Determining risk factors predicting miscarriage among couples undergoing assisted reproductive treatment: a systematic review.

This systematic review was conducted in 2025 in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, version 2020 [ 34 ]. This study was approved by the Ethics Committee for Biomedical Research of Iran University of Medical Sciences (IR.IUMS.REC.1402.964). To obtain the best and most relevant articles, the following study inclusion and exclusion criteria were considered: Studies that included a variety of risk factors related to a miscarriage (first and second trimester). The study population was couples with primary or secondary infertility and undergoing treatment with one of the ART methods. Articles were considered in English and with full access. Studies that included a variety of risk factors related to a miscarriage (first and second trimester). The study population was couples with primary or secondary infertility and undergoing treatment with one of the ART methods. Articles were considered in English and with full access. Reviews, letters, editorials, short reports, book chapters, conference, policy briefs, protocols and articles where full text are not available. Studies in which the study population and type of infertility (primary or secondary) were not specified. Studies that exclusively examined miscarriage in non-infertile individuals or studies that focused on other aspects of infertility such as screening, effects, complications, and the like. Studies that lacked clear reporting on factors influencing miscarriage. Studies that only aimed to investigate the impact of specified risk factors. and studies that focused on specific patients (diabetes, tuberculosis, heart disease, etc.). Reviews, letters, editorials, short reports, book chapters, conference, policy briefs, protocols and articles where full text are not available. Studies in which the study population and type of infertility (primary or secondary) were not specified. Studies that exclusively examined miscarriage in non-infertile individuals or studies that focused on other aspects of infertility such as screening, effects, complications, and the like. Studies that lacked clear reporting on factors influencing miscarriage. Studies that only aimed to investigate the impact of specified risk factors. and studies that focused on specific patients (diabetes, tuberculosis, heart disease, etc.). A comprehensive search was conducted to obtain relevant studies in the databases PubMed, Scopus, Web of Science, Ovid, BSCO Host, IEEE, Embase, Proquest, and Cochrane Library between January 1, 2014 and February 1, 2025. Gray literature and Google Scholar search engine were also reviewed. In addition, to increase the comprehensiveness of the study, sources and citations in articles were tracked and the authors’ publications and academic records were reviewed to obtain additional articles. Studies were reviewed based on a search strategy including key concepts “Risk Factors”, “abortion”, “assisted reproductive technology”, “Prediction”, synonyms and various combinations of them with logical operators (AND, OR) (Supplementary 1). A manual search of the reference lists of the included studies was also conducted. In order to facilitate the elimination of duplicate studies and better resource management, EndNote version 9 was used. Then, the titles and abstracts of the obtained articles were screened independently by two researchers (SAFA, ML). In case of disagreement, an attempt was made to resolve the problem through group discussion first, and secondly, the opinion of a third person was used for the final conclusion. Data were collected using a data collection form in MS-Word software version 2019. Key data collected included author name, year of publication, country, research objective, study type, participants/sample size, ART methods, parameter characteristics (demographics, lifestyle, and medical history, clinical, treatment), and major findings. In this study, due to the increasing use of ART in infertility treatment, the role and impact of demographic and lifestyle factors were also investigated in various studies. Factors effective in predicting miscarriage such as clinical parameters and treatment were also investigated. In this study, due to the different types of studies, the CASP checklist was used to assess the quality of the study. This checklist had various types depending on the type of study and consisted of 10 to 12 questions, which were given one of 3 answers (yes, no, can’t tell) and at the end, the points were collected and the final score was announced. The JBI checklist was also used to assess the risk of bias. The assessment was conducted by two independent researchers (EM, MS). The checklist consisted of 8 to 12 questions, and the possible answers for each question included (4 modes: yes, no, unclear, and not applicable). In this study, the results were reported descriptively. In general, considering the progress of treatment in infertility, miscarriage depends on various risk factors and triggers according to the patient’s condition, which was attempted in this study, while identifying the risk factors, to provide an appropriate classification. The parameters identified in this study were classified into three main groups including demographic and lifestyle, clinical, and treatment, which makes it more transparent and understandable for people involved in the treatment process. Knowing and understanding these factors leads to better success and effectiveness of the treatment outcome. Considering that the study population was treated with different methods and there was a lot of heterogeneity between participants and studies, it was not possible to perform a meta-analysis.

In this study, after searching the databases, 612 articles were identified. After removing duplicates ( n  = 253), and the lack of relevance of the title and abstract to the research objectives ( n  = 295), the number of articles for access to the full text was 64 articles. Subsequently, 6 articles were excluded due to lack of access to the full text and 58 studies were fully reviewed. Based on the eligibility criteria, 41 studies were excluded and finally 17 studies were considered eligible for inclusion in the study. The diagram of the study selection process according to PRISMA is shown in Fig.  1 . Fig. 1 The PRISMA diagram The PRISMA diagram The quality and risk of bias for each study were assessed using the CASP and JBI checklists, focusing on key methodological aspects such as consistency between research components, data collection, and ethical approval. Studies were rated as “high”, “moderate”, or “low” based on a scoring system appropriate to each type of study. All studies were considered acceptable and included in the data collection (Supplementary 2 and 3). Summary of the main information collected from the included articles was shown in Table 1 . Accordingly, the most studies entered in 2022 ( n = 5) [ 35 – 39 ], 2023 ( n = 4) [ 16 , 20 , 40 – 42 ], and the contribution of the years 2014 [ 43 ], 2015 [ 44 ], 2016 [ 45 ], 2018 [ 46 ], 2020 [ 47 ], 2021 [ 48 ], 2024 [ 15 ] is one study. Table 1 The main characteristics of the included studies Author, Year (ref) Country