24‐h Holter monitoring | |||||
| SDNN (ms) | 151.9 ± 33.9 | 149.5 ± 23.0 | 143.0 ± 32.0 | 147.0 ± 35.7 |
F (3.69) = 0.254 P = 0.858 |
| SDNN index | 70.4 ± 15.6 | 70.6 ± 12.9 | 68.9 ± 17.6 | 66.7 ± 19.0 |
F (3.69) = 0.217 P = 0.884 |
| RMSSD | 45.6 ± 13.5 | 45.6 ± 12.4 | 45.8 ± 17.9 | 40.5 ± 14.0 |
F (3.69) = 0.653 P = 0.584 |
| Triangular index | 42.4 ± 12.0 | 42.8 ± 9.7 | 40.9 ± 9.7 | 43.2 ± 12.2 |
F (3.69) = 0.162 P = 0.922 |
Note: Lowercase superscript letters (a–d) indicate the subgroups (a: overweight/obese PCOS; b: normal‐weight PCOS; c: overweight/obese control; d: normal‐weight control). Significant pairwise differences between groups are indicated accordingly. Data are presented as mean ± SD unless otherwise specified. Variables marked with * are presented as median (min–max).
Abbreviations: A2C, apical two‐chamber; A3C, apical three‐chamber; A4C, apical four‐chamber; ET, ejection time; IVCT, isovolumetric contraction time; IVRT, isovolumetric relaxation time; LVGLS, left ventricular global peak longitudinal strain; MPI, myocardial performance index; NN, artifact‐free interbeat intervals; PCOS, polycystic ovary syndrome; RMSSD, root mean square of successive differences; SD, standard deviation; SDNN, standard deviation of NN intervals.
When the overall PCOS and control groups were compared (without BMI stratification), LA end‐systolic diameter (31.52 ± 3.68 vs 29.42 ± 3.95 mm, P = 0.030) and LA/Ao ratio (1.32 ± 0.18 vs 1.24 ± 0.15, P < 0.05) were significantly higher in the PCOS group than in controls. All other conventional and M‐mode ECHO parameters were within normal limits, with no significant differences between the PCOS and control groups (P > 0.05).
TDI parameters did not differ significantly between PCOS and control groups (P > 0.05). However, LV global peak longitudinal strain (LVGLS) apical four‐chamber (A4C) was significantly lower in the PCOS group (−18.1 ± 3.3 vs −20.2 ± 3.8, P = 0.016). HRV parameters showed no significant differences between the PCOS and control groups (P > 0.05). See Tables S2 and S3 for further ECHO and Holter data.
To assess the independent effects of PCOS and obesity, a two‐way ANOVA was conducted using three BMI categories: normal‐weight, overweight, and obesity. BMI had a similar impact across groups for most variables, except LVGLS apical three‐chamber (A3C) (P < 0.05). A significant interaction between group and BMI was found for LVGLS A3C (F [2, 70] = 3.522, P < 0.05, partial η 2 = 0.105), indicating a moderate effect size and supporting the practical relevance of the observed interaction. While no differences were seen in the normal‐weight and overweight groups, adolescents with obesity and PCOS had significantly lower LVGLS A3C than obese controls (P < 0.01). The results are shown in Table 4.
TABLE 4.
Two‐way ANOVA test results.
