Idiopathic hirsutism and dyslipidemia: A systematic review, meta‐analysis, and meta‐regressions

Despite existing several kinds of studies on idiopathic hirsutism (IH) and its association with metabolic disturbances, a limited number of studies have assessed lipid profiles in patients with IH with conflicting and inconclusive results. This study aimed to evaluate the lipid profiles in women with IH, compared to those of polycystic ovary syndrome (PCOS) and healthy controls.


| INTRODUC TI ON
Hirsutism is defined as the presence of excessive terminal hair in the androgenic area. Idiopathic hirsutism (IH) is considered the most prevalent cause of hirsutism. The prevalence of IH is estimated to be 5%-20%, which is different between ethnicities with more prevalence in Mediterranean ancestry. [1][2][3][4] This prevalent disorder is referred to as hirsutism without known causes and defined as hirsutism, along with normal serum androgen and normal ovulatory function after excluding other underlying etiologies, such as polycystic ovary syndrome (PCOS). 1,2,5 A recent meta-analysis estimated the prevalence of IH by 7.74%. 6 There is evidence demonstrating the association of hyperandrogenism disorders, particularly PCOS, with metabolic disturbances like dyslipidemia, which has been defined as increased concentrations of triglyceride (TG) or low-density lipoprotein cholesterol (LDL-C) or decreased values of high-density lipoprotein cholesterol (HDL-C), or receiving lipid-lowering therapy 7 ; findings remained significant even after adjusting to obesity. [8][9][10][11] Moreover, epidemiologic studies indicate that IH as a mild hyperandrogenism condition with a peripheral increased androgen activity might also be associated with some metabolic disturbances, such as dyslipidemia. [12][13][14] Since some investigators have described mild forms of ovarian and adrenal functional hyperandrogenism in some patients with IH, it is logical that IH increases the risk of metabolic disturbances in a similar manner of PCOS. 15,16 Despite existing several kinds of studies on IH and its association with metabolic disturbances, 17-32 a limited number of studies have assessed lipid profiles in patients with IH with conflicting and inconclusive results. [17][18][19][20][21][22][23] The contradictory results of studies might be due to the small sample size, low quality of studies, use of different diagnostic criteria, and lack of considering potential confounders.
This study has been the first meta-analysis investigating the lipid profiles in women with IH, compared to those of PCOS and healthy controls, before and after adjustment for potential confounders.

| MATERIAL AND ME THODS
This meta-analysis was conducted based on the guidelines for the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 33 to compare the Standardized Mean Difference (SMD) of dyslipidemia parameters between patients with IH, PCOS, and healthy controls.
The study protocol has been registered in the PROSPERO (Registration number: CRD42021264873).

| Search strategy
We systematically searched Pub-Med, Scopus, and EMBASE for retrieving published up to July 2021 investigating lipid parameters in women with IH, compared to PCOS, and healthy controls.
Search in PubMed was performed initially, based on MESH terms.
Search strategies were based on "titles, abstracts, and keywords" and were almost similar for all databases. A search was limited to human studies published in the English language. Time was not limited. Reference lists of included studies were reviewed for additional articles. Searches were performed by two reviewers independently.
The full electronic search strategies are shown in Table S1.

| Eligibility criteria
Observational studies, including cross-sectional, case-control, and cohort studies were eligible to be included in the meta-analysis if they assessed dyslipidemia in IH patients in comparison with healthy and/ or PCOS controls. We included studies in which IH was diagnosed as existing hirsutism with normo-ovulation and normal androgen levels, and PCOS was diagnosed based on the Rotterdam, NIH, or AES criteria. Any IH and PCOS diagnosis criteria were eligible to be included.
Studies were excluded if they had one or more of the following criteria: (1) Non-observational designs, (2) Lack of assessing women with IH as the study population, (3) unknown IH diagnostic criteria, (4) unknown PCOS diagnostic criteria or using other criteria, and (5) Studies without assessing at least one of lipid parameters.

| Study selection
Search results were screened based on predefined eligibility criteria. All relevant studies were included if they had assessed at least one of the lipid parameters, including total cholesterol, triglyceride, LDL-C (low-density lipoprotein cholesterol), HDL-C (high-density lipoprotein cholesterol), non-HDL-C. All references were entered in EndNote software. Firstly, we performed an initial selection based on the titles. Then, duplicates were deleted and the abstracts of the Conclusion: This study demonstrated that the majority of lipid parameters were not significantly different in women with IH, compared to those of healthy and PCOS controls.