study type Participants/ sample size ART methods Risk Factors (Baseline (demographics, lifestyle, and medical history), Clinical, Treatment , major findings Ashrafi et al. / 2014 [ 43 ] Iran cross-sectional 234 pregnant women with PCOS and 234 non-PCOS women with ART conception IVF/ICSI Baseline : Age, parity, irregular menses and hirsutism cases, menstrual irregularity, and menarche age. Clinical :serum triglyceride level, pregestational metformin use- No significant differences were found in terms of pre-pregnancy BMI and weight gain during pregnancy among groups. -Metformin consumption was associated with a reduction in GDM prevalence, significantly ( p  = 0.01). Groups were significantly different in terms of mean serum levels of triglyceride and cholesterol ( p  = 0.02 and p  = 0.04, respectively). Yang et al. /2015 [ 44 ] China Retrospective cohort Womenwho reached the ongoing pregnancy stage, 7,598 cases were divided into three groups: group 1, lack of pregnancy ( n  = 6,651); group 2, BP ( n  = 520); and group 3, SA ( n  = 427 IVF Baseline :Age, BMI. Clinical : basic FSH level and E2 level. Treatment : endometrium thickness on hCG administration, secondary infertility proportion Multivariate analysis between BP and SA groups showed that only the age ( P  = 0.037) and endometrium thickness on hCG administration day ( P  = 0.029) can result in the differences between BP and SA groups Wang et al. /2016 [ 45 ] China cross-sectional observational 751 Women with at least one spontaneous miscarriage(irrespective of conception method (natural or ART)) Not Mention Baseline :Age, Age of spouse, Menarche age, BMI, Education Clinical : Ovulation disorder, Uterine factors, Cervical factors, Fallopian tube factors, Pelvic cavity factors, Endocrine disorders, Immune factors, Chromosomal abnormalities, Reproductive tract infections, Male infertility. Higher education level and a pelvic cavity disorder were associated with a lower risk of RSM (OR = 0.27 and 0.46, respectively). - Late spontaneous miscarriages were more frequent in Women with RSM than in those with a SSM (31.5% vs. 14.2%, respectively) and were associated with a history of uterine cavity procedures (OR = 2.095) and cervical factors related to infertility (OR = 4.136, 95% CI: 1.012–16.90). Hu et al. /2018 [ 46 ] China A retrospective cohort ET cycles. 4165 cycles in Nanjing and 1320 in Changzhou IVF/ICSI Baseline :maternal age, maternal BMI (kg/m2), paternal BMI, and infertility type Clinical : COH protocol Treatment :Total Gn dose, fertilization methods, cycle type, no. of embryos transferred, cleavage-stage embryo or blastocyst, serum hCG level 14 days after transfer - Couples in ART pregnancies demonstrated a significantly increased risk of PL as maternal age (HR = 1.31, Ptrend < 0.001) grows. - In contrast to fresh cycle, women who received frozen cycle embryo had a significant increased risk of early miscarriage ( P  < 0.001), while frozen cycle was linked with lower risk of late miscarriage ( P  = 0.045). − 4 factors (maternal age, COH protocol, cycle type and serum hCG level 14 days after transfer) had an independent impact on miscarriage mainly before 12 weeks of gestational age. - Clinicians may make it better to evaluate a Women risk of PL based on the maternal age at the time of treatment, COH protocol, cycle type and serum hCG level 14 days after transfer and the gestational week of the fetus. Boynukalin et al. /2020 [ 47 ] Turkey retrospective cohort 707 FBTs after PGT-A ICSI Baseline :Maternal age, paternal age, BMI, infertility diagnose, number of previous attempts, number of previous miscarriage Treatment :Duration of stimulation, total Gn dosage used, E2 and P4 levels on trigger day, endometrial thickness, E2 and P4 levels 6 days before the FBT cycle, mitoscore, ICM score, trophectoderm score and day of embryo biopsy A high BMI, an embryo biopsy on day 6 and a high number of miscarriages negatively affect the live birth rate after single euploid FBT Li et al. /2021 [ 48 ] China retrospective cohort study Women successfully conceived with ART. 16,097 Women (2,378 had an abortion and 13,719 did not have one). Not Mention Baseline :female age, male age and female cause of infertility, BMI Clinical :FSH, E2, LH, AMH Treatment :controlled ovulation induction protocol, PRL, number of treatment cycles, number of transferred embryos, method of fertilization, embryo stage, number of fetuses and controlled ovulation induction protocol, dose of Gn -Found that when female age > 32 year (cut-off point), age and the risk of spontaneous abortion were positively correlated. -There was a linear positive correlation between AFC and live birth. -The older the male age, the higher the incidence of abortion. -Basal P, AMH and PRL in the abortion group were also significantly lower than those in the non-abortion group. -There was a statistical difference between the number of treatment cycles, embryos and fetuses at pregnancy diagnosis between the two groups Bilibio et al. / 2022 [ 35 ] Brazil prospective cohort 422 blastocysts from 135 Womenwho had undergone preimplantation genetic analysis after ICSI ICSI Baseline :maternal age, paternal age, Infertility time, repeat abortion, male factor, endometriosis, PCOS Clinical :AFC, tubal factor, Treatment :Ovarian reserve, and semen quality (Seminal collection type, Semen concentration, motility, preparation) cryopreserved oocyte, parameters on D1 (pro-nucleus, type of nuclear precursor corpuscles, polar corpuscles), D2 and D3 (number of blastomeres, embryonic classification), sperm with normal morphology, TE quality, and ICM quality -Analysis revealed maternal age and type C TE as the main risk factors for aneuploidy. -Poor ovarian reserve ( p   38 and > 36 years, respectively), AFC  38 year, type-C embryo with trophectoderm, which showed a positive predictive value of 88.6% and a specificity of 97.5%. Wang et al. / 2022 [ 38 ] China Retrospective cross-sectional 2709 first pregnancy ET cycles IVF/ ICSI Baseline :Female age, male age, female BMI, Male smoking, number of previous miscarriages, Adverse pregnancy history, and IUI unfertilized history. IVF failed ET history Clinical :AMH level, cycle type Treatment :ET stage, number of transferred embryos, thickness of endometrium, isolated tubal factor, isolated diminished ovarian reserve factor, Female chromosome, Male chromosome, Embryo transfer stage, Male factor. The EPL rate of the first-time pregnancies for infertility Women undergoing a fresh/frozen-thaw ET cycle was 14.1% The odds of EPL after frozen-thaw ETwere higher than those after fresh ET (1.11–2.27) -A thin endometrium on the day of ET increased the odds of EPL (1.01–3.33) -The risk factors for EPL were age 40 and older, obesity, frozen-thaw cycle, thin endometrium, and non-isolated tubal factor. Shuai et al. / 2022 [ 36 ] China retrospective cohort total of 35,076 Women, including 15,557 women in the fresh-ET group and 19,519 women in the frozen-ET group IVF/ICSI Baseline :maternal age, BMI, and infertility diagnosis Treatment :Ovarian-stimulation protocols (for fresh-transfer cycles), fresh/frozen-thawed ETs, endometrial preparation protocols (for frozen-transfer cycles), stage/number of transferred embryos, insemination method (for fresh-transfer cycles), and