| PCOS | Control | F | P | η 2 | |
|---|---|---|---|---|---|
| Mean ± SD | Mean ± SD | ||||
| WHR | |||||
| Normal‐weighta | 0.76 ± 0.03 | 0.71 ± 0.03 | F 1 = 7.966 | P 1 < 0.01 | η 2 1 = 0.102 |
| Overweightb | 0.79 ± 0.05 | 0.77 ± 0.05 | F 2 = 28.294 | P 2 < 0.001a‐b,a‐c,b‐c | η 2 2 = 0.447 |
| Obesec | 0.85 ± 0.08 | 0.82 ± 0.04 | F 3 = 0.402 | P 3 > 0.05 | η 2 3 = 0.011 |
| SBP | |||||
| Normal‐weighta | 110.0 ± 8.1 | 113.0 ± 7.2 | F 1 = 0.368 | P 1 > 0.05 | η 2 1 = 0.005 |
| Overweightb | 119.4 ± 4.6 | 117.5 ± 5.4 | F 2 = 27.221 | P 2 < 0.001a‐b,a‐c,b‐c | η 2 2 = 0.437 |
| Obesec | 127.0 ± 5.9 | 123.0 ± 5.9 | F 3 = 2.050 | P 3 > 0.05 | η 2 3 = 0.055 |
| DBP | |||||
| Normal‐weighta | 72.9 ± 6.1 | 72.4 ± 5.2 | F 1 = 3.256 | P 1 > 0.05 | η 2 1 = 0.044 |
| Overweightb | 78.3 ± 3.5 | 76.5 ± 5.3 | F 2 = 29.821 | P 2 < 0.001a‐b,a‐c,b‐c | η 2 2 = 0.460 |
| Obesec | 85.0 ± 3.3 | 81.0 ± 3.2 | F 3 = 0.856 | P 3 > 0.05 | η 2 3 = 0.024 |
| HOMA‐IR | |||||
| Normal‐weighta | 1.7 ± 0.4 | 1.8 ± 0.7 | F 1 = 4.121 | P 1 < 0.05 | η 2 1 = 0.058 |
| Overweightb | 3.0 ± 1.2 | 2.3 ± 0.5 | F 2 = 16.580 | P 2 < 0.001a‐c,b‐c | η 2 2 = 0.331 |
| Obesec | 5.4 ± 4.1 | 3.5 ± 1.5 | F 3 = 2.263 | P 3 > 0.05 | η 2 3 = 0.063 |
| TyG index | |||||
| Normal‐weighta | 8.0 ± 0.4 | 7.9 ± 0.3 | F 1 = 5.004 | P 1 < 0.05 | η 2 1 = 0.069 |
| Overweightb | 8.2 ± 0.4 | 8.0 ± 0.4 | F 2 = 4.152 | P 2 < 0.05a‐c | η 2 2 = 0.109 |
| Obesec | 8.4 ± 0.4 | 8.1 ± 0.4 | F 3 = 0.342 | P 3 > 0.05 | η 2 3 = 0.010 |
| HDL‐C | |||||
| Normal‐weighta | 55.9 ± 11.5 | 53.7 ± 7.7 | F 1 = 0.046 | P 1 > 0.05 | η 2 1 = 0.001 |
| Overweightb | 45.4 ± 11.2 | 54.2 ± 6.5 | F 2 = 0.974 | P 2 > 0.05 | η 2 2 = 0.027 |
| Obesec | 56.2 ± 26.7 | 51.7 ± 11.2 | F 3 = 1.546 | P 3 > 0.05 | η 2 3 = 0.043 |
| Triglyceride* (mg/dL) | |||||
| Normal‐weighta | 77.9 ± 26.6 | 68.8 ± 22.5 | F 1 = 6.680 | P 1 < 0.05 | η 2 1 = 0.088 |
| Overweightb | 91.7 ± 42.4 | 76.0 ± 34.3 | F 2 = 3.873 | P 2 < 0.05 | η 2 2 = 0.101 |
| Obesec | 116.0 ± 41.1 | 80.8 ± 33.8 | F 3 = 1.057 | P 3 > 0.05 | η 2 3 = 0.030 |
| LV mass | |||||
| Normal‐weighta | 99.1 ± 16.3 | 95.6 ± 18.2 | F 1 = 3.084 | P 1 > 0.05 | η 2 1 = 0.048 |
| Overweightb | 119.8 ± 36.2 | 118.0 ± 15.6 | F 2 = 19.569 | P 2 < 0.001 | η 2 2 = 0.391 |
| Obesec | 155.5 ± 31.7 | 128.5 ± 28.4 | F 3 = 1.541 | P 3 > 0.05 | η 2 3 = 0.048 |
| LA end‐systolic diameter (mm) | |||||
| Normal‐weighta | 29.6 ± 2.8 | 27.5 ± 3.3 | F 1 = 5.166 | P 1 < 0.05 | η 2 1 = 0.078 |
| Overweightb | 31.8 ± 3.9 | 31.8 ± 2.4 | F 2 = 13.874 | P 2 < 0.001 | η 2 2 = 0.313 |
| Obesec | 35.5 ± 1.5 | 31.8 ± 4.3 | F 3 = 1.291 | P 3 > 0.05 | η 2 3 = 0.041 |
| LVGLS A2C | |||||
| Normal‐weighta | −19.9 ± 3.2 | −19.9 ± 4.6 | F 1 = 1.626 | P 1 > 0.05 | η 2 1 = 0.025 |
| Overweightb | −18.4 ± 3.6 | −20.3 ± 2.6 | F 2 = 2.385 | P 2 > 0.05 | η 2 2 = 0.070 |
| Obesec | −16.7 ± 2.6 | −18.3 ± 4.1 | F 3 = 0.452 | P 3 > 0.05 | η 2 3 = 0.014 |
| LVGLS A3C | |||||
| Normal‐weighta | −18.4 ± 3.3 | −17.4 ± 3.8 | F 1 = 0.702 | P 1 > 0.05 | η 2 1 = 0.012 |
| Overweightb | −19.1 ± 2.6 | −18.2 ± 4.6 | F 2 = 0.428 | P 2 > 0.05 | η 2 2 = 0.014 |
| Obesec | −15.5 ± 3.5 | −19.7 ± 2.0 | F 3 = 3.522 | P 3 < 0.05* | η 2 3 = 0.105 |
| LVGLS A4C | |||||
| Normal‐weighta | −18.5 ± 2.4 | −20.1 ± 3.7 | F 1 = 6.540 | P 1 < 0.05 | η 2 1 = 0.091 |
| Overweightb | −19.0 ± 4.3 | −21.4 ± 3.6 | F 2 = 1.822 | P 2 > 0.05 | η 2 2 = 0.053 |
