Gynecological Endocrinology ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/igye20 Association between hyperandrogenism and adverse pregnancy outcomes in patients with different polycystic ovary syndrome phenotypes undergoing in vitro fertilization/intracytoplasmic sperm injection: a systematic review and metaanalysis Linna Ma, Yurong Cao, Yue Ma & Jun Zhai To cite this article: Linna Ma, Yurong Cao, Yue Ma & Jun Zhai (2021): Association between hyperandrogenism and adverse pregnancy outcomes in patients with different polycystic ovary syndrome phenotypes undergoing in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis, Gynecological Endocrinology, DOI: 10.1080/09513590.2021.1897096 To link to this article: https://doi.org/10.1080/09513590.2021.1897096 View supplementary material Published online: 11 Mar 2021 Submit your article to this journal Article views: 71 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=igye20 GYNECOLOGICAL ENDOCRINOLOGY https://doi.org/10.1080/09513590.2021.1897096 ORIGINAL ARTICLE Association between hyperandrogenism and adverse pregnancy outcomes in patients with different polycystic ovary syndrome phenotypes undergoing in vitro fertilization/intracytoplasmic sperm injection: a systematic review and meta-analysis Linna Maa,b, Yurong Caoa,b, Yue Maa,b and Jun Zhaia,b a Center for Reproductive Medicine, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China; bHenan Key Laboratory of Reproduction and Genetics, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China ABSTRACT ARTICLE HISTORY Objective: To study the association between hyperandrogenism (HA) and adverse pregnancy outcomes in patients with different polycystic ovary syndrome phenotypes undergoing in vitro fertilization (IVF)/ intracytoplasmic sperm injection (ICSI) Methods: We reviewed all eligible articles published up to October 2020 after searching in PubMed, Embase, Cochrane Library, Web of Science, Wanfang Data, and CNKI databases The primary outcomes were the clinical pregnancy rate (CPR), miscarriage rate (MR), and live birth rate (LBR), whereas the secondary outcomes were the number of retrieved oocytes and endometrial thickness Risk ratios (RRs) or mean differences with 95% confidence intervals (CIs) were calculated to estimate the HA impact on IVF/ ICSI outcomes in patients with polycystic ovary syndrome (PCOS) phenotypes Results: Of the 789 trials identified, nine retrospective studies involving 3037 patients with PCOS were included Compared to the PCOS group with normal androgen levels, the PCOS group with HA exhibited increased MR (RR: 1.56, 95% CI: 1.13, 2.16); the CPR (RR: 0.88, 95% CI: 0.77, 1.01) and LBR (RR: 0.79, 95% CI: 0.55, 1.11) were not significantly different between these groups Subgroup analysis revealed that the CPR was lower in the polycystic ovarian (PCO)-morphology þ HA þ oligo-anovulation (AO) group than in the PCO þ AO group (RR: 0.81, 95% CI: 0.67, 0.99) Among Asians, the PCOS/HA group had increased MR (RR: 1.56, 95% CI: 1.06, 2.31) and showed thinner endometrial thickness However, among Caucasians, no differences were observed between the two groups Conclusions: HA may have adverse effects on clinical pregnancy and miscarriage outcomes in different PCOS phenotypes, particularly among Asians Received 12 November 2020 Revised February 2021 Accepted 25 February 2021 Published online 11 March 2021 Introduction Polycystic ovary syndrome (PCOS) is the most common endocrine and metabolic disorder among women PCOS is characterized by hyperandrogenism (HA), presence of polycystic ovaries, and anovulation following the exclusion of other endocrine causes such as hyperprolactinemia and nonclassical congenital adrenal hyperplasia [1,2] The prevalence of PCOS in reproductive-age women ranges from to 10% [3,4] As women with PCOS are infertile, they often receive assisted reproductive technology (ART) fertility treatment [5] While the etiology of PCOS remains largely unknown, PCOS likely represents a complex multigenic disease under environmental influences [6] HA is considered as a PCOS phenotype according to the diagnostic criteria from the European Society of Human