concomitant gynaecological disorders In addition, frozen-thawed transfer was a risk factor for ESA as compared with fresh transfer ( P  = 0.000). -Blastocyst transfer was a risk factor for ESA as compared with cleavage transfer -Maternal age, BMI, number of transferred embryos, and frozen-thawed transfer were independent risk factors for ESA in ART treatment cycles. Trindade et al. / 2022 [ 37 ] Brazil Retrospective case-control 499 Women submitted to IVF with a clinical pregnancy IVF/ICSI Baseline :Maternal and paternal age, female BMI, history of EP, previous miscarriage, previous pelvic surgery history, history of TBI Clinical :Endometriosis, male factor, ovulation factor, basal FSH and AFC Treatment : GnRH agonist or antagonist protocol, number of mature oocytes retrieved, number of transferred embryos, fresh or frozen ET, endometrial thickness prior to ET, sperm concentration and the day of ET, Controlled ovarian stimulation, IVF, endometrial preparation and ET, ORP and PGT cycles, Cleavage-stage In high-risk Women, a single blastocyst transfer seems to be a reasonable approach to decrease the ectopic pregnancy risk. -Tubal factor infertility ( p =0.005), previous miscarriage history ( p <0.05); number of cleavage-stage ETs ( p =0.028); ≥2 transferred embryos ( p =0.025), all associated with greater EP risk. Zhan et al. / 2022 [ 39 ] China cross-sectional 410 expectant couples (289 (70%) in model group, and the rest 121 (30%) into the validation group) Not Mention Baseline :Name, age, height, and weight, menstruation, age, Menstrual period, Menstrual blood volume, Dysmenorrhea childbearing history, as well as lifestyle including time of exercise, utilization of electronic equipment, Sleeping duration, occupational pressure, smoking and alcohol, and the frequency of unprotected sex, depression, BMI of male, Wife anxiety, Abortion, Extra-uterine pregnancy, Algopareunia, Basic disease (Hypertension, Diabetes mellitus, Thyroid diseases), Gynaecological disease: (PCOS, Endometriosis, Myoma of uterus, Ovarian disease) Gynecological surgery: (Fallopian tube surgery, Surgery for ovarian cysts, Hyster myomectomy, Metacyesis) Clinical :uterine size, endometrial thickness, AMH, FSH, LH, as well as FPG. - The risk factors for the predictive model included female age and occupational pressure, gynaecological disease, AMH, FSH, FPG, depression, as well as male smoking. The AUC for model A and model B was 0.954 (0.931 ~ 0.978) and 0.955 (0.931 ~ 0.979), respectively. Eldin et al. / 2023 [ 40 ] Egypt retrospective cohort 195 pregnancies on day 14–17 after ET ICSI Baseline :maternal age, BMI, status, Clinical :Baseline hormonal profiles. Treatment :Treatment protocol, quality and number of embryos The risk of early pregnancy loss was associated with older age and a smaller number of transferred embryos. Guo et al. / 2023 [ 41 ] China prospective cohort 169 pregnancies IVF/ICSI Baseline :Age, weight, height, BMI, racial origin, smoking habit, and parity; infertility type, length and cause(s) of infertility, Maternal mean arterial pressure Clinical :BasalFSH, ultrasound markers including mean gestational sac diameter, fetal heart activity, crown rump length and mUTPI and biochemical biomarkers including maternal sFlt-1, PlGF, kisspeptin and glycodelin-A Treatment :fertilization method, number, type and stage of ET, MAP - Significant prediction for miscarriage before 13 weeks’ gestation was provided by maternal age, fetal heart activity, mUTPI, and serum glycodelin-A. -A combination of maternal age, fetal heart activity, mUTPI, and serum glycodelin-A at 6 weeks’ gestation could effectively identify IVF/ET pregnancies at risk of first-trimester miscarriage Ozer et al. / 2023 [ 20 ] Turkey retrospective cohort 3805 good-quality FET cycles were IVF Baseline :female age, paternal age, BMI, duration of infertility, diagnosis of infertility, PCOS, history of RPL, RIF, severe male infertility, adenomyosis and endometriosis Treatment :endometrial preparation protocols (natural/artificial), embryo quality (top/good), number of previous IVF cycles, endometrial preparation, number of transferred embryos, number of total oocytes History of RPL, RIF, advanced female age, presence of PCOS, and high BMI (> 30 kg/m2) were the factors that increased first trimester pregnancy loss Yuan et al. / 2023 [ 16 ] China Retrospective cross-sectional 1,017 infertile women treated with IVF-ET IVF Baseline :Age, female BMI, fasting blood glucose, blood type, years of infertility, a history of uterine surgery, type of infertility, Clinical :type of cycle, genetic factors, including chromosomal abnormalities in men and women, female hormone levels, including the level of FSH, LH, E2, progesterone), testosterone, PRL, and AMH, thyroid hormone levels for serum- FT3, serum- FT4, TSH, TG-Ab, and TPO-Ab; immune factors, including ACA and antinuclear antibody; coagulation function tests, including PT, APTT, TT, FIB, and D-dimer; infection factors: leucorrhea test, Chlamydia trachomatis, Neisseria gonorrhoeae, TORCH test, etc.; abnormal ovarian structure: abnormal number, size, polycystic changes, or space occupying lesions in the ovary; uterine structural lesions, such as uterine malformation, leiomyoma, adenomyosis, scar diverticulum; male sperm abnormalities. Treatment :the number of sinus follicles, the number of eggs obtained, the number of high-quality embryos, the thickness and type of endometrium on the day of transfer, and the quality of the embryo The AUC score and the F1 score with the training set of the XGBoost model (0.877 ± 0.014 and 0.730 ± 0.019, respectively) were significantly higher than those of the logistic model (0.713 ± 0.013 and 0.568 ± 0.026, respectively) Ouyang et al. / 2023 [ 42 ] China Retrospective cross-sectional 13,977 infertile women after IVF-ET IVF/ICSI Baseline :Maternal age, BMI, Infertility duration, Transfer cycle, Infertility type, Cause of infertility Treatment :14-day HCG (mIU/ml), day-14 (blastocysts on day 12), serum βHCG levels, endometrial thickness on transfer, Insemination methods, Number of transferred embryos, Intrauterine hematomas (GSD ; YSD  r; EL ; embryonic heart rate) The AUC of this scoring system were 0.884 (95% CI 0.870–0.899) and 0.890 (95% CI 0.878–0.903) in the training set and test set, respectively Zhang et al. / 2024 [ 15 ] China Retrospective cross-sectional On the first cycles for IVF/ICSI treatment. 20,322(After excluding specific cases, a total of 6,724 cycles were analyzed) IVF/ ICSI Baseline :Age, female BMI, number of good quality embryos, lower AFC, basal testosterone, type of infertility, the number of spontaneous abortions Clinical :tubal factor, the number of induced abortions, basal FSH (IU/L) Treatment :LH levels on hCG day, number of retrieved oocytes, 2PN and MII oocytes, doses of Gn, Gn days, and E2 and progesterone on hCG day, number of good quality embryos, number of available embryos, LH (IU/L) on hCG day, and tubal factor. The significance of population differences and regional variations - for the elder subgroup (age ≥ 35 year), female age, basal FSH levels, and number of available embryos were significant risk factors, while for the younger subgroup (age < 35 year), female age, BMI, number of spontaneous abortions, and number of good quality embryos.