| Obesec | −16.6 ± 3.2 | −19.4 ± 4.2 | F 3 = 0.148 | P 3 > 0.05 | η 2 3 = 0.005 |
Note: BMI, calculated as weight in kilograms divided by the square of height in meters. F 1, P 1,η 2 1: F, P, η 2 values for the comparison between PCOS and control groups. F 2, P 2, η 2 2: F, P, η 2 values for the comparison between normal weight‐overweight‐obese groups. F 3, P 3,η 2 3: F, P, η 2 values for the interaction effect between group (PCOS/control) and BMI. Lowercase superscript letters (a–d) indicate the subgroups (a: overweight/obese PCOS; b: normal‐weight PCOS; c: overweight/obese control; d: normal‐weight control). Significant pairwise differences between groups are indicated accordingly.
Abbreviations: A2C, apical two‐chamber; A3C, apical three‐chamber; A4C, apical four‐chamber; ANOVA, analysis of variance; DBP, diastolic blood pressure; HDL‐C, high‐density lipoprotein cholesterol; HOMA‐IR, homeostasis model assessment of insulin resistance; LA, left atrial; LV, left ventricular; LVGLS, left ventricular global peak longitudinal strain; PCOS, polycystic ovary syndrome; SBP, systolic blood pressure; TyG, triglyceride/glucose; WHR, waist‐to‐hip ratio.
No significant difference was observed between the PCOS and control groups in the normal and overweight categories (P > 0.05). However, a significant difference was found in the obese group (P < 0.01), with lower values in the PCOS group.
4. DISCUSSION
PCOS in adolescents is increasingly recognized as a condition with significant metabolic and CV implications beyond reproductive abnormalities. In our study, the LA end‐systolic diameter and LA/Ao ratio were significantly higher in the PCOS group than in the controls. In the subgroup analysis, LA end‐systolic diameter did not differ between the normal‐weight PCOS and overweight/obese control groups, suggesting that PCOS alone may be associated with a similar increase in LA end‐systolic diameter as obesity. However, when obesity and PCOS coexisted, the LA/Ao ratio was significantly higher. Additionally, LA diameter has been found to be associated with all‐cause mortality in adults 19 and has been reported to be increased in adults with PCOS independent of obesity. 20 In the current study, although LA diameters remained within normal limits, the higher values in the PCOS group at diagnosis may indicate early or subclinical cardiac changes rather than overt diastolic dysfunction.
LV mass was significantly higher in overweight/obese groups compared to normal‐weight groups. Additionally, although not statistically significant, normal‐weight PCOS patients had a higher LV mass than normal‐weight controls. Patel et al. 21 found no difference in BMI‐adjusted LV mass index between obese PCOS and obese controls. However, a meta‐analysis primarily involving adult studies concluded that PCOS is associated with increased LV mass even after adjusting for obesity. 20 When considering both our study and existing literature, any observed increase in LV mass in adolescents with PCOS appears to be largely mediated by obesity, with no independent effect of PCOS demonstrated. However, in adults, PCOS alone is considered an independent risk factor for increased LV mass, independent of obesity. Thus, while LV mass may not reliably indicate early cardiac changes in adolescents with PCOS, it remains a complication that should be monitored over time.
