Research Assignment

ORIGINAL ARTICLE

Psychological distress, anger and quality of life in polycystic ovary syndrome: associations with biochemical, phenotypical and socio-demographic factors

Lidia Borghia , Daniela Leonea, Elena Vegnia, Valentina Galianob, Corina Lepadatub, Patrizia Sulpiziob and Emanuele Garziab

aDepartment of Health Sciences, Unit of Clinical Psychology, University of Milan, Milan, Italy; bReproductive Medicine Unit, San Paolo Hospital, University of Milan, Milan, Italy

ABSTRACT Objective: To investigate the association between polycystic ovary syndrome (PCOS) and psy- chological disturbances, including anger. To analyze whether the biochemical/phenotypical fea- tures of PCOS play a role in the type and severity of psychological disorders. Material and methods: This case–control study included 30 PCOS patients meeting NIH criteria and 30 non-PCOS women referring to Reproductive Medicine Unit for infertility. Complete clinical and biochemical screening and the self-reported psychological data [Symptom Check List 90-R (SCL-90-R); Short-Form Health Survey 36 (SF-36); and State-Trait Anger Expression Inventory-2 (STAXI-2)] were collected. Statistical analyses were performed with SPSS-21. Results: Compared with control women, women with PCOS reported significantly higher scores on SCL-90-R scales of somatization, anxiety, hostility, psychoticism, overall psychological distress and a number of symptoms. At STAXI-2, patients with PCOS scored higher in trait-anger and in the outward expression of anger, while lower in outward anger-control; PCOS patients had sig- nificantly lower scores on SF-36 scales of physical functioning and bodily pain. Hirsutism was dir- ectly associated with anxiety. Regarding the associations between phenotypical/biochemical features and psychological distress in PCOS patients, results showed that waist-to-hip ratio is inversely related to anxiety, psychoticism, hostility and to the indexes of psychological distress; such inverse relationship was also seen between plasmatic levels of testosterone and trait-anger, and between total cholesterol and hostility. Conclusions: Results were consistent with the previous literature on the well-being of PCOS women (in particular for anxiety and quality of life [QoL]) but failed to find evidence for depres- sion. The relationship between psychological distress and the features of the syndrome high- lighted the role of hirsutism. With respect to hyperandrogenemia, our data rejected its involvement in the elevated negative mood states and affects. Adopting an interdisciplinary approach in the PCOS patients’ care, anger showed to be common and deserves major consideration.

ARTICLE HISTORY Received 7 June 2016 Accepted 14 March 2017

KEYWORDS Anger; hirsutism; PCOS; psychological distress; quality of life

Introduction

Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders of reproductive-aged women [1,2]. The great variability of PCOS’ prevalence, ranging from 2% to 26% [3,4], is due to the hetero- geneity of the gynecological and endocrine signs and symptoms, and to the difficulties for a general agree- ment over the diagnostic criteria. In fact, the 1990 National Institute of Health (NIH) criteria included only hyperandrogenism and anovulation [5], while the Rotterdam workshop of the European Society for Human Reproduction and Embryology/American

Society for Reproductive Medicine (ESHRE/ASRM) [6] added a third criterion (ultrasonographic evidence of polycystic ovaries) and stated that any two of the three features were sufficient for PCOS diagnosis, lead- ing to an increase of the prevalence of the syndrome [7]. Also the Androgen Excess Society (AES), dealt with the PCOS classification highlighting the crucial role of hyperandrogenemia [8].

PCOS is accompanied by an enhanced risk of insulin resistance, hyperinsulinemia and long-term health implications, including type-2 diabetes mellitus, dyslipi- demia and artherosclerosis [9–11].

CONTACT Lidia Borghi lidia.borghi@unimi.it Department of Health Sciences, University of Milan, San Paolo University Hospital, Via di Rudin�ı 8, 20142 Milan, Italy � 2017 Informa UK Limited, trading as Taylor & Francis Group

JOURNAL OF PSYCHOSOMATIC OBSTETRICS & GYNECOLOGY, 2018 VOL. 39, NO. 2, 128–137 http://dx.doi.org/10.1080/0167482X.2017.1311319

It is well-established that PCOS women suffer from impaired emotional well-being and markedly reduced quality of life (QoL) [12–15]. Studies suggested that a significant proportion of PCOS patients may suffer from anxiety and depression, which indeed constitutes a comorbidity in a large number of patients [16–19], as well as associations with mental disorders such as bipolar and posttraumatic stress disorders [20,21]. There is conflicting evidence on the association between biochemical markers and psychological out- comes: some studies have shown that hirsutism, body mass index (BMI) [22,23], acne [17,24], hyperandrogene- mia and insulin resistance [25] are associated with anx- iety and greater psychotic symptoms, whereas others failed to find this association [26]. The link between depressive disorders and hyperandrogenemia/hyperan- drogenism in PCOS patients has been shown inconclu- sive in several studies [27–29]. In addition, the role of physical appearance, particularly obesity and hirsutism has been relatively well-established: obesity is likely to be the most important factor in reducing QoL [30–33], and negative body image was strongly associated with depression [34,35]. However, the relationship between the phenotypical/biochemical features of PCOS and the type and severity of the psychological disorders remains unclear. A recent study by Moran et al. [14] brings some insight into the role of PCOS phenotypes (NIH and non-NIH diagnostic criteria) for psychological wellbeing, showing that non-NIH phenotypes exhibited similar psychological profiles as NIH PCOS, indicating increased psychological dysfunction in PCOS patients, even in milder reproductive phenotypes. However, women with NIH PCOS appear to have worse QoL in some areas than women with non-NIH PCOS.