- These 7 independent risks to build a predictive model is: age ( P  < 0.001), female BMI ( P  = 0.008), number of spontaneous abortions ( P  = 0.015), number of induced abortions ( P  = 0.325), basal FSH ( P  = 0.214), endometrium thickness on hCG day ( P  = 0.345) and number of Good quality embryos ( P  = 0.021) ART Assisted reproductive thechnology, GDM Gestational diabetes mellitus, PCOS polycystic ovary syndrome, IVF in vitro fertilization, ICSI Intracytoplasmic sperm injection, BMI Body mass index, BP Biochemical pregnancy, SA Spontaneous abortion, SSM Single spontaneous miscarriage, FSH Follicular-stimulating hormone, E2 Estradiol, hCG Human Chorionic Gonadotropin, RSM Recurrent spontaneous miscarriages, ET Embryo transfer, COH Controlled ovarian hyperstimulation, Gn Gonadotropin, PL Progesterone Luteal support, FBT Frozen-warmed blastocyst transfer, PGT-A Preimplantation genetic testing for aneuploidy, PRL Prolactin, P4 Progesterone, ICM Inner Cell Mass, LH Luteinizing hormone, AMH Anti-mullerian hormone, TE Trophectoderm, EPL Early pregnancy loss, IUI Intrauterine Insemination, ESA Early spontaneous abortion, EP Ectopic pregnancy, TBI Tubal factor infertility, AFC Antral follicle count, GnRH Gonadotropin-releasing hormone, OPR Ongoing Pregnancy Rate, FPG Fasting plasma glucose, AUC Area under the curve, mUTPI Mean uterine artery pulsatility index, sFlt-1 serum soluble fms-like tyrosine kinase-1, PlGF Placental growth factor, MAP Mean arterial pressure, FET frozen-thawed embryo transfer, RPL Recurrent pregnancy loss, RIF Recurrent implantation failure, XGBoost eXtreme Gradient Boosting, FT3 Serum-free triiodothyronine, FT4 Serum-free thyroxine, TSH Thyroid-stimulating hormone, TG-Ab Thyroglobulin antibody, TPO-Ab Thyroid Peroxidase Antibodies, ACA Anticardiolipin antibody, PT Prothrombin time, APTT Activated partial thromboplastin time, TT Thrombin time, FIB Plasma fibrinogen, TORCH (Toxoplasmosis, Other, Rubella, Cytomegalovirus, Herpes simplex virus), βHCG β-human chorionic gonadotropin, GSD Gestational sac diameter, YSD Yolk sac diameter, EL Embryonic length, 2PN Two Pronuclei, MII Metaphase II The main characteristics of the included studies Ashrafi et al. / 2014 [ 43 ] Iran Baseline : Age, parity, irregular menses and hirsutism cases, menstrual irregularity, and menarche age. Clinical :serum triglyceride level, pregestational metformin use- No significant differences were found in terms of pre-pregnancy BMI and weight gain during pregnancy among groups. -Metformin consumption was associated with a reduction in GDM prevalence, significantly ( p  = 0.01). Groups were significantly different in terms of mean serum levels of triglyceride and cholesterol ( p  = 0.02 and p  = 0.04, respectively). Yang et al. /2015 [ 44 ] China Womenwho reached the ongoing pregnancy stage, 7,598 cases were divided into three groups: group 1, lack of pregnancy ( n  = 6,651); group 2, BP ( n  = 520); and group 3, SA ( n  = 427 Baseline :Age, BMI. Clinical : basic FSH level and E2 level. Treatment : endometrium thickness on hCG administration, secondary infertility proportion Wang et al. /2016 [ 45 ] China 751 Women with at least one spontaneous miscarriage(irrespective of conception method (natural or ART)) Baseline :Age, Age of spouse, Menarche age, BMI, Education Clinical : Ovulation disorder, Uterine factors, Cervical factors, Fallopian tube factors, Pelvic cavity factors, Endocrine disorders, Immune factors, Chromosomal abnormalities, Reproductive tract infections, Male infertility. Higher education level and a pelvic cavity disorder were associated with a lower risk of RSM (OR = 0.27 and 0.46, respectively). - Late spontaneous miscarriages were more frequent in Women with RSM than in those with a SSM (31.5% vs. 14.2%, respectively) and were associated with a history of uterine cavity procedures (OR = 2.095) and cervical factors related to infertility (OR = 4.136, 95% CI: 1.012–16.90). Hu et al. /2018 [ 46 ] China Baseline :maternal age, maternal BMI (kg/m2), paternal BMI, and infertility type Clinical : COH protocol Treatment :Total Gn dose, fertilization methods, cycle type, no. of embryos transferred, cleavage-stage embryo or blastocyst, serum hCG level 14 days after transfer - Couples in ART pregnancies demonstrated a significantly increased risk of PL as maternal age (HR = 1.31, Ptrend < 0.001) grows. - In contrast to fresh cycle, women who received frozen cycle embryo had a significant increased risk of early miscarriage ( P  < 0.001), while frozen cycle was linked with lower risk of late miscarriage ( P  = 0.045). − 4 factors (maternal age, COH protocol, cycle type and serum hCG level 14 days after transfer) had an independent impact on miscarriage mainly before 12 weeks of gestational age. - Clinicians may make it better to evaluate a Women risk of PL based on the maternal age at the time of treatment, COH protocol, cycle type and serum hCG level 14 days after transfer and the gestational week of the fetus. Boynukalin et al. /2020 [ 47 ] Turkey Baseline :Maternal age, paternal age, BMI, infertility diagnose, number of previous attempts, number of previous miscarriage Treatment :Duration of stimulation, total Gn dosage used, E2 and P4 levels on trigger day, endometrial thickness, E2 and P4 levels 6 days before the FBT cycle, mitoscore, ICM score, trophectoderm score and day of embryo biopsy Li et al. /2021 [ 48 ] China Women successfully conceived with ART. 16,097 Women (2,378 had an abortion and 13,719 did not have one). Baseline :female age, male age and female cause of infertility, BMI Clinical :FSH, E2, LH, AMH Treatment :controlled ovulation induction protocol, PRL, number of treatment cycles, number of transferred embryos, method of fertilization, embryo stage, number of fetuses and controlled ovulation induction protocol, dose of Gn -Found that when female age > 32 year (cut-off point), age and the risk of spontaneous abortion were positively correlated. -There was a linear positive correlation between AFC and live birth. -The older the male age, the higher the incidence of abortion. -Basal P, AMH and PRL in the abortion group were also significantly lower than those in the non-abortion group. -There was a statistical difference between the number of treatment cycles, embryos and fetuses at pregnancy diagnosis between the two groups Bilibio et al. / 2022 [ 35 ] Brazil Baseline :maternal age, paternal age, Infertility time, repeat abortion, male factor, endometriosis, PCOS Clinical :AFC, tubal factor, Treatment :Ovarian reserve, and semen quality (Seminal collection type, Semen concentration, motility, preparation) cryopreserved oocyte, parameters on D1 (pro-nucleus, type of nuclear precursor corpuscles, polar corpuscles), D2 and D3 (number of blastomeres, embryonic classification), sperm with normal morphology, TE quality, and ICM quality -Analysis revealed maternal age and type C TE as the main risk factors for aneuploidy. -Poor ovarian reserve ( p   38 and > 36 years, respectively), AFC  38 year, type-C embryo with trophectoderm, which showed a positive predictive value of 88.6% and a specificity of 97.5%. Wang et al. / 2022 [ 38 ] China Baseline :Female age, male age, female BMI, Male smoking, number of previous miscarriages, Adverse