IVSTs was higher in overweight/obese groups regardless of PCOS status, while remaining within normal limits. PWTs was higher in overweight/obese PCOS compared with normal‐weight PCOS and controls, with no significant difference between overweight/obese PCOS and BMI‐matched controls, indicating a weight‐driven rather than PCOS‐specific effect. One study reported similar IVST but increased PWT in adolescents with PCOS. 18 A meta‐analysis including both adolescents and adults found increased IVST and PWT in diastole in PCOS, even after excluding obese individuals. 20 These findings suggest that LV hypertrophy in adolescents with PCOS primarily affects the posterior wall, with possible septal involvement over time. An increase in IVST and PWT at end diastole should be considered a warning sign for LV hypertrophy and dysfunction. Our study showed that while all systolic and diastolic values were within normal limits, diastolic values were highest in the overweight/obese PCOS group, only systolic differences reached statistical significance.
Lower strain measurements on ECHO indicate subclinical impairment of LV myocardial function. In our study, LVGLS A4C was significantly lower in the PCOS group, while other strain values showed a nonsignificant trend in the same direction. Subgroup analyses revealed the lowest strain values in obese PCOS patients, though not statistically significant. Patel et al. 21 compared GLS and global circumferential strain between obese PCOS and obese control adolescents and found no significant differences, while adult studies found that while conventional ECHO parameters did not differ between groups, GLS was lower in the PCOS group. 22 Demirelli et al. 23 similarly demonstrated reduced strain values in young adult PCOS patients. Although the average strain values remained within normal reference ranges and BMI‐stratified subgroup analyses did not reveal statistically significant differences, the lower LVGLS A4C observed in the overall PCOS group may reflect subtle functional changes not captured by conventional ECHO. We believe that strain ECHO is valuable in detecting early‐stage functional impairment in PCOS patients. However, given that cardiac abnormalities tend to emerge with aging, its use may be prioritized in patients with a high CV risk profile during follow‐up.
A meta‐analysis of 30 studies, including three adolescent cohorts, found that PCOS is associated with preclinical LV diastolic dysfunction without affecting systolic function. 20 Early diastolic dysfunction is a predictor of heart disease and mortality, supporting the importance of early CV risk detection in PCOS patients. 24 , 25 Intervening during adolescence offers an opportunity to mitigate long‐term complications.
No significant differences in HRV parameters were observed between PCOS and control groups. In the only other adolescent study, frequency‐domain analysis from 10‐min ECG recordings demonstrated an increased low‐frequency/high‐frequency ratio in PCOS, suggesting altered autonomic balance. 26 Another study including adolescents and young adults reported lower very low frequency values in PCOS, without differences in other HRV indices. 27 In adults, a meta‐analysis showed reduced HRV in PCOS overall, with alterations evident in normal‐weight and overweight individuals but not in those with obesity. 28 Collectively, these findings suggest that autonomic dysfunction in PCOS may emerge progressively and in parallel with increasing cardiometabolic risk. To our knowledge, this is the first study evaluating time‐domain HRV parameters in adolescents with PCOS, underscoring the need for larger studies in this population.
In our study, since patients were evaluated at the time of PCOS diagnosis, the number of patients meeting the criteria for metabolic syndrome was low; however, the presence of metabolic abnormalities during adolescence was evident. A study comparing adolescents and adults with PCOS found lower BP, LDL‐C, and total cholesterol in adolescents, but similar rates of glucose intolerance and insulin resistance. 29 Another study showed that the prevalence of metabolic syndrome in adolescents varies based on the diagnostic criteria used, with the Pediatric Endocrine Society criteria identifying the highest prevalence. 30 In our study, insulin levels and HOMA‐IR were significantly higher in overweight/obese adolescents with PCOS than in all other groups, with HOMA‐IR exceeding established cutoff values only in this subgroup, suggesting that clinically relevant insulin resistance emerges primarily in the context of obesity. However, the components of metabolic syndrome become more pronounced with age. In this context, early diagnosis can play a crucial role in preventing and effectively managing comorbidities.