Reproduction and Embryology/American Society for Reproductive Medicine consensus meeting held in Rotterdam, the National Institutes of Health criteria, and the Androgen Excess Society criteria The prevalence of hyperandrogenemia is fairly high among patients with PCOS, ranging from 70 to 78% [7] There has been increasing controversy regarding the potential adverse effects of HA on pregnancy outcomes after infertility treatment [8,9] Whether antiandrogen pretreatment is beneficial to PCOS patients with HA who have received ART treatment is currently unknown HA reportedly does not affect the pregnancy outcomes in women with PCOS andmore oocytes can be obtained from patients with HA during controlled ovarian stimulation, which is more conducive to the acquisition of highquality embryos [8] Nevertheless, HA adversely affects the clinical pregnancy rate (CPR) during ART treatment [10] This study aimed to study the association between HA and adverse pregnancy outcomes in patients with PCOS via a systematic review and meta-analysis to provide guidance for HA pretreatment prior to in vitro fertilization (IVF) in patients with PCOS Methods Search strategy Articles published from January 1950 until May 2020 were searched in PubMed, Embase, Cochrane Library, Web of Science, Wanfang Data, and CNKI databases to identify eligible studies from English- and Chinese-language journals Electronic database searches were supplemented with manual search and literature CONTACT Jun Zhai bestzhai2005@163.com No.1, Jianshe East Road, Zhengzhou 450052, China Supplemental data for this article can be accessed here ß 2021 Informa UK Limited, trading as Taylor & Francis Group KEYWORDS Hyperandrogenism; in vitro fertilization/intracytoplasmic sperm injection; polycystic ovary syndrome; pregnancy outcome; meta-analysis China Belgium China Israel China China Turkey Li [13] De Vos [14] Yang [15] Eldar-Geva [16] Wu [17] Luo [18] Selc¸uk [19] Eftekhar [12] Country Size Age BMI Total T (ng/ml) No of retrieved oocytes Endometrial thickness on ET day Clinical pregnancy rate Miscarriage rate Live birth rate Size Age BMI Iran Eftekhar [12] Ramezanali [10] Iran Ramezanali [10] Study Table Basic characteristics of included studies Biochemical HA Clinical HA Treatment protocol PCO ỵ HA 83 30.8 4.6 27.1 ± 3.9 ± 0.3 11.2 ± 6.3 9.6 ± 1.6 14/38 3/14 11/14 66 28.41 ± 4.73 28.69 3.30 AO ỵ HA 103 32.2 5.33 27.5 ± 4.9 1.1 ± 0.5 11.0 ± 7.3 9.8 ± 1.6 19/72 0/19 19/19 56 29.52 ± 4.54 28.42 ± 2.95 PCO ỵ AO 32 32.4 4.4 26.4 4.0 0.3 ± 0.1 11.2 ± 7.1 9.9 ± 1.2 8/15 1/8 7/8 102 27.33 ± 5.16 28.87 ± 4.08 (continued) Long agonist GnRH antagonist Serum testosterone !1.97 noml/l Ferriman Gallwey score >8 Long agonist Testosterone >3 nmol/l, free androgen index >4.5 and/or androstenedione >12 nmol/l Testosterone >60 ng/dl and/or serum D4A >3.8 ng/ml GnRH antagonist testosterone level !2.2 nmol/l or androstenedione level !12 nmol/l Total serum testosterone >52 ng/ Ferriman–Gallwey score >8 GnRH antagonist dl and calculated free testosterone >0.64 ng/dl Severe endometriosis, hydrosalpinx, Total testosterone concentration Ferriman–Gallwey score >8 Long agonist uterine factor, severe male factor, age >0.5 ng/ml and/orfree or antagonist ! 40, DOR (AMH < ng/ml, FSH > testosterone >3.5 pg/ml 12 IU/l), smokers, diabetic Severe male factor, severe Total testosterone Antagonist endometriosis, hydrosalpinx, history levels above the 95th of any endocrine disorder except percentile (0.481 ng/ml) of the PCOS, and history of surgical level found in the group hysteroscopy Long agonist Exclusion criteria PCOS was based on the Rotterdam criteria Aged 18–41 years old with Congenital adrenal hyperplasia, ultrasound diagnosis of polycystic Cushing’s syndrome, androgenicovary morphology secreting tumors, oocyte donation, PGT, severe male factor Aged 20–35 years who FSH 12 IU/l, Overweight/obese, endometriosis, POF, who were previous ovarian surgical history, undergoing first IVF cycle hydrosalping, severe oligoasthenospermia or azoospermia, and systemic diseases such as diabetes mellitus and hypo-or hyperthyroidism Aged 20–39 years, both ovaries present, no previous ovarian operation, no current hormone therapy PCOS based on Rotterdam criteria Other hyperandrogen diseases PCOS based on Rotterdam criteria Hyperprolactinemia, thyroid disease, congenital adrenal hyperplasia, Cushing syndrome, androgensecreting tumor, 21-hydroxylase deficiency atypical adrenal hyperplasia, genetic diseases; repeated abortions, endometriosis, pelvic anatomical factors, reproductive tract infections, etc Rotterdam criteria in each group Age ! 