As yet little attention has been paid to the symp- toms of anger and aggressiveness in PCOS patients. This gap in knowledge is of concern, given the poten- tial role of hyperandrogenemia and high levels of testosterone in inducing aggressiveness [36,37].

Hypothesizing that PCOS women would show higher rates of psychological alterations, the present study aimed to: 1) assess the psychological distress, anger and QoL in PCOS women compared to a control group of women; and 2) investigate the relationship between the psychological characteristics and the phenotypical/ biochemical features of the syndrome in PCOS women.

Materials and methods

Participants

We progressively enrolled a convenience sample of 30 PCOS women referring to an outpatient clinic of

Gynecological Endocrinology of a University Hospital in northern Italy during a period of four months. We chose to select PCOS patients who met the more strict and conservative NIH diagnostic criteria which include the presence of oligomenorrhea (cycles lasting >35 days) or amenorrhea (no periods in 6 months) and hyperandrogenemia/hyperandrogenism (hirsutism or obvious acne or pronounced alopecia). Additionally, we decided to enroll PCOS patients that were all infer- tile at the moment they filled out the questionnaires, in order to control for the potential confounding role of infertility on psychological outcomes. Different pitu- itary, adrenal, ovarian, thyroid or metabolic diseases have been excluded. All patients had bilateral polycys- tic ovaries morphology on ultrasound. Fifteen PCOS patients were taking metformin (500 mg three times daily) according to our PCOS treatment protocol for insulin-resistant patients.

We enrolled a control group of 30 women, age- matched with the PCOS women, from consecutive women seen in the same outpatient clinic who met the following inclusion criteria: history of regular men- strual cycle, absence of severe gynecologic and non- gynecologic diseases. Our PCOS sample was entirely constituted by infertile women at the moment they filled out the questionnaires; therefore, with the aim to limit the potential effect of the unfulfilled wish to conceive in our results, we admitted infertile women to the control group but not for endogenous reasons (e.g. male infertility).

Exclusion criteria, in both groups, included a prior psychiatric diagnosis, current use of psychiatric medications and difficulties with Italian language comprehension.

Women of both groups have had spontaneous onset of puberty and regular sexual development. Women were screened with detailed clinical and gyne- cological examination, pelvic ultrasonography, plasma endocrine and metabolic evaluation.

Measures and instruments

Socio-demographic data

Age, relational status, educational level and profes- sional condition were collected.

Psychological distress

Participants’ psychological distress was measured using the Symptom Checklist-90-Revision (SCL-90-R) [38–40], a multidimensional, self-report questionnaire of 90 items on a 5-point Likert scale (ranging from “0 ¼ no problem” to “4 ¼ very serious”). The questionnaire

JOURNAL OF PSYCHOSOMATIC OBSTETRICS & GYNECOLOGY 129

evaluates a broad range of psychological problems and symptoms of psychopathology clustered in 9 pri- mary scales (somatization, obsessive-compulsive, inter- personal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation and psychoticism). It also includes 3 Global Indices: Global Severity Index (GSI), designed to measure overall psychological dis- tress; Positive Symptom Distress Index (PSDI), measur- ing the intensity of symptoms; Positive Symptom Total (PST), i.e. number of self-reported symptoms. Higher scores on scales and indexes of the SCL-90-R indicate higher distress. The Italian version showed good internal consistency for all subscales (Cronbach’s alpha values between 0.70 and 0.96) [41].

Anger and aggressiveness

Participants’ trait anger and aggressiveness were eval- uated using the State-Trait-Anger-Expression-Inventory, version 2 (STAXI-2) [42,43], a self-report questionnaire of 57-items on a 4-point scale. The instrument meas- ures five different dimensions; we decided to leave out the first, that is how a person feels angry at a given moment (State-Anger), and to focus on the other four: how frequently, easily, and intensely the person feels angry (Trait-Anger) and what the person does when feeling angry (e.g. express or control the anger) (Anger Expression-In, Anger Expression-Out, Anger Control-In, Anger Control-Out). Higher scores indicate higher levels of trait-anger, and prevalence on the modality to cope with anger. The Italian version showed good internal consistency (Cronbach's alpha coefficient ranges from 0.74 to 0.95) [44].

Quality of life (QoL)

Health-related QoL was assessed using the Italian ver- sion of the Short Form Health Survey (SF-36) [45,46], an instrument composed by eight subscales, namely Physical Functioning, Physical Role Function, Bodily Pain, General Health, Vitality, Social Functioning, Emotional Role Function and Mental Health. The sub- scales scores range between 0 and 100, with lower scores indicating poorer QoL and a higher concern in the specific domain. The Italian version showed good internal consistency for all the subscales (Cronbach's alpha values range from 0.77 to 0.93) [46]. In addition, the subscales are combined to yield two summary health status measures, the Physical Sum scale and the Psychological Sum scale [45,47].

Phenotypical and biochemical features

As regards the phenotypical features, BMI (kg/m2) was calculated from height and weight measurements.

Abdominal circumference (AC) was recorded and waist-to-hip (W/H) ratio calculated. Hirsutism scoring was done using the modified Ferriman–Gallwey score (mFG score). Nine body areas were scored on a 0–4 scale for facial and body terminal hair growth distribu- tion, with a mFG score of �6 considered to be hirsut- ism. Presence of acne and/or acanthosis nigricans was noted.

Laboratory investigations and ultrasound examina- tions were performed in the early follicular phase (days 3–5) of spontaneous bleeding or withdrawal bleeding induced with medroxyprogesterone acetate. Blood samples were drawn at 8:00 am after overnight bed rest and fasting, subjected to centrifugation to separate the corpuscular phase and serum examin- ation which were stored at �20 �C until the moment of analysis.