A meta‐analysis reported that adolescents with PCOS were over three times more likely to have metabolic syndrome compared to controls, especially in the presence of obesity. 31 Our data confirmed that overweight/obese PCOS patients had higher waist‐to‐hip ratios, BP, triglycerides, insulin, and HOMA‐IR, and lower HDL‐C than other groups. However, differences between obese PCOS and obese controls did not reach statistical significance, possibly due to small sample size or shared pathophysiological mechanisms. Obesity and PCOS often coexist, and both are characterized by insulin resistance and hyperandrogenism, which may exacerbate one another. PCOS in adolescence has even been described as a “postpubertal central obesity syndrome.” Their overlapping metabolic profiles suggest similar comorbidities and CV consequences, even if these are not yet clinically apparent. Therefore, early detection of metabolic and CV risks is essential to prevent long‐term complications.
Limitations of our study include the relatively small sample size and under‐recruitment of PCOS subgroups, which may have reduced statistical power and contributed to several non‐significant findings. Recruitment was further limited by the COVID‐19 pandemic and participants' reluctance to undergo 24‐h Holter monitoring. The cross‐sectional design precludes causal inference and limits the assessment of longitudinal changes. Obesity may also have acted as a confounding factor, as BMI showed stronger effects than PCOS for several parameters in the two‐way ANOVA analyses. Although differences in LA dimensions and strain parameters were observed, all measurements remained within normal pediatric reference ranges, rendering their clinical relevance uncertain. Finally, echocardiographic assessments were performed by a single operator, which reduces interobserver variability but may limit generalizability. Despite these limitations, assessing treatment‐naïve adolescents and the use of strain ECHO and HRV strengthen the methodological scope beyond conventional parameters.
In conclusion, we demonstrated certain changes in selected cardiometabolic markers in adolescents with PCOS who were overweight or obese at diagnosis. Larger prospective and longitudinal studies are needed to clarify the distinct effects of PCOS and obesity. In those with both conditions, weight management targeting insulin resistance should be prioritized. Early cardiac evaluation in adolescents with PCOS should primarily focus on assessing modifiable CV risk factors—such as BP, lipid profile, fasting glucose, and insulin resistance—rather than routine cardiac imaging or Holter monitoring. Advanced cardiac investigations may be reserved for patients with clinical indications or abnormal findings on initial assessment.
AUTHOR CONTRIBUTIONS
DAA: Conceptualization, methodology, data collection, writing the original draft, clinical interpretation, funding acquisition and final approval of the manuscript. MPK: Conceptualization, methodology, statistical analysis, review and editing, clinical interpretation and final approval of the manuscript. MÖ: Data collection, statistical analysis and final approval of the manuscript. İE: Conceptualization, methodology, data collection, review and editing, clinical interpretation, final approval of the manuscript. HHA: Data collection, review and editing, clinical interpretation, final approval of the manuscript. TK: Review and editing, clinical interpretation, final approval of the manuscript. NK: Conceptualization, methodology, review and editing, clinical interpretation, final approval of the manuscript. OD: Review and editing, clinical interpretation, final approval of the manuscript. SA: Conceptualization, methodology, review and editing, clinical interpretation, funding acquisition, final approval of the manuscript.
FUNDING INFORMATION
This study was supported by Hacettepe University Scientific Research Projects Coordination Unit (project no.: THD‐2021‐19437). The funder had no role in the study design; data collection, analysis, or interpretation; manuscript writing; or the decision to submit the article for publication.
CONFLICT OF INTEREST STATEMENT
All authors declare that they have no actual or perceived conflicts of interest related to this work.
Supporting information
Table S1. Comparison of anthropometric data, metabolic syndrome parameters and laboratory findings between PCOS and control groups.
Table S2. Comparison of conventional and M‐mode echocardiographic findings between PCOS and control groups.
Table S3. Comparison of tissue Doppler imaging, strain echocardiography, and 24‐h Holter monitoring data between PCOS and control groups.
ACKNOWLEDGMENTS
The authors would like to thank Dr. Ayşe Gül Güven for assistance with data collection.