pregnancy history, and IUI unfertilized history. IVF failed ET history Clinical :AMH level, cycle type Treatment :ET stage, number of transferred embryos, thickness of endometrium, isolated tubal factor, isolated diminished ovarian reserve factor, Female chromosome, Male chromosome, Embryo transfer stage, Male factor. The EPL rate of the first-time pregnancies for infertility Women undergoing a fresh/frozen-thaw ET cycle was 14.1% The odds of EPL after frozen-thaw ETwere higher than those after fresh ET (1.11–2.27) -A thin endometrium on the day of ET increased the odds of EPL (1.01–3.33) -The risk factors for EPL were age 40 and older, obesity, frozen-thaw cycle, thin endometrium, and non-isolated tubal factor. Shuai et al. / 2022 [ 36 ] China Baseline :maternal age, BMI, and infertility diagnosis Treatment :Ovarian-stimulation protocols (for fresh-transfer cycles), fresh/frozen-thawed ETs, endometrial preparation protocols (for frozen-transfer cycles), stage/number of transferred embryos, insemination method (for fresh-transfer cycles), and concomitant gynaecological disorders In addition, frozen-thawed transfer was a risk factor for ESA as compared with fresh transfer ( P  = 0.000). -Blastocyst transfer was a risk factor for ESA as compared with cleavage transfer -Maternal age, BMI, number of transferred embryos, and frozen-thawed transfer were independent risk factors for ESA in ART treatment cycles. Trindade et al. / 2022 [ 37 ] Brazil Baseline :Maternal and paternal age, female BMI, history of EP, previous miscarriage, previous pelvic surgery history, history of TBI Clinical :Endometriosis, male factor, ovulation factor, basal FSH and AFC Treatment : GnRH agonist or antagonist protocol, number of mature oocytes retrieved, number of transferred embryos, fresh or frozen ET, endometrial thickness prior to ET, sperm concentration and the day of ET, Controlled ovarian stimulation, IVF, endometrial preparation and ET, ORP and PGT cycles, Cleavage-stage In high-risk Women, a single blastocyst transfer seems to be a reasonable approach to decrease the ectopic pregnancy risk. -Tubal factor infertility ( p =0.005), previous miscarriage history ( p <0.05); number of cleavage-stage ETs ( p =0.028); ≥2 transferred embryos ( p =0.025), all associated with greater EP risk. Zhan et al. / 2022 [ 39 ] China 410 expectant couples (289 (70%) in model group, and the rest 121 (30%) into the validation group) Baseline :Name, age, height, and weight, menstruation, age, Menstrual period, Menstrual blood volume, Dysmenorrhea childbearing history, as well as lifestyle including time of exercise, utilization of electronic equipment, Sleeping duration, occupational pressure, smoking and alcohol, and the frequency of unprotected sex, depression, BMI of male, Wife anxiety, Abortion, Extra-uterine pregnancy, Algopareunia, Basic disease (Hypertension, Diabetes mellitus, Thyroid diseases), Gynaecological disease: (PCOS, Endometriosis, Myoma of uterus, Ovarian disease) Gynecological surgery: (Fallopian tube surgery, Surgery for ovarian cysts, Hyster myomectomy, Metacyesis) Clinical :uterine size, endometrial thickness, AMH, FSH, LH, as well as FPG. - The risk factors for the predictive model included female age and occupational pressure, gynaecological disease, AMH, FSH, FPG, depression, as well as male smoking. The AUC for model A and model B was 0.954 (0.931 ~ 0.978) and 0.955 (0.931 ~ 0.979), respectively. Eldin et al. / 2023 [ 40 ] Egypt Baseline :maternal age, BMI, status, Clinical :Baseline hormonal profiles. Treatment :Treatment protocol, quality and number of embryos Baseline :Age, weight, height, BMI, racial origin, smoking habit, and parity; infertility type, length and cause(s) of infertility, Maternal mean arterial pressure Clinical :BasalFSH, ultrasound markers including mean gestational sac diameter, fetal heart activity, crown rump length and mUTPI and biochemical biomarkers including maternal sFlt-1, PlGF, kisspeptin and glycodelin-A Treatment :fertilization method, number, type and stage of ET, MAP - Significant prediction for miscarriage before 13 weeks’ gestation was provided by maternal age, fetal heart activity, mUTPI, and serum glycodelin-A. -A combination of maternal age, fetal heart activity, mUTPI, and serum glycodelin-A at 6 weeks’ gestation could effectively identify IVF/ET pregnancies at risk of first-trimester miscarriage Baseline :female age, paternal age, BMI, duration of infertility, diagnosis of infertility, PCOS, history of RPL, RIF, severe male infertility, adenomyosis and endometriosis Treatment :endometrial preparation protocols (natural/artificial), embryo quality (top/good), number of previous IVF cycles, endometrial preparation, number of transferred embryos, number of total oocytes Baseline :Age, female BMI, fasting blood glucose, blood type, years of infertility, a history of uterine surgery, type of infertility, Clinical :type of cycle, genetic factors, including chromosomal abnormalities in men and women, female hormone levels, including the level of FSH, LH, E2, progesterone), testosterone, PRL, and AMH, thyroid hormone levels for serum- FT3, serum- FT4, TSH, TG-Ab, and TPO-Ab; immune factors, including ACA and antinuclear antibody; coagulation function tests, including PT, APTT, TT, FIB, and D-dimer; infection factors: leucorrhea test, Chlamydia trachomatis, Neisseria gonorrhoeae, TORCH test, etc.; abnormal ovarian structure: abnormal number, size, polycystic changes, or space occupying lesions in the ovary; uterine structural lesions, such as uterine malformation, leiomyoma, adenomyosis, scar diverticulum; male sperm abnormalities. Treatment :the number of sinus follicles, the number of eggs obtained, the number of high-quality embryos, the thickness and type of endometrium on the day of transfer, and the quality of the embryo Baseline :Maternal age, BMI, Infertility duration, Transfer cycle, Infertility type, Cause of infertility Treatment :14-day HCG (mIU/ml), day-14 (blastocysts on day 12), serum βHCG levels, endometrial thickness on transfer, Insemination methods, Number of transferred embryos, Intrauterine hematomas (GSD ; YSD  r; EL ; embryonic heart rate) On the first cycles for IVF/ICSI treatment. 20,322(After excluding specific cases, a total of 6,724 cycles were analyzed) Baseline :Age, female BMI, number of good quality embryos, lower AFC, basal testosterone, type of infertility, the number of spontaneous abortions Clinical :tubal factor, the number of induced abortions, basal FSH (IU/L) Treatment :LH levels on hCG day, number of retrieved oocytes, 2PN and MII oocytes, doses of Gn, Gn days, and E2 and progesterone on hCG day, number of good quality embryos, number of available embryos, LH (IU/L) on hCG day, and tubal factor. The significance of population differences and regional variations - for the elder subgroup (age ≥ 35 year), female age, basal FSH levels, and number of available embryos were significant risk factors, while for the younger subgroup (age < 35 year), female age, BMI, number of spontaneous abortions, and number of good quality embryos.