The hormonal evaluation and the metabolic assess- ment consisted of assays of specific plasmatic compounds listed in Table 1. Insulin resistance was evaluated using the homeostasis model analysis (HOMA) [fasting glucose (mmol/l) � fasting insulin (mU/ml)/22.5] [48] and the fasting glucose-to-insulin ratio (GIR) (mg/1024 U) [49,50]. Finally, the free andro- gen index (FAI) [T (nmol/l)/SHBG �100] [51] was calcu- lated for each participant. TSH, FT4, Cortisol, Prolactin and 17-OH-progesterone assays allowed to recognize in advance the presence of other endocrine dysfunc- tions that can mime the features of the syndrome, rea- son for exclusion from the study. The intra-assay and inter-assay coefficients of variation (CV) for each bio- chemical or hormonal parameter were evaluated, and the values of the CVs were in any case respectively lower than 6 and 11%.

Data analysis

Two-tailed statistical analysis was performed using SPSS for Windows 21.0 software (SPSS Inc., Chicago, IL) with statistical significance set at a level of p � 0.05.

To describe demographic and hormonal data, descriptive statistics (means, standard deviations, fre- quencies) were performed, and Pearson’s Chi-square analyses and t-tests were used to assess differences between PCOS and controls.

First, data were assessed for normality using the Kolmogorov–Smirnov tests. As data were mainly non- normally distributed and given the small sample size, non-parametric tests were applied. The differences in SCL90-R subscales, STAXI-II subscales and SF-36 dimen- sions mean scores between PCOS patients and control groups were tested through Mann–Whitney U tests. The median and interquartile range (IQR) of the data

130 L. BORGHI ET AL.

were presented. The analyses were also performed splitting the PCOS group in the Metformin sub-group and the non-Metformin sub-group.

Spearman partial correlations, controlling for time since diagnosis and treatment, were used to explore the presence of associations between the psycho- logical variables (SCL-90-R, STAXI-II and SF-36 dimen- sions) that altered in PCOS patients and the biochemical/phenotypical variables or the socio-demo- graphic characteristics among the PCOS patients. Finally, associations among all intervening variables were summed up in order to create a conceptual synthesis of factors related to psychological maladjust- ment in PCOS patients.

Ethical approval

The study protocol was approved by the Hospital Ethics Committee. All participants gave an informed written consent before entering the study.

Results

Socio-demographic and clinical characteristics

Out of 43 PCOS women who met study criteria, 30 agreed to participate with a response rate of 70%, and out of 40 control women who met study criteria, 30 agreed to participate with a response rate of 75%.

Participants’ socio-demographic characteristics are shown in Table 2, with no significant differences

between groups. Table 3 summarizes the clinical and the hormonal/biochemical data. The two groups, except for total cholesterol, significantly differ for all the weighted features. The study group clearly shows either greater prevalence of obesity (BMI) and abdom- inal fat disposition (W/H) or hyperandrogenemia (T, FAI), hyperandrogenism (mFG) and hyperinsuline- mia (basal insulinemia, HOMA, GIR). No women pre- sented acne; 3 patients of the study group showed bilateral acanthosis nigricans on the thighs.

Psychological distress (SCL-90-R)

Compared with controls, PCOS patients showed higher levels of psychological distress in all the SCL-90-R dimensions and significantly increased scores on the primary scales Somatization (U ¼ 237.5, p ¼ 0.002), Anxiety (U ¼ 275.5, p ¼ 0.009), Hostility (F ¼ 260.5, p ¼ 0.005), Paranoid ideation (U ¼ 307.5, p ¼ 0.033) and Psychoticism (U ¼ 241, p ¼ 0.002); PCOS patients also showed a higher score in two indexes: Global Severity Index (U ¼ 286, p ¼ 0.015); Positive Symptoms Total (U ¼ 276.5, p ¼ 0.010) (Table 4).

Anger and aggressiveness (STAXI-II)

PCOS women scored higher in Trait-Anger (U ¼ 253, p ¼ 0.003) and Anger Expression-Out (U ¼ 305,

Table 1. Laboratory characteristics of hormonal and biochemical data. Plasmatic biochemical compound Method Laboratory tool

Sensitivity of the analytical method Intra-assay CV% Inter –assay CV%

Follicle-stimulating hormone (FSH)

Chemiluminescence Immuno Assay

Vitros 5600 0.66 IU/l <2.1 <10L

Luteinizing hormone (LH) Chemiluminescence Immuno Assay

Vitros 5600 0.215 IU/l <1.5 <11

17 b Estradiol (17-b E2) Chemiluminescence Immuno Assay

Vitros 5600 6.36 pg/mL <4.7 <9.3

17-OH-Progesterone (17-OH Pg)

RIA Immulite 2000 Xpi TEK 2 ORM/MET 010

<10.5

Thyroid-stimulating hormone (TSH)

Chemiluminescence Immuno Assay

Vitros 5600 0.014 lUI/mL <2.1 <2.2

Prolactin Chemiluminescence Immuno Assay

Vitros 5600 1.4 ng/dl <1 <5.6

Androstenedione Chemiluminescence Immuno Assay

Liaison Diasorin 2.4 lg/dL <3.7 <10

Dehydroepiandrosterone sul- fate (DHEA-S)

Chemiluminescence Immuno Assay

Liaison Diasorin 0.7 lg/dL <5.8 <10

Fasting insulin Non Isotopic Immunoassay

Tosoh AIA 0.5 lU/mL <2.3 <4.6

Fasting glucose Enzymatic Colorimetric Assay

Vitros 5600 20 mg/dL <1.2 <2.2

Sex hormone binding globulin (SHBG)

Chemiluminescence Immuno Assay

Immulite 2000 Xpi ORM/ MET 012

1 nmol/L <6 <8

Total testosterone Chemiluminescence Immuno Assay

Vitros 5600 4.9 ng/dL <2.4 <7.4

Cortisol Chemiluminescence Immuno Assay

Vitros 5600

Total cholesterol Enzymatic Colorimetric Assay

Vitros 5600

JOURNAL OF PSYCHOSOMATIC OBSTETRICS & GYNECOLOGY 131

p ¼ 0.031) and lower in Anger Control-Out (U ¼ 298.5, p ¼ 0.025) when compared with the controls (Table 4).