Aygün Ari D, Pehlivantürk Kizilkan M, Öztürk M, et al. Subclinical cardiac alterations and cardiometabolic profile in adolescents with newly diagnosed polycystic ovary syndrome: A cross‐sectional comparative echocardiographic study. Int J Gynecol Obstet. 2026;175:310‐321. doi: 10.1002/ijgo.71017
The abstract of this study was published in the Journal of Adolescent Health as part of the proceedings of the SAHM 2025 Annual Meeting (Poster no. 209), held March 4–7, 2025, in Baltimore, Maryland, USA.
DATA AVAILABILITY STATEMENT
Research data are not shared.
REFERENCES
- 1. Ibáñez L, Oberfield SE, Witchel S, et al. An International Consortium Update: pathophysiology, diagnosis, and treatment of polycystic ovarian syndrome in adolescence. Horm Res Paediatr. 2017;88(6):371‐395. [DOI] [PubMed] [Google Scholar]
- 2. Kelsey MM, Zeitler PS. Insulin resistance of puberty. Curr Diab Rep. 2016;16(7):64. [DOI] [PubMed] [Google Scholar]
- 3. Allen LA, Shrikrishnapalasuriyar N, Rees DA. Long‐term health outcomes in young women with polycystic ovary syndrome: a narrative review. Clin Endocrinol (Oxf). 2022;97(2):187‐198. [DOI] [PubMed] [Google Scholar]
- 4. Alvarez YR, Pico M, Ashokprabhu N, et al. Polycystic ovarian syndrome: a risk factor for cardiovascular disease. Curr Atheroscler Rep. 2023;25(12):1003‐1011. [DOI] [PubMed] [Google Scholar]
- 5. Lorenz MW, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction of clinical cardiovascular events with carotid intima‐media thickness: a systematic review and meta‐analysis. Circulation. 2007;115(4):459‐467. [DOI] [PubMed] [Google Scholar]
- 6. Okoth K, Chandan JS, Marshall T, et al. Association between the reproductive health of young women and cardiovascular disease in later life: umbrella review. BMJ. 2020;371:m3502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Wekker V, van Dammen L, Koning A, et al. Long‐term cardiometabolic disease risk in women with PCOS: a systematic review and meta‐analysis. Hum Reprod Update. 2020;26(6):942‐960. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Yarali H, Yildirir A, Aybar F, et al. Diastolic dysfunction and increased serum homocysteine concentrations may contribute to increased cardiovascular risk in patients with polycystic ovary syndrome. Fertil Steril. 2001;76(3):511‐516. [DOI] [PubMed] [Google Scholar]
- 9. Jacewicz‐Święcka M, Kowalska I. Polycystic ovary syndrome and the risk of cardiometabolic complications in longitudinal studies. Diabetes Metab Res Rev. 2018;34(8):e3054. [DOI] [PubMed] [Google Scholar]
- 10. Teede HJ, Tay CT, Laven JJE, et al. Recommendations from the 2023 international evidence‐based guideline for the assessment and management of polycystic ovary syndrome. Eur J Endocrinol. 2023;189(2):G43‐G64. [DOI] [PubMed] [Google Scholar]
- 11. Witchel SF, Oberfield S, Rosenfield RL, et al. The diagnosis of polycystic ovary syndrome during adolescence. Horm Res Paediatr. 2015;83(6):376‐389. [DOI] [PubMed] [Google Scholar]
- 12. Hampl SE, Hassink SG, Skinner AC, et al. Clinical practice guideline for the evaluation and treatment of children and adolescents with obesity. Pediatrics. 2023;151(2):e2022060640. [DOI] [PubMed] [Google Scholar]
- 13. Yildiz BO, Bolour S, Woods K, Moore A, Azziz R. Visually scoring hirsutism. Hum Reprod Update. 2010;16(1):51‐64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Martin KA, Anderson RR, Chang RJ, et al. Evaluation and treatment of hirsutism in premenopausal women: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2018;103(4):1233‐1257. [DOI] [PubMed] [Google Scholar]
- 15. Kurtoğlu S, Hatipoğlu N, Mazıcıoğlu M, Kendirci M, Keskin M, Kondolot M. Insulin resistance in obese children and adolescents: HOMA‐IR cut‐off levels in the prepubertal and pubertal periods. J Clin Res Pediatr Endocrinol. 2010;2(3):100‐106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Moon S, Park JS, Ahn Y. The cut‐off values of triglycerides and glucose index for metabolic syndrome in American and Korean adolescents. J Korean Med Sci. 2017;32(3):427‐433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zimmet P, Alberti KG, Kaufman F, et al. The metabolic syndrome in children and adolescents ‐ an IDF consensus report. Pediatr Diabetes. 2007;8(5):299‐306. [DOI] [PubMed] [Google Scholar]