- These 7 independent risks to build a predictive model is: age ( P  < 0.001), female BMI ( P  = 0.008), number of spontaneous abortions ( P  = 0.015), number of induced abortions ( P  = 0.325), basal FSH ( P  = 0.214), endometrium thickness on hCG day ( P  = 0.345) and number of Good quality embryos ( P  = 0.021) ART Assisted reproductive thechnology, GDM Gestational diabetes mellitus, PCOS polycystic ovary syndrome, IVF in vitro fertilization, ICSI Intracytoplasmic sperm injection, BMI Body mass index, BP Biochemical pregnancy, SA Spontaneous abortion, SSM Single spontaneous miscarriage, FSH Follicular-stimulating hormone, E2 Estradiol, hCG Human Chorionic Gonadotropin, RSM Recurrent spontaneous miscarriages, ET Embryo transfer, COH Controlled ovarian hyperstimulation, Gn Gonadotropin, PL Progesterone Luteal support, FBT Frozen-warmed blastocyst transfer, PGT-A Preimplantation genetic testing for aneuploidy, PRL Prolactin, P4 Progesterone, ICM Inner Cell Mass, LH Luteinizing hormone, AMH Anti-mullerian hormone, TE Trophectoderm, EPL Early pregnancy loss, IUI Intrauterine Insemination, ESA Early spontaneous abortion, EP Ectopic pregnancy, TBI Tubal factor infertility, AFC Antral follicle count, GnRH Gonadotropin-releasing hormone, OPR Ongoing Pregnancy Rate, FPG Fasting plasma glucose, AUC Area under the curve, mUTPI Mean uterine artery pulsatility index, sFlt-1 serum soluble fms-like tyrosine kinase-1, PlGF Placental growth factor, MAP Mean arterial pressure, FET frozen-thawed embryo transfer, RPL Recurrent pregnancy loss, RIF Recurrent implantation failure, XGBoost eXtreme Gradient Boosting, FT3 Serum-free triiodothyronine, FT4 Serum-free thyroxine, TSH Thyroid-stimulating hormone, TG-Ab Thyroglobulin antibody, TPO-Ab Thyroid Peroxidase Antibodies, ACA Anticardiolipin antibody, PT Prothrombin time, APTT Activated partial thromboplastin time, TT Thrombin time, FIB Plasma fibrinogen, TORCH (Toxoplasmosis, Other, Rubella, Cytomegalovirus, Herpes simplex virus), βHCG β-human chorionic gonadotropin, GSD Gestational sac diameter, YSD Yolk sac diameter, EL Embryonic length, 2PN Two Pronuclei, MII Metaphase II The studies were geographically distributed in different countries, with the most number of studies in China ( n = 11) [ 15 , 16 , 36 , 38 , 39 , 41 , 42 , 44 – 46 , 48 ]. Turkey ( n = 2) [ 20 , 47 ], Brazil ( n = 2) [ 35 , 37 ], Egypt [ 40 ], and Iran [ 43 ] also have conducted studies in this field. In this study, 9 studies were conducted as cohort [ 20 , 35 , 36 , 40 , 41 , 44 , 46 – 48 ], 6 studies as cross-sectional [ 15 , 16 , 38 , 39 , 42 , 43 ], and 1 study as observational [ 45 ] and case-control [ 37 ]. In most studies, the two methods of IVF/ICSI ( n = 8) [ 15 , 36 – 38 , 41 – 43 , 46 ] were used together in the study population, while in other studies only samples with IVF [ 16 , 20 , 44 ] or ICSI [ 35 , 40 , 47 ] were used, and 3 studies “not mention” [ 39 , 45 , 48 ]. Overall, the risk factors were categorized into three main characteristics: Baseline (demographics, lifestyle, and medical history), Clinical, and Treatment. Most studies ( n = 10) included all three main characteristics [ 15 , 16 , 35 , 37 , 38 , 40 , 41 , 44 , 46 , 48 ], some studies included only two main characteristics such that 4 studies [ 20 , 36 , 42 , 47 ] included baseline and treatment characteristics and 3 studies [ 39 , 43 , 45 ] included baseline and clinical characteristics. This characteristic is one of the most important categories in ART treatment, which has been mentioned in all selected studies and is one of the key factors in predicting miscarriage in patients undergoing treatment. This characteristic includes demographic information including age, parental BMI, menstrual age, menstruation, parity, which significantly affect fertility treatments. Also, due to the industrialization of societies, unhealthy habits and lifestyles such as poor nutrition, physical inactivity, tobacco use, alcohol, depression and anxiety have been introduced as risk factors that are significantly associated with an increased risk of miscarriage. On the other hand, the couple’s medical history and genetics, including the type and duration of infertility, the number of previous pregnancy attempts, the number of previous miscarriages, previous fertility problems, and history of surgery, lead to an increased risk of miscarriage. This section included risk factors that examined hormones such as levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), progesterone (P), testosterone (T), prolactin (PRL), anti-Müllerian hormone (AMH), thyroid-stimulating hormone (TSH), and antral follicle count. Also, genital tract infections, abnormal ovarian structure, abnormal egg number and size, polycystic changes, ovulation disorders, fallopian tube factors, uterine factors, and chromosomal abnormalities in women and men effectively affect the outcome of treatment and are very helpful in identifying couples at risk of miscarriage. Accurate evaluation of these parameters by physicians are essential, and paying attention to these risk factors will lead to designing an appropriate treatment plan for each couple, leading to better conditions in treatment and a reduction in miscarriage rates. This group of risk factors is related to the type of treatment and treatment protocols that affect the miscarriage rate. The complications of each treatment method such as IVF, ICSI vary depending on the implementation differences. Factors such as total gonadotropin dose (Gn), cycle type, serum hCG level after transfer (IU/L) are among the parameters that are included in the treatment category. Also, the levels of estradiol and progesterone hormones on the trigger day and the controlled ovulation induction protocol, prolactin (PRL), the number of treatment cycles, and the number of embryos transferred affect the miscarriage rate. Also, examining ovarian reserve and semen quality (type of semen collection, concentration, motility, preparation) and sperm with normal morphology, frozen eggs, embryo quality on days 2 and 3, and paying attention to the physical and mental condition of the couple have been introduced as influential factors that play a very important role in reducing the risk of miscarriage among couples undergoing ART treatment and are very helpful in the success of infertility treatment with minimal complications.

One of the major reasons for the failure of ART treatment is miscarriage. In this study, risk factors for miscarriage in couples undergoing ART treatment were identified and classified into three main groups. The classification of risk factors (into baseline, clinical and treatment characteristics) determined the extent of the risk factors’ impact and the clarity of their impact on miscarriage. Addressing these risk factors, in addition to affecting miscarriage, can also affect overall reproductive health. By being aware of these factors, couples gain better control over their disease management by increasing their knowledge by receiving appropriate counseling and early detection of potential factors leading to miscarriage. Physicians also take a Comprehensive and personalized approach to managing ART patients by using risk factors. Overall, ART has significantly increased the chances of conceiving a child for infertile patients, However, further research is needed to uncover influencing factors and their relationship to improve outcomes for couples undergoing treatment.