Quality of life (SF-36)

PCOS patients had significantly lower scores on Physical Function (U ¼ 213.5, p < 0.001), Bodily Pain (U ¼ 201, p < 0.001) SF-36 scales (Table 4) and on the Physical Sum scale (v2 ¼ 806, p ¼ 0.005) indicating markedly reduced QoL in the field of physical health compared with the controls; no differences were found for the other scales and for the Mental Sum scale.

No differences were found between Metformin PCOS sub-group and non-Metformin PCOS sub-group for any of the psychological and QoL scales.

Correlation between psychological data and biochemical/phenotypical data or socio-demographic characteristics in PCOS women

Psychological distress. In PCOS Patients, Anxiety was found to be directly related to mFG (q ¼ 0.26, p ¼ 0.045) and inversely to W/H (q ¼ �0.38, p ¼ 0.038); Hostility was negatively associated with W/H (q ¼ �0.4, p ¼ 0.028) and total levels of Cholesterol (q ¼ �0.4, p ¼ 0.028); Psychoticism negatively correlated with W/H (q ¼ �0.41, p ¼ 0.023); GSI and PST Positive Symptoms

Total correlated both with W/H (respectively q ¼ �0.46, p ¼ 0.011 and q ¼ �0.46, p ¼ 0.014); Somatization did not correlate with any biochemical variable. No correla- tions were found between socio-demographic charac- teristics and outcomes of psychological distress.

Anger. Trait-Anger was negatively associated with total Testosterone (q ¼ �0.40, p ¼ 0.030) and W/H (q ¼ �0.40, p ¼ 0.028). Anger Expression-Out and Anger Control-Out were not associated with any bio- chemical variable, while Anger Control-Out positively correlated with educational level (q ¼ 0.432, p ¼ 0.019)

Quality of life. SF-36 scales altered in PCOS patients compared with controls (Physical Functioning, Bodily Pain, Physical Sum scale) did not correlate with any biochemical variable. Physical Functioning positively correlated with educational level (q ¼ 0.371, p ¼ 0.047).

Figure 1 summarized the associations among all intervening variables and represents a conceptual syn- thesis of factors related to psychological maladjust- ment in PCOS patients.

Discussion

This study reports a comprehensive exploration of psy- chological functioning and QoL of PCOS patients, as well as their specific biochemical features and the associations between these outcomes. The study

Table 2. Patients’ socio demographical characteristics. PCOS Controls p

Age, mean (sd) 33.7 (5.7) 35.3 (4.6) 0.259 Relationship status, % (n) 0.535 Single 23.3% (7) 30% (9) Cohabiting/married 66.7% (20) 66.7% (20) Divorced/Widow 10% (3) 3.3% (1)

Educational level, % (n) 0.056 Primary 37.9% (11) 23.3% (7) High school 51.7% (15) 40% (12) Graduate 10.3% (3) 36.7% (11)

Profession, % (n) 0.136 Professional conditions 66.7% (20) 83.3% (25) Non-professional conditions (housewives, students, unemployed) 33.3% (10) 16.7% (5)

Pearson’s Chi-square analyses and t-tests were used to assess differences between PCOS and controls.

Table 3. Patients phenotypical and biochemical data. PCOS (n ¼ 30) Controls (n ¼ 30) Mean ± sd Mean ± sd p

Body Mass Index – BMI (kg/m2) 33.35 ± 5.85 23.46 ± 3.43 <0.001 Waist to hip ratio – W/H (cm) 0.94 ± 0.08 0.77 ± 0.05 <0.001 Modified Ferriman–Gallwey score A – mFG a 12.13 ± 7.51 5.87 ± 3.08 <0.001 Total Testosterone – T (nmol/L) 2.01 ± 0.9 1.35 ± 0.62 0.001 Free Androgen Index – FAI (%) 11.14 ± 8.65 2.41 ± 1.37 <0.001 Basal insulinaemia (lU/ml) 17.53 ± 9.77 8.78 ± 4.39 <0.001 Homeostasis Model Assessment – HOMA 4.1 ± 2.46 1.96 ± 1.02 <0.001 Glucose-to-insulin ratio – GIR 7.14 ± 4.73 13.74 ± 8.44 <0.001 Sex Hormone Binding Globulin – SHBG (nmol/L) 24.42 ± 9.44 63.49 ± 30.76 <0.001 Total cholesterol (mg/dl) 191.97 ± 39.65 193.80 ± 18.05 0.819

T-tests were used to assess differences between PCOS and controls.

132 L. BORGHI ET AL.

includes an evaluation of feelings of anger and reac- tions, that were not frequently explored in PCOS patients.