- 18. Çetin M, Tunçdemir P, Karaman K, et al. Cardiovascular evaluation and serum paraoxonase‐1 levels in adolescents with polycystic ovary syndrome. J Obstet Gynaecol. 2020;40(1):90‐95. [DOI] [PubMed] [Google Scholar]
- 19. Bouzas‐Mosquera A, Broullón FJ, Álvarez‐García N, et al. Left atrial size and risk for all‐cause mortality and ischemic stroke. CMAJ. 2011;183(10):E657‐E664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Mirzohreh ST, Panahi P, Zafardoust H, et al. The role of polycystic ovary syndrome in preclinical left ventricular diastolic dysfunction: an echocardiographic approach: a systematic review and meta‐analysis. Cardiovasc Endocrinol Metab. 2023;12(4):e0294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Patel SS, Truong U, King M, et al. Obese adolescents with polycystic ovarian syndrome have elevated cardiovascular disease risk markers. Vasc Med. 2017;22(2):85‐95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Erdoğan E, Akkaya M, Bacaksız A, et al. Subclinical left ventricular dysfunction in women with polycystic ovary syndrome: an observational study. Anadolu Kardiyol Derg. 2013;13(8):784‐790. [DOI] [PubMed] [Google Scholar]
- 23. Demirelli S, Degirmenci H, Ermis E, et al. The importance of speckle tracking echocardiography in the early detection of left ventricular dysfunction in patients with polycystic ovary syndrome. Bosn J Basic Med Sci. 2015;15(4):44‐49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Mogelvang R, Sogaard P, Pedersen SA, et al. Cardiac dysfunction assessed by echocardiographic tissue Doppler imaging is an independent predictor of mortality in the general population. Circulation. 2009;119(20):2679‐2685. [DOI] [PubMed] [Google Scholar]
- 25. Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association joint committee on clinical practice guidelines. Circulation. 2022;145(18):e895‐e1032. [DOI] [PubMed] [Google Scholar]
- 26. Velusami D, Krishnamurthy S, Abiharini S. Association between QTc and Poincare plot in polycystic ovary syndrome in young adolescents: a cross‐sectional study. J Family Med Prim Care. 2024;13(11):5320‐5324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Geronikolou S, Bacopoulou F, Chryssanthopoulos S, Cokkinos DV, Chrousos GP. Hypertension predisposition and thermoregulation delays in adolescents with polycystic ovary syndrome: a pilot study. Children (Basel). 2022;9(3):316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Mirzohreh ST, Panahi P, Heidari F. Exploring heart rate variability in polycystic ovary syndrome: implications for cardiovascular health: a systematic review and meta‐analysis. Syst Rev. 2024;13(1):194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. de Medeiros SF, de Medeiros MAS, Barbosa BB, Yamamoto MMW, Maciel GAR. Comparison of metabolic and obesity biomarkers between adolescent and adult women with polycystic ovary syndrome. Arch Gynecol Obstet. 2021;303(3):739‐749. [DOI] [PubMed] [Google Scholar]
- 30. Akgül S, Bonny AE. Metabolic syndrome in adolescents with polycystic ovary syndrome: prevalence on the basis of different diagnostic criteria. J Pediatr Adolesc Gynecol. 2019;32(4):383‐387. [DOI] [PubMed] [Google Scholar]
- 31. Fu L, Xie N, Qu F, Zhou J, Wang F. The association between polycystic ovary syndrome and metabolic syndrome in adolescents: a systematic review and meta‐analysis. Reprod Sci. 2023;30(1):28‐40. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Comparison of anthropometric data, metabolic syndrome parameters and laboratory findings between PCOS and control groups.
Table S2. Comparison of conventional and M‐mode echocardiographic findings between PCOS and control groups.
Table S3. Comparison of tissue Doppler imaging, strain echocardiography, and 24‐h Holter monitoring data between PCOS and control groups.
Data Availability Statement
Research data are not shared.