In recent years, significant advances have been achieved in the treatment of infertility through ART. However, miscarriage remains a major complication with a relatively high incidence among ART outcomes. In the present study, a wide range of risk factors for miscarriage among couples undergoing ART were identified and classified into three main categories: baseline, clinical, and treatment-related characteristics. The findings show that these factors, in addition to being numerous, can also affect pregnancy outcomes in some cases. Including demographic, lifestyle, and medical history elements that may affect fertility outcomes. Age is one of the most important risk factors in both men and women, and almost all studies emphasize its importance. Increasing age is associated with decreased egg quality and mitochondrial disorders, which ultimately increase the risk of miscarriage. This finding is consistent with several studies that have shown that after the age of 35, and especially after the age of 40, the risk of miscarriage increases significantly with an increase in chromosomal abnormalities and other fertility challenges [ 15 , 38 , 46 ]. Menarche age has also been mentioned as a risk factor for successful miscarriage in some studies [ 43 , 45 ]. Paternal age is also an important factor in reproductive success [ 35 , 48 ]. Body mass index (BMI) was also identified as an important factor. Evidence suggests that high BMI is associated with systemic inflammation, insulin resistance, and adverse changes in the endometrial environment, which can reduce the likelihood of successful implantation. Our findings are consistent with previous reviews that have shown a direct association between obesity and an increased risk of miscarriage in ART [ 20 , 47 ]. However, in the study [ 43 ], no relationship was observed between pre-pregnancy BMI and weight gain during pregnancy. Education level is often associated with better access to reliable information and resources about fertility and miscarriage, leading to better access to health care services and support organizations (counseling). However, the relationship between education level and miscarriage prevention had mixed results, with some studies confirming this relationship [ 49 ], while others found it ineffective [ 45 ] so the need for further research in the future is emphasized. Lifestyle is a potential and multifaceted factor that influences the prevention of miscarriage in women undergoing ART. A proportional diet and rich in essential nutrients improve fertility outcomes. Studies have shown that consuming foods rich in vitamins (D, E), folic acid, omega-3 fatty acids, and antioxidants are essential for fertility [ 11 ]. Also, physical activity, moderate and regular exercise, leads to better stress management and weight control, and has a positive effect on the quality of infertility treatment. Some physical activities lead to better hormonal balance and better stress management and reduce the impact of psychological stress in infertility treatment [ 39 ]. On the other hand, excessive or high-intensity exercise may have adverse effects on fertility outcomes, and its amount needs to be adjusted in consultation with a physician. lifestyle factors including smoking and alcohol consumption can reduce gamete quality and negatively impact pregnancy outcome through increased oxidative stress. These results highlight the importance of corrective interventions before ART initiation, such as smoking cessation counseling and dietary modification In a study [ 39 ], in addition to risk factors including female age and job pressure, the effect of other risk factors such as job stress, smoking, depression, and anxiety on the prediction of miscarriage has been mentioned. The relationship between medical history and miscarriage prevention in women undergoing ART, involves several key factors. In various studies, factors such as previous pregnancy outcomes, history of any miscarriage, and complications of previous pregnancies have been identified as influential. A history of various surgeries, such as fallopian tube and pelvic surgeries, are among the risk factors for miscarriage that may have inflamed the patient’s reproductive system. The presence and development of underlying diseases or chronic disease related to women such as polycystic ovary syndrome, high blood pressure, autoimmune disorders, lead to an increased risk of miscarriage and infertility treatment failure [ 50 , 51 ]. Various laboratory parameters, endocrine, hormonal and genetic disorders played a fundamental role in determining the risk of miscarriage in women undergoing ART. By carefully managing and monitoring these parameters, one can hope for successful pregnancy outcomes and reduce the incidence of miscarriage in the ART population. Monitoring and regulating hormone levels is essential for a successful pregnancy and improves uterine conditions and ovulation. Various hormones are important in successful fertility, the most important of which are LH, E2, P, T, PRL, AMH and basal FSH [ 15 , 48 ]. Also, hyperthyroidism or hypothyroidism plays an important role in fertility and reducing miscarriage, which is related to the TSH hormone. These findings reinforce the importance of endocrine balance and ovarian reserve in successful implantation [ 16 ]. Among the clinical indicators, ovarian reserve was of particular importance. Low AMH and high FSH levels not only indicate reduced egg quality and quantity, but are also associated with an increased risk of miscarriage. This finding is consistent with previous studies [ 52 , 53 ] that have identified ovarian reserve indicators as predictors of pregnancy outcome. Also, hormonal disorders and underlying diseases such as hypothyroidism or hyperthyroidism and autoimmune diseases (especially antiphospholipid syndrome) have been identified in some studies as factors contributing to increased miscarriage. These findings emphasize the importance of screening and management of diseases before ART [ 54 , 55 ]. Some studies [ 16 , 35 , 45 , 47 ] examined the effect of genetic factors, including chromosome abnormalities in women and men, and factors such as endocrine disorders, fallopian tube factors, and pelvic cavity factors were expressed as influential parameters in miscarriage. This characteristic included various risk factors that refer to the stages of treatment of infertile couples and are effective in predicting miscarriage in couples undergoing ART treatment. One of the most important risk factors in this group is the type of treatment method that is performed under specific conditions, depending on the cause of infertility (male or female). The miscarriage rate in the ICSI method was higher than the IVF method, which is more effective in this regard because the male factor of infertility. Ovulation stimulation protocols (agonist and antagonist) are among the very important risk factors that are effective in reducing miscarriage [ 37 , 48 ]. The quality and number of embryos transferred on the day of transfer affect the outcome of infertility treatment and live birth [ 15 ]. With an increase in the number of embryos transferred, the chance of pregnancy increases, but on the other hand, the risk of pregnancy and multiple births increases, which can lead to miscarriage. Embryo quality reduce risk of miscarriage. Poor-quality embryos are more likely to have genetic abnormalities and cell division disorders, which increases the risk of miscarriage. Our findings are consistent with studies [ 20 , 48 ] that have shown that the use of preimplantation genetic screening (PGT) can help reduce the risk of miscarriage. Differences between fresh and frozen embryo transfer have been reported in some studies [ 46 ]. Evidence suggests that in patients at high risk of OHSS or endometrial insufficiency, the use of frozen embryos may have a better outcome. This finding highlights the importance of personalized decision-making in the choice of embryo transfer method.The results showed that the risk of early miscarriage was higher in women receiving frozen embryos than in women with fresh embryos, however, frozen cycles were associated with a lower risk of late miscarriage. The studies [ 15 , 42 , 44 ] also showed that identifying risk factors of maternal age, Controlled ovarian hyperstimulation (COH) protocol, cycle type and serum hCG level 14 days after transfer were effective in predicting both types of miscarriage. Embryo quality is related to parameters such as number, embryo grading, and day of biopsy or transfer outcomes [ 35 , 47 ]. It should be