Overall, PCOS patients showed significantly more pronounced emotional distress, assessed with the scales of the SCL-90-R, compared with the controls; in particular, they showed more symptoms of somatiza- tion, anxiety, hostility, psychoticism, a greater overall psychological distress (GSI) and reported a greater number of symptoms (PST). Our data strongly support more recent studies [30,52–55] on PCOS emotional outcomes: anxiety is significant in PCOS patients, is mainly focused on the somatic and body-related aspects and is frequently underdiagnosed. Considering depression, unlike some previous evidence [52,53,56], we regard the association with PCOS as inconclusive.

In our study, PCOS women reported higher levels of trait-anger and a higher tendency to express their anger outward, while, showing a lower tendency to control their anger. . This is an interesting and rela- tively new finding [57,58] suggesting that anger expression is usually an underestimated aspect, deserving of major consideration when caring for PCOS patients.

Furthermore, PCOS patients showed an altered QoL but only in respect to physical health; in particular to

Physical Function and Bodily Pain (resulting also in an altered Physical Sum scale), while they did not report problems in the emotional dimension of SF36. In this field, previous studies suggested that PCOS and its associated symptoms have a negative effect on both physical and emotional dimensions measured either with SF-36 [23,30] or with other instruments [58]. PCOS The absence of an impact on emotional QoL could be seen as a contradiction of what emerged by SCL-90-R and STAXI-2 unless we look at the instru- ments used. Compared to SCL90R and STAXI that evaluate frequency and intensity of symptoms or feel- ings experienced, SF-36 investigates QoL dimensions that are more mediated by the awareness of the impact that one's emotional status exerts on occupa- tional or social functioning. Lastly, some authors [23,30] suggested that the physical aspect of QoL may be closely related to the PCOS features since they are best predicted by phenotypical aspects such as obesity and hirsutism, and were found to be mostly associated with the NIH phenotype [14].

Consistent with previous studies [30,53], in our PCOS sample psychological distress and QoL were not associated with socio-demographic factors (age, mari- tal status and education); only a lower educational level was related to a lower perceived QoL in a

Table 4. Psychological symptoms and health-related QoL in patients with PCOS and control women. PCOS (n ¼ 30) Controls (n ¼ 30) Median (IQR) Median (IQR) Mann–Withney U test

Psychological distress (SCL-90-R) Somatization 0.75 (0.42–1.45) 0.46 (0.08–0.66) 237.5�� Obsessive-compulsive 0.45 (0.20–0.85) 0.45 (0.10–1.00) 416.5 Interpersonal sensitivity 0.39 (0.22–0.89) 0.22 (0.00–0.91) 358.5 Depression 0.70 (0.21–0.94) 0.48 (0.13–0.98) 401.5 Anxiety 0.45 (0.20–0.93) 0.20 (0.10–0.45) 275.5�� Hostility 0.50 (0.29–0.83) 0.33 (0.00–0.54) 260.5�� Phobic anxiety 0.07 (0.00–0.29) 0.00 (0.00–0.14) 337.5 Paranoid ideation 0.67 (0.29–1.00) 0.08 (0.00–1.00) 307.5� Psychoticism 0.30 (0.1–0.43) 0.00 (0.00–0.23) 241.0�� Global Severity Index 0.52 (0.33–0.84) 0.26 (0.16–0.80) 286.0� Positive Symptom Distress Index 33 (25.75–45.50) 19 (13.25–35.75) 276.5� Positive Symptom Total 1.3950 (1.23–1.68) 1.43 (1.17 –1.88) 445.0

Anger and aggressiveness (STAXI-II) Trait-Anger 8 (6.75–9.25) 6 (5–8.25) 253.0�� Anger Expression-Out 15 (12.75–16.5) 13 (11–15) 411.5� Anger Expression-In 17 (11.75–19.25) 14.5 (11.75–17.25) 305.0 Anger Control-Out 20 (15–23.25) 23 (18.75–25) 439.5� Anger Control-In 23 (18–27.25) 23 (19–25.25) 298.5

Health-related QoL (SF-36) Physical Functioning 90 (75–96.25) 100 (95–100) 213.5��� Physical Role Functioning 100 (75–100) 100 (87.5–100) 413.5 Bodily Pain 61 (48.75–84) 84 (74–100) 201.0��� General Health 72 (50.75–83) 76 (67–82) 322.0 Vitality 60 (45–70) 65 (52.5–75) 378.0 Social Functioning 87 (62–87) 100 (68–100) 337.0 Emotional Role Functioning 100 (57.75–100) 100 (66–100) 430.5 Mental Health 66 (52–84) 76 (58–88) 353.5

�p < 0.05. ��p < 0.01. ���p < 0.001.

JOURNAL OF PSYCHOSOMATIC OBSTETRICS & GYNECOLOGY 133

physical domain (physical functioning) and to a lower capacity to control the outward expression of anger, which remains in accordance with the literature on the associations between lower educational level and psy- chiatric disorders [59].

The present study failed in its aim to demonstrate a role of phenotypical or biochemical PCOS features in the pathogenesis of psychological distress. Nevertheless, some associations between psychological outcomes and biochemical/hormonal characteristics, particularly for those data that are related to body image, such as the hirsutism and the W/H ratio, seem to be interesting and need to be deepened. In particu- lar, hirsutism was found to be associated with anxiety. Consistent with previous studies [23,53,60], hirsutism seemed to be the most distressing aspect for the PCOS patients’ emotional wellbeing, and it has also been found to be directly related to anxiety among the general population [61].