40 years; BMI > 35 kg/m2; FSH > 12 mIU/ml; endometriosis; recurrent implantation failure, systemic illness, endocrine disorders PCO ỵ AO ỵ HA 168 32.1 ± 4.7 27.6 ± 3.9 1.3 ± 0.6 11.3 ± 6.7 9.9 ± 1.3 25/77 4/25 21/25 127 28.56 ± 4.83 29.15 ± 3.11 Aged 18–40 years, PCOS was based on the Rotterdam criteria Diagnosed with PCOS who underwent the first IVF/ICSI cycle Inclusion criteria L MA ET AL Wu [17] Eldar-Geva [16] Yang [15] De Vos [14] Li [13] Study Country Size Age BMI T (nmol/l) No of retrieved oocytes Endometrial thickness Clinical pregnancy rate Miscarriage rate Live birth rate Size Age BMI T (nmol/l) No of retrieved oocytes Endometrial thickness Clinical pregnancy rate Miscarriage rate Size Age BMI Basal T (ng/ml) No of retrieved oocytes Endometrial thickness Clinical pregnancy rate Miscarriage rate Live birth rate Size Age BMI Total T (g/l) No of retrieved oocytes Endometrial thickness on hCG day Clinical pregnancy rate Miscarriage rate Live birth rate Size Age BMI Basal T (ng/ml) No of retrieved oocytes Endometrial thickness on hCGday Clinical pregnancy rate Miscarriage rate Live birth rate Total T (ng/ml) No of retrieved oocytes Endometrial thickness on ET day Clinical pregnancy rate Miscarriage rate Live birth rate Table Continued Inclusion criteria HA PCOS 538 30.29 ± 3.49 23.24 ± 3.73 4.72 ± 2.75 13.19 ± 6.20 9.83 ± 1.81 301/536 61/301 5/15 NA PCOS 181 30.40 ± 4.02 22.20 ± 3.17 1.43 ± 0.52 12.77 ± 5.62 10.18 ± 1.55 77/180 7/77 (continued) 4/10 PCO ỵ HA 19 27.9 4.1 27.7 ± 6.1 2.9 ± 1.1 19.2 ± 8.5 10/50 11/54 HA PCOS 244 29.1 ± 3.4 24.4 ± 4.3 1.6 ± 2.0 16.5 ± 9.6 10.3 ± 1.8 71/162 24/71 47/71 PCO ỵ AO 175 29.9 (3.7) 22.7 (2027) 0.32 (0.250.41) 12 (719) PCO ỵ AO 117 29.4 3.6 22.0 ± 3.6 0.4 ± 0.2 13.8 ± 6.7 11.7 ± 2.3 60/126 9/60 0.48 ± 0.25 20.35 ± 9.60 9.27 ± 1.00 26/102 5/26 Treatment protocol 59/156 NA PCOS 330 29.4 ± 3.5 23.9 ± 4.1 0.9 ± 0.3 15.1 ± 8.2 10.7 ± 1.7 109/236 22/109 87/109 PCO þ AO 10 29.8 ± 2.1 27.2 ± 7.5 1.6 0.4 21.0 2.3 PCO ỵ HA 54 29.1 (3.6) 25.6 (2331) 0.56 (0.460.74) 13.5 (921) PCO ỵ AO þ HA 66 29.3 (3.8) 27.0 (24–33) 0.55 (0.46–0.77) 11 (7–18) 0.69 ± 0.39 19.85 ± 8.79 9.18 ± 0.98 12/66 0/12 Clinical HA PCO ỵ HA 18 28.1 4.3 20.4 ± 2.3 1.4 ± 1.0 14.9 ± 6.2 11.5 ± 2.1 9/18 2/9 0.94 ± 0.33 18.73 ± 7.74 9.13 0.94 17/56 2/17 Biochemical HA PCO ỵ AO ỵ HA 54 30.3 3.9 22.6 3.3 1.2 ± 0.6 12.2 ± 7.1 11.6 ± 2.6 25/58 5/25 1.20 ± 0.45 22.79 ± 11.03 9.35 ± 1.22 35/127 6/35 Exclusion criteria GYNECOLOGICAL ENDOCRINOLOGY 17/55 PCO ỵ AO 55 29.01 ± 3.84 24.28 ± 4.21 158/316 19/158 Clinical HA 64/77 PCO ỵ AO 316 31.2 2.7 21.4 ± 2.2 2.1 ± 1.4 16.0 ± 8.0 L MA ET AL Treatment protocol 14/38 AO ỵ HA 38 30.32 ± 4.07 24.53 ± 2.96 Biochemical HA Inclusion and exclusion criteria The following studies were included in the analysis: (1) observational studies such as cohort, case–control, and cross-sectional studies; (2) studies investigating PCOS patients with or without HA or those evaluating PCOS patients with different phenotypes, the diagnosis of which was in accordance with the 2003 Rotterdam criteria; and (3) studies that had CPR, miscarriage rate (MR), live birth rate (LBR), number of retrieved oocytes, and endometrial thickness as the primary or secondary outcomes However, the following studies were excluded: (1) repeated studies, studies with incomplete data, conference abstracts without detailed contents, and literature reviews; (2) studies with missing, inexact, or contradictory experimental results; (3) studies that did not specify the diagnostic criteria for PCOS; (4) studies that did not compare the pregnancy outcomes between the PCOS patients with and without HA or among patients with different PCOS phenotypes; and (5) studies that did not involve