noted that sometimes blastocyst stage transfer is suitable for implantation, but due to chromosomal abnormalities, there is a possibility of increased risk of miscarriage in certain contexts [ 36 ]. In addition, Yuan et al. [ 16 ], Ouyang et al. [ 42 ] using new data-driven tools (machine learning models), provided suitable models for estimating the risk of miscarriage that use a variety of biochemical, hormonal, and fetal data to increase prediction accuracy and emphasize the presence of a personal physician in care. This study has several important clinical implications Addressing modifiable risk factors, such as body mass index, smoking and alcohol consumption, and metabolic disorders (diabetes and metabolic diseases) can increase the likelihood of successful pregnancy and reduce the risk of miscarriage. Therefore, the need for comprehensive screening and correction of modifiable risk factors before initiating ART has an impact on positive pregnancy outcomes. careful monitoring of ovarian reserve markers, including AMH and FSH, as well as a thorough assessment of the endometrium prior to embryo transfer, is crucial. Such evaluations enable clinicians to identify high-risk patients and implement tailored management strategies accordingly. selecting an appropriate stimulation protocol, making individualized decisions regarding fresh versus frozen embryo transfer, and prioritizing high-quality embryos can all play a decisive role in improving ART outcomes. Overall, these findings highlight that a structured, patient-centered, and risk-based approach can support clinicians in optimizing ART outcomes and minimizing the likelihood of miscarriage. Considering that infertility treatment is a very complex and multifaceted process, it seems that some of the risk factors effective in diagnosing miscarriage, despite their role and importance, have received less attention. It is suggested that further studies be conducted focusing on issues such as lifestyle and behavioral habits such as nutrition, physical activity, etc. In studies related to miscarriage, more attention has been paid to risk factors related to women, while in infertility treatment, the male factor is effective in terms of clinical and emotional support and companionship of the woman, so it is suggested that male factors should also be studied in addition to female factors. It is also suggested that other aspects such as family, socio-economic, mental and psychological issues of couples should be addressed in future studies. On the other hand, with the increasing use of artificial intelligence in various fields including data analysis, it is suggested that these algorithms be used to analyze big data in the field of infertility and discover new and effective risk factors, considering the development of machine learning and deep learning algorithms. In addition, longitudinal studies, integrating genetic and immunological profiles, are recommended to improve miscarriage diagnosis in the future. In this study, the focus was on original articles and specific types of studies, which may have missed some risk factors. Also, due to resource limitations, it was not possible to access all databases and full texts of articles in this study. Due to the selection of articles published in English, it is possible to exclude some relevant studies that have been published in other languages. Some complications of ART treatment such as multiple births and ovarian hyperstimulation syndrome (OHSS) were ignored. Also, the effect of underlying and chronic diseases such as (diabetes, tuberculosis, etc.) was also excluded, which could affect the comprehensiveness of risk factors. In this study, since the results were reported descriptively, the ability to quantitatively assess the relative importance of different risk factors was limited. Moreover, due to the considerable heterogeneity in study designs, participant populations, and reported risk factors, conducting a meta-analysis was not feasible. Nevertheless, we believe that the descriptive classification provided in this review offers a useful framework for clinicians and researchers and may serve as a foundation for developing predictive models in future studies.

According to the definition of the World Health Organization (WHO), infertility means the failure to achieve pregnancy after one year (or more) of regular unprotected sexual intercourse [ 1 ]. Infertility is one of the major health problems that has caused many challenges and troubles for society, especially couples. According to studies, about 50 to 80 million people suffer from some form of infertility around the world [ 2 , 3 ]. Treatment of infertility depends on various factors, including personal preferences of each individual, the duration of infertility, the age of the couple, and the causes of infertility in both men and women, and some causes of infertility cannot be treated [ 4 – 7 ]. Assisted reproductive technologies (ART), as one of the ways to treat infertility, include methods in which egg and sperm are fertilized outside the individual’s body in the laboratory, and the resulting embryo is transferred to the mother’s uterus [ 8 , 9 ]. The use of ART methods is a very complex, time-consuming process and depends on various factors such as the couple’s history, clinical and laboratory factors, and treatment protocol [ 10 , 11 ]. The success rate of ART has been reported to be about 25 to 30%, while it requires a lot of time and emotional, psychological, and economic support, and the possibility of failure in the early stages is relatively high [ 9 , 12 ]. ART also sometimes leads to complications such as ovarian hyperstimulation syndrome, ectopic pregnancy, miscarriage, etc [ 13 , 14 ]. Miscarriage is a common complication of ART. A miscarriage is the spontaneous loss of a fetus before the 20th week of pregnancy, meaning that the fetus cannot survive outside the uterus [ 15 , 16 ]. It is estimated that 10 to 25% of all pregnancies end in miscarriage. If a miscarriage occurs during the first 13 weeks of pregnancy (the first trimester), it is known as early miscarriage, which accounts for about 80% of miscarriages [ 17 ]. This rate is significantly higher in couples undergoing ART than in natural conception [ 18 ]. A miscarriage between the 14th week and before the 20th week is called late miscarriage. Factors that influence miscarriage include parental age, lifestyle, maternal health, history of recurrent miscarriages, hormonal disorders, smoking, and alcohol consumption [ 17 , 19 , 20 ]. The use of new technologies, such as mobile applications, decision support systems, fuzzy logic and artificial intelligence, especially machine learning algorithms, is increasing in the medical fields [ 21 – 24 ]. Machine learning uses the analysis of large volumes of data to discover patterns and relationships [ 25 ]. Given the complexity and large number of factors affecting ART treatment, this method is a suitable tool for diagnosing miscarriage [ 26 , 27 ]. Using machine learning, the relationship between features and the target is examined, the most effective features are selected (feature selection), and a prediction model is built [ 28 , 29 ]. In addition to identifying risk factors, these models can reduce miscarriages caused by ART treatments by integrating with health information systems and clinical decision support systems, also that can be used as a complementary technology alongside physicians and accurate decision-making or designing personalized treatment plans [ 30 , 31 ]. According to the results of previous studies [ 32 , 33 ], in addition to physical problems and complications such as bleeding and infection in women, miscarriage also affects their mental health and causes feelings of depression, stress, and self-blame. These issues can significantly affect women’s daily life and interpersonal relationships. Also, despite the progress in treatment, the rate of miscarriage is still high, especially in women undergoing ART treatment, so identifying the factors that diagnose miscarriage in women undergoing ART treatment is effective in reducing their physical and psychological complications. This study was conducted to identify the risk factors predicting miscarriage among couples undergoing ART treatment, in order to increase knowledge, to identify the impact of factors affecting miscarriage, and to take preventive measures (counseling, education) in terms of reducing the incidence of miscarriage in ART treatment.

Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3. Supplementary Material 4. Supplementary Material 4.