Testosterone revealed a peculiar association with anger. Previous studies found an association between elevated testosterone levels and physical symptoms of PCOS, but failed to find the role of testosterone in PCOS emotional problems; nonetheless, PCOS patients report high mood disturbances [23,29,62,63]. Our find- ings in PCOS women revealed an inverse relationship between Testosterone and aggressiveness compared to the one expected: the most elevated trait-anger

scores were associated with lower plasmatic levels of Testosterone, and vice versa. This is partially consistent with Weiner et al. [57] findings that reported a curvi- linear relationship between plasmatic Testosterone and negative effects in PCOS patients. They tried to explain this unexpected association through a model of activational influences of testosterone on adult female behavior: the moods tended to moderate at the extreme levels of the hormone because of compe- tition on receptor binding sites and adjustments in the number of receptors.

Furthermore, the levels of total cholesterol revealed an inverse association with another aspect of anger, that is hostility. Few literature has investigated the relationship between cholesterol and anger; how- ever, one study [64] reported that lower levels of total cholesterol were associated with symptoms of anger/ hostility in the postpartum.

Unexpectedly, and in contrast with previous studies on the role of obesity [30–32], W/H ratio was inversely related to anxiety, hostility, psychoticism, a greater overall psychological distress and a greater number of symptoms.

The study has some critical points. The sample size of patients was small and only one center was involved, which may limit the possibility of generaliz- ing. Nevertheless, in order to identify a more strictly diagnosed PCOS population, and to clearly highlight

Figure 1. Overview of the associations between biochemical/phenotypical features, socio-demographic data and psychological out- comes. �p < 0.05, ��p < 0.001.

134 L. BORGHI ET AL.

any psychological distress linked to the PCOS syn- drome, we resorted to NIH criteria and performed a complete clinical and biochemical screening. However, the correlations between psychological and biochem- ical data should be considered with caution and future studies should investigate the conformity of these associations across different PCOS phenotypes (i.e. if there are differences among NIH-PCOS versus non-NIH PCOS). The presence of a control group com- parable for socio-economic characteristics could par- tially control for confounders, as the role of educational level should be investigated in further studies with larger samples due to the impossibility in the present study to adjust for its potential confound- ing role (in comparing PCOS patients and controls). Moreover, it could be considered a limitation that half of the study group patients were taking Metformin according to our PCOS treatment protocol in insulin- resistant patients. Nonetheless the results obtained in the Metformin PCOS sub-group with regard to psy- chological functioning, feelings of anger and QoL were not significantly different from the non- Metformin PCOS sub-group.

Finally, our PCOS sample was entirely composed of infertile women at the moment they filled out the questionnaires, so it could be difficult to distinguish whether alterations in the emotional well-being and QoL were due to infertility and to the potential unful- filled wish to conceive or to the PCOS condition. We tried to control this aspect recruiting a control group of infertile women, but further studies with a fertile controlled group could clarify the role of infertility on PCOS patients [59].

The points of strength of the present study are to provide a more comprehensive evaluation of PCOS psychological functioning, including the dimension of aggressiveness, that is a value-added contribution rela- tive to previous findings, and further to explore the associations between psychological characteristics and biochemical features. However, the study revealed some methodological issues, in particular in terms of external validity, that temper the conclusions that can be drawn and have implications for future research.

In conclusion, PCOS patients are at risk to suffer from psychological disturbances such as anxiety and aggressiveness and have implication for health-related QoL. Both biochemical and phenotypical factors seem to be associated with the wellbeing of PCOS women (e.g. testosterone and anger) but in an inconclusive way. Our preliminary findings might help clinicians pay attention to aspects of emotional wellbeing of patients with PCOS that differ from anxiety or depression, as the trait of aggressiveness and anger expression.

Future studies with larger samples should investigate the role of phenotypical and biochemical features of PCOS in the type and severity of psychological dis- tress, exploring the direction of the associations and clarifying the confounding/predicting role of the edu- cational level.

Acknowledgements

The authors would like to thank all of the women who so graciously volunteered to be participants in this study.

Disclosure statement

The authors report no conflicts of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

ORCID

Lidia Borghi http://orcid.org/0000-0001-9581-9921

References

1. Diamanti-Kandarakis E, Kouli CR, Bergiele AT, et al. A survey of the polycystic ovary syndrome in the Greek island of Lesbos: hormonal and metabolic profile. J Clin Endocrinol Metab 1999;84:4006–11.

2. Azziz R, Woods KS, Reyna R, et al. The prevalence and features of the polycystic ovary syndrome in an unse- lected population. J Clin Endocrinol Metab 2004;89: 2745–9.

3. Knochenhauer ES, Key TJ, Kahsar-Miller M, et al. Prevalence of the polycystic ovary syndrome in unse- lected black and white women of the southeastern United States: a prospective study. J Clin Endocrinol Metab 1998;83:3078–82.

4. Asuncion M, Calvo RM, San Millan JL, et al. A prospect- ive study of the prevalence of the polycystic ovary syn- drome in unselected Caucasian women from Spain. J Clin Endocrinol Metab 2000;85:2434–8.

5. Zawadzki JK, Dunaif A, Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In: Dunaif A, Givens JR, Haseltine FP, Merriam GR, eds. Polycystic ovary syndrome. Boston (MA): Blackwell Scientific Publications; 1992:377–84.

6. The Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril 2004;81:19–25.

7. March WA, Moore VM, Willson KJ, et al. The prevalence of polycystic ovary syndrome in a community sample

JOURNAL OF PSYCHOSOMATIC OBSTETRICS & GYNECOLOGY 135

assessed under contrasting diagnostic criteria. Hum Reprod 2010;2:544–51.

8. Azziz R, Carmina E, Dewailly D, et al. Criteria for defin- ing polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an androgen excess soci- ety guideline. J Clin Endocrinol Metab 2006;91: 4237–45.