outcomes Two investigators filtered the articles, extracted the data, and independently conducted quality assessment Any disagreements were resolved through discussion or by a third investigator until a consensus was reached Data, including general characteristics, type of studies, participants, factors, and outcomes, were extracted according to a predesigned table PCOS was diagnosed based on different classifications Cases of polycystic ovarian morphology (PCO) ỵ oligo-anovulation (AO) ỵ HA, PCO ỵ HA, and HA ỵ AO were included into the HA group, whereas those with PCO ỵ AO were included into the normal androgen level (NA) group The Newcastle–Ottawa Scale was used to assess study quality [11] Scores were reported as total values of 29/85 5/29 227/301 PCO ỵ AO ỵ HA 85 30.7 ± 2.4 21.6 ± 2.1 4.5 ± 1.4 15.4 ± 5.2 Literature filtering, data extraction, and quality assessment Inclusion criteria Exclusion criteria review Our search strategy based on the following key terms and Medical Subject Headings was employed: ((hyperandrogenism or hyperandrogenic), (polycystic ovary syndrome, PCOS, or different polycystic ovary syndrome phenotypes), and (pregnancy, infertility, IVF, assisted reproductive technology or ART)) In all searches, limitations were set for human characteristics and clinical trials Selc¸uk [19] Size Age BMI T No of retrieved oocytes Endometrial thickness Clinical pregnancy rate Miscarriage rate Live birth rate Size Age BMI T (nmol/l) No of retrieved oocytes Endometrial thickness Clinical pregnancy rate Miscarriage rate Live birth rate Live birth rate Luo [18] Study Table Continued Country Statistical analysis Meta-analysis was performed using Review Manager version 5.3 (Cochrane Collaboration, Copenhagen, Denmark) and Stata version 12.0 (StataCorp, College Station, TX, USA) Dichotomous variables were expressed as risk ratios (RRs) with 95% confidence intervals (95% CIs), whereas continuous variables were presented as mean differences (MDs) with 95% CIs The enrolled articles were examined for heterogeneity using the v2 test and I2 Studies were considered heterogeneous when p and I2 ! 50%; when p > and I2 < 50%, studies were regarded to show no heterogeneity when no heterogeneity among various studies was detected, meta-analyses were performed using a fixed-effects model Otherwise, meta-analyses were conducted using a random-effects model, the source of heterogeneity was further analyzed, and possible confounding factors were explored in subgroup analysis Differences with p < 05 were considered statistically significant Publication bias was assessed using Begg’s GYNECOLOGICAL ENDOCRINOLOGY Figure Effect of hyperandrogenism on IVF/ICSI outcomes (A) clinical pregnancy rate, (B) miscarriage rate, and (C) live birth rate and Egger’s tests, with values of p > 05 indicating no publication bias Results Literature search results Overall, 789 articles were initially searched After reading their titles and abstracts, 746 articles were removed Subsequently, 43 articles were further evaluated by full-text reading After screening, nine retrospective articles were ultimately included in the meta-analysis [10,12–19] The screening flowchart and results are presented in Supplemental Figure Characteristics and quality assessment of the included studies Overall, the nine included articles were cohort studies involving 3037 patients; of these patients, 1719 and 1318 were assigned to the HA gand NA group All studies reported the average age and body mass index of patients, and only two studies did not indicate the serum androgen levels [18,19] The age and BMI of all patients were comparable The basic characteristics and quality assessment of the included studies are shown in Table and Supplemental Table 1, respectively Primary outcomes Clinical pregnancy rate Eight articles involving a total of 2388 patients with PCOS (1348 and 1040 patients in the PCOS/HA and PCOS/NA groups, respectively) reported the CPR [10,12,13,15–19] Meta-analysis was conducted using a random-effects model because of the heterogeneity among various studies (p ¼ 008, I2 ¼ 63%) The results indicated an insignificant difference in the CPR between