K E Y W O R D S
dyslipidemia, idiopathic hirsutism, polycystic ovary syndrome remaining article were reviewed by one reviewer. Any disagreement in the selection of abstracts was resolved by other reviewers. Finally, the full text of all selected studies was obtained for data processing.
Any disagreement in abstract and full-text selection was resolved by other reviewers.

| Data extraction
Two reviewersindependently extracted data from full-text articles.
Disagreements were resolved by a senior reviewer. In addition to lipid parameters the following information was gathered from each study: Title, authors, year of publication, study design, diagnostic criteria for IH and PCOS, mean (SD) of age and body mass index (BMI), and adjustments or matching.

| Quality assessment
The quality of included studies was assessed by two reviewers independently while they were blinded to article characteristics, such as study authors, institution, and journal name. Newcastle-Ottawa scale for case-control studies and Appraisal tool for Cross-Sectional Studies (AXIS) were used to assess the quality of articles were scored as follows: If a study obtained ≥70% of the highest level of these tools, it was considered as high quality, those with 40%-70% F I G U R E 1 Study flow diagram of the study.  as moderate, and those with 20%-40% as low, and those with <20% as very low quality. 34,35 We assessed quality of the results, and method sections for each study.

| Risk of bias assessment
The Cochrane Collaboration's tools were applied to evaluate the risk of bias for included papers, which have been developed for different studies, such as case-control and cohorts. Review author's judgments were categorized as "low risk," "high risk," and "unclear risk" of bias (either low or high risk of bias). 36

| Statistical analysis
The current meta-analysis was conducted to assess the standardized mean difference (SMD) of lipid parameters in IH patients, compared with healthy and PCOS controls. We also ran a subgroup analysis to explore the SMD of lipid parameters in studied groups based on study design (cross-sectional vs case-control  (Table 1 and Tables S2 and S3).

TA B L E 2
The results of meta-regression analysis concerning the effects of potential confounders on the association between lipid parameters and idiopathic hirsutism (IH), compared to healthy and polycystic ovary syndrome (PCOS) controls Parameter Potential confounder

| Meta-regression and subgroup analysis
The results of meta-regression adjusted for confounding variables are presented in Table 2; although age was not a significant confounder, body mass index (BMI) seems to have a confounding effect on the primary results.

| DISCUSS ION
This meta-analysis indicates higher SMD of TC and lower SMD of HDL-C in women with IH, compared to those of the healthy controls, whereas there were no significant differences in the SMDs of TG and LDL-C between study groups. This study also revealed no difference in the levels of lipid profiles of women with IH in comparison with PCOS patients, except for TG.
Although there is uncertainty over the diagnosis of IH, mainly due to the lack of a unique definition for this endocrine disorder, recent evidence considers it as existing hirsutism with average circulating concentrations of androgens, regular menstrual cycles, and normal ovaries after excluding other androgen excess disorders. 5 Although the pathogenesis of IH is not entirely elucidated, several mechanisms might be involved in its pathogenesis. Primary increased skin 5α-reductase activity despite normal androgen levels is considered the main event in women with IH. 43 Differences in androgen receptors due to genetic variations in women with IH might explain the presence of hirsutism in the absence of increased androgen levels. 44 Abnormalities in peripheral androgen metabolism were reported in women with IH 45 ; however, they have normal levels of testosterone, 17α-hydroxyprogesterone, and dehydroepiandrosterone sulfate. 46 In addition to hormonal and clinical alterations in women with IH, there is also evidence suggesting a link between hyperandrogenism and metabolic disturbances, 47 even in non-obese women has been previously documented. 48  found that women with IH had a higher SMD of TC and a lower SMD of HDL-C than healthy controls, whereas there were no significant differences in the SMDs of TG and LDL-C between the study groups.
Since remarkable number of women with PCOS are characterized by biochemical findings of hyperandrogenism, 16

| CON CLUS ION
The results of this study revealed that the majority of lipid parameters were not significantly different in women with IH, compared to those of the healthy and PCOS controls. Our study findings can give clinicians some clues to avoid unnecessary screening tests for lipid parameters in woman with IH without having additional risk factors.
However, further comprehensive longitudinal studies of women in different ethnicities and subgroups of BMI that mainly focus on lipid parameters with well-defined populations are still needed for the precise evaluation of the associations of lipid parameters in patients with IH, compared to those of healthy and PCOS groups.

AUTH O R CO NTR I B UTI O N S
M.A and L.M were involved in search in databases, quality assessment, study selection, data extraction, and critical discussion. M.A also contributed to design of the study, supervision, search in databases, quality assessment, study selection, data extraction, critical discussion, data analysis and interpretation, manuscript drafting, and revising the manuscrIpt. A.S contributed to statistical analysis, interpreting data, and manuscript drafting. F.R.T was involved in study design, quality assessment, data analysis, revising manuscript, and critical discussion.

FU N D I N G I N FO R M ATI O N
None.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

E TH I C A L A PPROVA L
Authors declare human ethics approval was not needed for this study.