9. Hardiman O, Pillay OS, Atiomo W. Polycystic ovary syn- drome and endometrial carcinoma. Lancet 2003; 361:1810–12.

10. Legro RS. Polycystic ovary syndrome and cardiovascu- lar disease: a premature association?. Endocr Rev 2003;24:302–12.

11. Ehrmann DA. Polycystic ovary syndrome. N Engl J Med 2005;352:1223–36.

12. Barnard L, Ferriday D, Guenther N, et al. Quality of life and psychological well being in polycystic ovary syn- drome. Hum Reprod 2007;22:2279–86.

13. Jones GL, Hall JM, Balen AH, Ledger WL. Health-related quality of life measurement in women with polycystic ovary syndrome: a systematic review. Hum Reprod Update 2008;14:15–25.

14. Moran LJ, Deeks AA, Gibson-Helm ME, Teede HJ. Psychological parameters in the reproductive pheno- types of polycystic ovary syndrome. Hum Reprod 2012;27:2082–8.

15. Moran L, Gibson-Helm M, Teede H, Deeks A. Polycystic ovary syndrome: a biopsychosocial understanding in young women to improve knowledge and treatment options. J Psychosom Obstet 2010;31:24–31.

16. Mansson M, Holte J, Landin-Wilhelmsen K, et al. Women with polycystic ovary syndrome are often depressed or anxious – a case control study. Psychoneuroendocrinology 2008;33:1132–8.

17. Benson S, Hahn S, Tan S, et al. Prevalence and implica- tions of anxiety in polycystic ovary syndrome: results of an internet-based survey in Germany. Hum Reprod 2009;24:1446–51.

18. Jedel E, Waern M, Gustafson D, et al. Anxiety and depression symptoms in women with polycystic ovary syndrome compared with controls matched for body mass index. Hum Reprod 2010;25:450–6.

19. Dokras A, Clifton S, Futterweit W, Wild R. Increased risk for abnormal depression scores in women with poly- cystic ovary syndrome: a systematic review and meta- analysis. Obstet Gynecol 2011;117:145–52.

20. Dobie DJ, Kivlahan DR, Maynard C, et al. Posttraumatic stress disorder in female veterans: association with self-reported health problems and functional impair- ment. Arch Intern Med 2004;164:394–400.

21. Klipstein KG, Goldberg JF. Screening for bipolar dis- order in women with polycystic ovary syndrome: a pilot study. J Affect Disord 2006;91:205–9.

22. Sonino N, Fava GA, Mani E, et al. Quality of life of hir- sute women. Postgrad Med J 1993;69:186–9.

23. Hahn S, Janssen OE, Tan S, et al. Clinical and psycho- logical correlates of quality-of-life in polycystic ovary syndrome. Eur J Endocrinol 2005;153:853–60.

24. Mallon E, Newton JN, Klassen A, et al. The quality of life in acne: a comparison with general medical condi- tions using generic questionnaires. Br J Dermatol 1999;140:672–6.

25. Livadas S, Chaskou S, Kandaraki AA, et al. Anxiety is associated with hormonal and metabolic profile in women with polycystic ovarian syndrome. Clin Endocrinol (Oxf) 2011;75:698–703.

26. Kerchner A, Lester W, Stuart SP, Dokras A. Risk of depression and other mental health disorders in women with polycystic ovary syndrome: a longitudinal study. Fertil Steril 2009;91:207–12.

27. Keegan A, Liao LM, Boyle M. ‘Hirsutism’: a psycho- logical analysis. J Health Psychol 2003;8:327–45.

28. Rasgon NL, Rao RC, Hwang S, et al. Depression in women with polycystic ovary syndrome: clinical and biochemical correlates. J Affect Disord 2003;74: 299–304.

29. Hollinrake E, Abreu A, Maifeld M, et al. Increased risk of depressive disorders in women with polycystic ovary syndrome. Fertil Steril 2007;87:1369–76.

30. Elsenbruch S, Hahn S, Kowalsky D, et al. Quality of life, psychosocial well-being, and sexual satisfaction in women with polycystic ovary syndrome. J Clin Endocrinol Metab 2003;88:5801–7.

31. Guyatt G, Weaver B, Cronin L, et al. Health-related quality of life in women with polycystic ovary syn- drome, a self-administered questionnaire, was vali- dated. J Clin Epidemiol 2004;57:1279–87.

32. Coffey S, Bano G, Mason HD. Health-related quality of life in women with polycystic ovary syndrome: a com- parison with the general population using the Polycystic Ovary Syndrome Questionnaire (PCOSQ) and the Short Form-36 (SF-36). Gynecol Endocrinol 2006;22:80–6.

33. Ching HL, Burke V, Stuckey BG. Quality of life and psy- chological morbidity in women with polycystic ovary syndrome: body mass index, age and the provision of patient information are significant modifiers. Clin Endocrinol (Oxf) 2007;66:373–9.

34. Himelein MJ, Thatcher SS. Depression and body image among women with polycystic ovary syndrome. J Health Psychol 2006a;11:613–25.

35. Himelein MJ, Thatcher SS. Polycystic ovary syndrome and mental health: a review. Obstet Gynecol Surv 2006b;61:723–32.

36. Archer J. Testosterone and human aggression: an evaluation of the challenge hypothesis. Neurosci Biobehav Rev 2006;30:319–45.

37. Van de Poll NE, van Goozen SHM. Hypothalamic involvement in sexuality and hostility: comparative psychological aspects. Prog Brain Res 1992;93:343–61.

38. Derogatis LR, Rickels K, Rock AF. The SCL-90 and the MMPI: a step in the validation of a new self-report scale. Br J Psychiatry 1976;128:280–9.