the PCOS/HA and PCOS/NA groups (RR: 0.94, 95% CI: 0.77, 1.15) (Figure 1(A)) Miscarriage rate Six articles involving a total of 995 patients with PCOS (557 and 438 patients in the PCOS/HA and PCOS/NA groups, respectively) evaluated the MR [10,12,13,15,17,18] Asno heterogeneity was observed (p ¼ 53, I2 ¼ 0.0%), a fixed-effects model was used for meta-analysis The results revealed a statistically significant difference in the MR between the PCOS/HA and PCOS/NA groups (RR: 1.56, 95% CI: 1.13, 2.16) (Figure 1(B)) Live birth rate Four articles involving a total of 884 patients (534 and 350 patients in the PCOS/HA and PCOS/NA group, respectively) investigated the LBR [10,14,15,17] Meta-analysis was performed L MA ET AL Table Meta-analysis of outcomes according to different phenotype, ethnicity Outcomes CPR Factor Phenotype Ethnicity Asians Caucasian Overall MR Ethnicity Asians Caucasian Overall LBR Ethnicity Asians Caucasian Overall Model used 0.81 (0.67, 0.99) 28 Fixed 0.90 (0.52, 1.58) 66 Random 0.85 (0.59, 1.24) Fixed 1.05 (0.74, 1.49) Fixed 1.08 (0.78, 1.50) 21 Fixed 1.08 (0.47, 2.48) 73 Random 0.88 (0.77, 1.01) 0.86 (0.73, 1.00) 16 36 Fixed Fixed 0.94 (0.72, 1.24) 27 Fixed 0.94 (0.77, 1.15) 63 Random 1.22 (0.71, 2.11) Fixed 0.45 (0.12, 1.68) Fixed 0.88 (0.32, 2.42) Fixed 2.41 (0.67, 8.63) Fixed 1.19 (0.48, 2.97) Fixed 0.61 (0.02, 19.34) 64 Random 1.37 (0.96, 1.95) 1.56 (1.06, 2.31) 0 Fixed Fixed 0.71 (0.29, 1.76) Fixed 1.56 (1.13, 2.16) Fixed 0.74 (0.32, 1.67) 85 Random 1.17 (0.87, 1.58) NA Fixed 0.71 (0.35, 1.43) 75 Random 0.85 (0.70, 1.02) NA Fixed 1.05 (0.72, 1.52) Fixed 1.27 (0.95, 1.70) NA Fixed 0.79 (0.56, 1.11) 0.83 (0.69, 1.00) 76 NA Random Random 0.74 (0.31, 1.75) 91 Random 0.84 (0.69, 1.12) 67 Random Phenotype (n) No of studies PCO ỵ AO ỵ HA (434) AO ỵ HA 197 PCO ỵ HA 186 PCO ỵ AO ỵ HA 295 PCO ỵ AO ỵ HA 349 AO ỵ HA (159) All HA (323) HA (489) HA (1348) PCO ỵ AO ỵ HA 434 AO ỵ HA 159 PCO ỵ HA (167) PCO þ AO þ HA (295) PCO þ AO þ HA 349 AO ỵ HA 159 All HA 134 HA 122 HA 557 PCO ỵ AO ỵ HA (234) AO ỵ HA (103) PCO ỵ HA (137) PCO ỵ AO ỵ HA (168) PCO ỵ AO ỵ HA (234) AO ỵ HA （103） All HA （71） HA （162） HA (534) PCO þ AO （567） PCO þ AO (189) PCO þ AO 261 AO ỵ HA 159 PCO ỵ HA 167 PCO þ HA (149) NA (678) NA (182) NA (1040) PCO ỵ AO 567 PCO ỵ AO (134) PCO ỵ AO (251) AO ỵ HA 159 PCO ỵ HA 167 PCO ỵ HA 149 NA 327 NA 34 NA (438) PCO ỵ AO (207) PCO ỵ AO (32) PCO ỵ AO (207) AO ỵ HA (103) PCO ỵ HA 137 PCO ỵ HA 83 NA 109 NA 164 NA 350 Risk ratio (95% CI) I2 (%) Phenotype (n) vs Note The results formatted in bold in the table are statistically significant Secondary outcomes individuals in the HA and NA groups, respectively) compared the number of retrieved oocytes between the HA and NA groups [15,17] No significant difference in the number of retrieved oocytes was identified between the two groups (MD: 0.71, 95% CI: À0.10, 1.53, I2 ¼ 14%) Furthermore, four studies included different phenotypes, with the number of retrieved oocytes being comparable [10,12,13,18] (Supplemental Figure 2, Supplemental Table 2) Number of retrieved oocytes Because of the different grouping methods that were used by the included studies, continuous variables could not be pooled Two articles involving a total of 1293 individuals (782 and 511 Endometrial thickness Two articles assessed endometrial thickness in the HA (782 patients) and NA groups (511 patients) on the day of human using a random-effects model due to the heterogeneity among various studies (p ¼ 03, I2 ¼ 67%) The results failed to show a significant difference between the PCOS/HA and PCOS/NA groups (RR: 0.84, 95% CI: 0.69, 1.12) (Figure 1(C)) GYNECOLOGICAL ENDOCRINOLOGY chorionic gonadotropin (hCG) administration or embryo transfer (ET) [15,17] The meta-analysis indicated that the endometrium was thinner in the HA group in the HA group than the in NA group, with the difference being statistically significant (MD: À0.37, 95% CI: 0.57, À0.17, I2 ¼ 0%) (Supplemental Figure 3, Supplemental Table 3) Sensitivity analysis Sensitivity analysis was conducted to identify the heterogeneity source After removing the included studies one by one, we found that the Man W study had a considerable influence on the heterogeneity of results (Supplemental Figure 4) After ruling out the study, the results revealed that the CPR did not differ between the two groups (RR: 0.88, 95% CI: 0.77, 1.01) Subgroup analysis Subgroup analysis according to different PCOS phenotypes and ethnicity was performed The subgroup analysis results are presented in Table The CPR was lower in the PCO ỵ AO þ HA than in the PCO þ AO group (RR: 0.81, 95% CI: 0.67, 0.99); nevertheless, no difference in other phenotypes was observed Concerning miscarriage, the results failed to show any difference in the MR between the subgroups Furthermore, there was no difference in the LBR between the subgroups Regarding to ethnicity, among Asians, the PCOS/HA group had an increased MR (RR: 1.56, 95% CI: 1.06, 2.31); additionally, the CPR (RR: 0.86, 95% CI: 0.73, 1.00) and LBR (RR: 0.83, 95% CI: 0.69, 1.00) were decreased, albeit without statistically significant difference However, no difference was observed among Caucasians Publication bias Begg’s and Egger’s tests were used to assess publication bias (Supplemental Figure 4) and showed values of p > 05 for CPR, MR, and LBR The results indicated that there was no publication bias Discussion This is the first meta-analysis that evaluated the HA effects of on the outcomes of IVF/intracytoplasmic sperm injection (ICSI) in patients with different PCOS phenotypes Nine articles were included in this t study In this meta-analysis, MR was significantly different between the PCOS/HA and PCOS/NA groups Nonetheless, the results of subgroup analysis indicated no difference in the MR among the phenotypes We speculate that these results may be attributable to the small sample size While we did not conclude that there were significant differences in the CPR and LBR between the PCOS/HA and PCOS/NA groups, our subgroup analysis revealed that the CPR decreased in the PCO ỵ HA ỵ AO group, compared to that in the PCO ỵ AO group (RR: 0.81, 95% CI: 0.67, 0.99) When other phenotypes were compared, no difference in the CPR was identified We analyzed the included studies from the literature and observed that patients with PCO ỵ HA ỵ AO had higher serum androgen levels than those with other phenotypes Yilmaz et al reported that compared to patients with other subtypes, those with PCO ỵ HA ỵ AO exhibited the highest incidence of metabolic syndrome [20] Different PCOS phenotypes not represent the same metabolic disorder Batcheller et al observed that the CPR was significantly lower in the PCOS/HA group than in the PCOS/NA group (39.5 vs 69.6%; odds ratio: 0.306, 95% CI: 0.12, 0.85, p ¼ 034) [21] The subgroup meta-analysis revealed that PCOS/HA had a negative effect on IVF/ICSI outcomes among Asians; however, no effect was detected among Caucasians The ethnicity impact on pregnancy outcomes remains unknown A possible reason is that the effects of ethnicity on endometrial thickness and endometrial receptivity in patients with PCOS vary across different races We performed a meta-analysis of endometrial thickness and showed that among Chinese individuals, the endometrium was thinner in the HA group than in the NA group on the day of hCG administration or ET Nevertheless, the number of retrieved oocytes did not differ among phenotypes HA above a certain level may reduce the clinical pregnancy outcomes of IVF/ICSI A possible mechanism by which HA influences the pregnancy outcomes of IVF/ICSI is its effects on endometrial receptivity and embryo implantation PCOS is reportedly an independent factor for recurrent pregnancy loss [22,23] Androgen levels are higher in women who had experienced recurrent miscarriages than in normal fertile controls The prevalence of hyperandrogenemia in recurrent miscarriages was approximately 11% [24] HA may have adverse effects on the endometrium Palomba et al revealed that Testosterone reportedly is a novel negative regulator of endometrial HOXA10 expression and also impairs the LIFSTAT3 pathway, and reduces the uterine avb3 integrin and glycodelin levels, which may contribute to poor endometrial function[25] HA affects the utilization of glucose by endometrial stromal cells, which is