39. Derogatis LR, The SCL-90-R: administration, scoring and procedures manual. 3rd ed. Minneapolis, MN: National Computer Systems; 1994.

40. Sarno I, Preti E, Prunas A, Madeddu F (eds). SCL-90-R. Symptom Checklist-90-R. Firenze; Giunti Organizzazioni Speciali; 2011.

41. Prunas A, Sarno I, Preti E, et al. Psychometric properties of the Italian version of the SCL-90-R: a study on a large community sample. Eur Psychiatry 2012;27:591–7.

136 L. BORGHI ET AL.

42. Spielberger CD. State-trait anger expression inventory- 2 (STAXI-2). Professional manual. Tampa (FL): Psychological Assessment Resources; 1999.

43. Spielberger CD, State-trait-anger-expression-inventory. Palo Alto (CA): Consulting Psychologist Press; 1988.

44. Comunian L, Manuale STAXI-2 state-trait anger expres- sion inventory-2. Adattamento italiano. Firenze: Organizzazioni speciali; 2004.

45. Ware JE, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473–83.

46. Apolone G, Mosconi P. The Italian SF-36 Health Survey: translation, validation and norming. J Clin Epidemiol 1998;51:1025–36.

47. Apolone G, Mosconi P, Ware J, Questionario sullo stato di salute SF-36. Manuale d’uso e guida all’interpretazione dei risultati [SF-36 quality of life questionnaire. User’s manual and guide to the inter- pretation of results]. Guerini, Milano; 2000.

48. Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and b-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–19.

49. Legro RS, Finegood D, Dunaif A. A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1998;83:2694–8.

50. Ducluzeau PH, Cousin P, Malvoisin E, et al. Glucose-to- insulin ratio rather than sex hormone-binding globulin and adiponectin levels is the best predictor of insulin resistance in nonobese women with polycystic ovary syndrome. J Clin Endocrinol Metab 2003;88:3626–31.

51. Morley JE, Patrick P, Perry HM. III. Evaluation of assays available to measure free testosterone. Metab Clin Exp 2002;51:554–9.

52. Deeks AA, Gibson-Helm ME, Paul E, Teede HJ. Is having polycystic ovary syndrome a predictor of poor psycho- logical function including anxiety and depression? Hum Reprod 2011;26:1399–407.

53. Cinar N, Kizilarslanoglu MC, Harmanci A, et al. Depression, anxiety and cardiometabolic risk in poly- cystic ovary syndrome. Hum Reprod 2011;26:3339–45.

54. Balikci A, Erdem M, Keskin U, et al. Depression, anxiety, and anger in patients with polycystic ovary syndrome. Arch Neuropsychiatry 2014;51:328–33.

55. Laggari V, Diareme S, Christogiorgos S, et al. Anxiety and depression in adolescents with polycystic ovary syndrome and Mayer-Rokitansky-K€uster-Hauser syn- drome. J Psychosom Obstet Gynaecol 2009;30:83–8.

56. Pastore LM, Patrie JT, Morris WL, et al. Depression symptoms and body dissatisfaction association among polycystic ovary syndrome women. J Psychosom Res 2011;71:270–6.

57. Weiner CL, Primeau M, Ehrmann DA. Androgens and mood dysfunction in women: comparison of women with polycystic ovarian syndrome to healthy controls. Psychosom Med 2004;66:356–62.

58. Barry JA, Hardiman PJ, Saxby BK, Kuczmierczyk A. Testosterone and mood dysfunction in women with polycystic ovarian syndrome compared to subfertile controls. J Psychosom Obstet 2011;32:104–11.

59. Erickson J, El-Gabalawy R, Palitsky D, Patten S, Mackenzie CS, Stein MB, Sareen J. Educational attain- ment as a protective factor for psychiatric disorders: findings from a nationally representative longitudinal study. Depress Anxiety 2016;33:1013–22.

60. De Niet JE, de Koning CM, Pastoor H, et al. Psychological well-being and sexarche in women with polycystic ovary syndrome. Hum Reprod 2010; 25:1497–503.

61. Drosdzol A, Skrzypulec V, Plinta R. Quality of life, men- tal health and self-esteem in hirsute adolescent females. J Psychosom Obstet Gynaecol 2010;31:168–75.

62. Tan S, Hahn S, Benson S, et al. Psychological implica- tions of infertility in women with polycystic ovary syn- drome. Hum Reprod 2008;23:2064–71.

63. McCook JG, Reame NE, Thatcher SS. Health-related quality of life issues in women with polycystic ovary syndrome. J Obstet Gynecol Neonatal Nurs 2005; 34:12–20.

64. Troisi A, Moles A, Panepuccia L, et al. Serum choles- terol levels and mood symptoms in the postpartum period. Psychiatry Res 2002;109:213–19.

� Current knowledge on the subject. � It is well-established that PCOS women suffer from impaired emotional well-being (in particular anxiety and depression) and markedly reduced

quality of life � It remains unclear the relationship between the phenotypical and biochemical features of PCOS and the type and severity of the psychological

disorders � Hyperandrogenemia and high levels of testosterone play a potential role in inducing aggressiveness in healthy people, while these associations

are unclear in PCOS patients.

� What this study adds. � PCOS women reported high level of trait anger and a tendency to over-express anger outward when compared with controls � Hirsutism directly correlated to anxiety in PCOS women � Testosterone, cholesterol total and waist-to-hip ratio showed an inverse relationship with the emotional distress

JOURNAL OF PSYCHOSOMATIC OBSTETRICS & GYNECOLOGY 137

Copyright of Journal of Psychosomatic Obstetrics & Gynecology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.