necessary for endometrial decidualization and embryo implantation [26] Subendometrial and endometrial blood flow (vascularization index) is considerably impaired in PCOS/HA women, compared to that in PCOS/NA patients (3.17 vs 1.59% and 1.11 vs 0.57%, respectively) [27] High androgen levels may have a negative impact on oocyte competence Hyperandrogenism may cause abnormal gene expression, which are essential for ovarian growth and cumulus expansion, such as IGF2R, TXNIP and HAS2 [28] Additionally, HA may have adverse effects on maternal and neonatal health Patients with PCOS, particularly the HA ỵ AO ỵ PCO phenotype, have a higher risk of developing gestational diabetes mellitus, pregnancy-induced hypertension, and premature delivery than the control group, leading to a lower LBR [29] Hyperandrogenism may contribute to the increased risk of obstetric and neonatal complications [30] This study had some limitations First, a limited number of cohort studies were included Additionally, the grouping methods of articles included in this meta-analysis were not identical Randomized controlled trials are required to determine whether PCOS patients with HA would benefit more from HA pretreatment prior to IVF or ICSI Conclusion This meta-analysis analyzed the influence of hyperandrogenemia on the pregnancy outcomes of IVF/ICSI in patients with PCOS Our results indicated that hyperandrogenemia might increase the MR in the PCOS/HA group Hyperandrogenemia had significant adverse effects on IVF/ICSI outcomes among Asians Ethical approval This article did not contain any studies with human participants performed by any of the authors 8 L MA ET AL Author contributions [13] Linna Ma was responsible for the writing of the paper, Wenrong Dai and Yunrong Cao were responsible for the collection of materials They all undertook the screening and sorting of data, Professor Zhai Jun gave guidance and modification of the article [14] [15] Disclosure statement No potential conflict of interest was reported by the author(s) Funding This research was financially supported by the National Natural Science Foundation of China [82071649] [16] [17] [18] [19] References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] Escobar-Morreale Hector F Polycystic ovary syndrome: definition, aetiology, diagnosis and treatment Nat Rev Endocrinol 2018;14(5): 270–284 Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group Revised consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome Fertil Steril 2003;81: 19–25 Bozdag G, Mumusoglu S, Zengin D, et al The prevalence and phenotypic features of polycystic ovary syndrome: a systematic review and meta-analysis Hum Reprod 2016;31(12):2841–2855 Li R, Zhang Q, Yang D, et al Prevalence of polycystics ovary syndrome in women in China: a large community-based study Hum Reprod 2013;28(9):2562–2569 Legro RS, Arslanian SA, Ehrmann DA, et al Diagnosis and treatment of polycystic ovary syndrome: an Endocrine Society Clinical Practice Guideline J Clin Endocrinol Metab 2013;98(12):4565–4592 Jayasena CN, Franks S The management of patients with polycystic ovary syndrome Nat Rev Endocrinol 2014;10(10):624–636 Kavardzhikova S, Pechlivanov B Clinical, hormonal and metabolic characteristics of different phenotypes of polycystic ovary syndrome, in Bulgarian population Akush Ginekol 2010;49(4):32–37 Gaddas M, Chaouache N, Ajina M, et al Does hyperandrogenic statute in the polycystic ovary syndrome constitute an obstacle to the success of in vitro fertilization? J Gynecol Obstet Biol Reprod 2016; 45(9):1091–1098 Cela V, Obino MER, Alberga Y, et al Ovarian response to controlled ovarian stimulation in women with different polycystic ovary syndrome phenotypes Gynecol Endocrinol 2018;34(6):518–523 Ramezanali F, Ashrafi M, Hemat M, et al Assisted reproductive outcomes in women with different polycystic ovary syndrome phenotypes: the predictive value of anti-M€ ullerian hormone Reprod Biomed Online 2016;32(5):503–512 Hinneburg I ROBINS-1: a tool for asssessing risk of bias in nonrandomised studies of interventions Med Monatsschr Pharm 2017; 40(4):175–177 Eftekhar M, Mirhashemi ES, Tabibnejad N Assisted reproductive outcomes in women with different polycystic ovary syndrome phenotypes Int J Gynecol Obstet 2019;144(2):147–152 [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] Li M, Song J, Chen SL, et al Association of the clinical characteristics and the IVF-ET outcome in infertile women with polycystic ovarian syndrome of different subtypes Nan Fang Yi Ke Da Xue Xue Bao 2009;29:224–227 De Vos M, Pareyn S, Drakopoulos P, et al Cumulative live birth rates after IVF in patients with polycystic ovaries: phenotype matters Reprod Biomed Online 2018;37(2):163–171 Yang W, Yang R, Yang S, et al Infertile polycystic ovary syndrome patients undergoing in vitro fertilization with the gonadotropinreleasing hormone-antagonist protocol: role of hyperandrogenism Gynecol Endocrinol 2018;34(8):715–718 Eldar-Geva T, Margalioth EJ, Gal M, et al Serum anti-Mullerian hormone levels during controlled ovarian hyperstimulation in women with polycystic ovaries with and without hyperandrogenism Hum Reprod 2005;20(7):1814–1819 Wu M, Xiao G, Yang J, et al J Effect of IVF-ET outcome on different types of PCOS patients treated with GnRH agonist long protocol Guangdong Med J 2018;39(01):103–106 Luo L, Liu J, Li J, et al Analysis of different subtypes of polycystic ovary syndrome and pregnancy outcome after in-vitro fertilization and embryo transfer J Prac Obste Gynecol 2010;26(10):787–790 € Selc¸uk S, Ozkaya E, Eser A, et al Characteristics and outcomes of in vitro fertilization in different phenotypes of polycystic ovary syndrome Turk J Obstet Gynecol 2016;13(1):1–6 Yilmaz M, Isaoglu U, Delibas IB, et al Anthropometric, clinical and laboratory comparison of four phenotypes of polycystic ovary syndrome based on Rotterdam criteria J Obstet Gynaecol Res 2011; 37(8):1020–1026 Batcheller AE, Rodriguez-Peterson KA, Martinez AM, et al Hyperandrogenic Polycystic Ovary Syndrome (PCOS) patients have reduced pregnancy rates compared to non-hyperandrogenic PCOS patients following in vitro fertilization (IVF) Reproductive Sciences 2013;20(3):177A–178A Banu J, Fatima P, Sultana P, et al Association of infertile patients having polycystic ovarian syndrome with recurrent miscarriage Mymensingh Med J 2014;23(4):770–773 Cocksedge KA, Saravelos SH, Metwally M, et al How common is polycystic ovary syndrome in recurrent miscarriage? Reprod Biomed Online 2009;19(4):572–576 Cocksedge KA, Saravelos SH, Wang Q, et al Does free androgen index predict subsequent pregnancy outcome in women with recurrent miscarriage? Hum Reprod 2008;23(4):797–802 Palomba S, Piltonen TT, Giudice LC Endometrial function in women with polycystic ovary syndrome: a comprehensive review Hum Reprod Update 2020; Rosas C, Or ostica L, Poblete C, et al Hyperandrogenism decreases GRP78 protein level and glucose uptake in human endometrial stromal cells Reprod Sci 2016;23(6):761–770 Lam P, Johnson I, Raine-Fenning N Endometrial blood flow is impaired in women with polycystic ovarian syndrome who are clinically hyperandrogenic Ultrasound Obstet Gynecol 2009;34(3): 326–334 Palomba S, Daolio J, La Sala GB Oocyte competence in women with polycystic ovary syndrome Trends Endocrinol Metab 2017;28(3): 186–198 Foroozanfard F, Asemi Z, Bazarganipour F, et al Comparing pregnancy, childbirth, and neonatal outcomes in women with different phenotypes of polycystic ovary syndrome and healthy women: a prospective cohort study Gynecol Endocrinol 2020;36(1):61–65 Palomba S, de Wilde Marlieke A, Angela F, et al Pregnancy complications in women with polycystic ovary syndrome Hum Reprod Update 2015;21(5):575–592 ... China ABSTRACT ARTICLE HISTORY Objective: To study the association between hyperandrogenism (HA) and adverse pregnancy outcomes in patients with different polycystic ovary syndrome phenotypes undergoing. .. ENDOCRINOLOGY https://doi.org/10.1080/09513590.2021.1897096 ORIGINAL ARTICLE Association between hyperandrogenism and adverse pregnancy outcomes in patients with different polycystic ovary syndrome. .. not compare the pregnancy outcomes between the PCOS patients with and without HA or among patients with different PCOS phenotypes; and (5) studies that did not involve outcomes Two investigators