The Ovarian Sensitivity Index (OSI) Significantly Correlates with Ovarian Reserve Biomarkers, Is More Predictive of Clinical Pregnancy than the Total Number of Oocytes, and Is Consistent in Consecutive IVF Cycles

A novel machine-learning framework based on early embryo morphokinetics identifies a feature signature associated with blastocyst development

BACKGROUND

Artificial Intelligence entails the application of computer algorithms to the huge and heterogeneous amount of morphodynamic data produced by Time-Lapse Technology. In this context, Machine Learning (ML) methods were developed in order to assist embryologists with automatized and objective predictive models able to standardize human embryo assessment. In this study, we aimed at developing a novel ML-based strategy to identify relevant patterns associated with the prediction of blastocyst development stage on day 5.

METHODS

We retrospectively analysed the morphokinetics of 575 embryos obtained from 80 women who underwent IVF at our Unit. Embryo morphokinetics was registered using the Geri plus® time-lapse system. Overall, 30 clinical, morphological and morphokinetic variables related to women and embryos were recorded and combined. Some embryos reached the expanded blastocyst stage on day 5 (BL Group, n = 210), some others did not (nBL Group, n = 365).

RESULTS

The novel EmbryoMLSelection framework was developed following four-steps: Feature Selection, Rules Extraction, Rules Selection and Rules Evaluation. Six rules composed by a combination of 8 variables were finally selected, and provided a predictive power described by an AUC of 0.84 and an accuracy of 81%.

CONCLUSIONS

We provided herein a new feature-signature able to identify with an high performance embryos with the best developmental competence to reach the expanded blastocyst stage on day 5. Clear and clinically relevant cut-offs were identified for each considered variable, providing an objective tool for early embryo developmental assessment.

Cytosine-guanine-guanine repeats of FMR1 gene negatively affect ovarian reserve and response in Chinese women

RESEARCH QUESTION

Do cytosine-guanine-guanine (CGG) repeats of the FMR1 gene affect ovarian function, ovarian response and assisted reproductive technology (ART) outcomes in Chinese women?

DESIGN

A retrospective cohort study of 5869 women who underwent 8932 ART cycles at Women's Hospital, School of Medicine, Zhejiang University between January 2018 and June 2021. Basic hormone level, oocyte yield, embryo quality and the rate of live birth were considered as main outcome measures to evaluate the effects of CGG repeats on ovarian function, ovarian response and ART outcomes.

RESULTS

The CGG repeats were negatively related to serum anti-Müllerian hormone (AMH), oestradiol, antral follicle count (AFC) and oocyte yield. A significant association was found between serum AMH, oestradiol and AFC even after age was controlled for. No statistically significant association, however, was found between CGG repeats and embryo quality or live birth rate. Ovarian function mediated the association between CGG repeats and ovarian response.

CONCLUSION

Increased CGG repeats on the FMR1 gene were associated with diminished ovarian function and poor ovarian response, and ovarian function played an intermediary role in the relationship between CGG repeats and ovarian response.

Evaluative effectiveness of follicular output rate, ovarian sensitivity index, and ovarian response prediction index for the ovarian reserve and response of low-prognosis patients according to the POSEIDON criteria: a retrospective study

The aim was to explore the implications of follicular output rate (FORT), ovarian sensitivity index (OSI), ovarian response prediction index (ORPI), and follicle-to-oocyte index (FOI) in low-prognosis patients defined by POSEIDON criteria. In total, 4030 fresh in vitro fertilization (IVF) cycles from January 2013 to October 2021 were included in this retrospective cohort analysis and were categorized into four groups based on the POSEIDON criteria. The FORT between Groups 1 and 2 (0.61 ± 0.34 vs. 0.65 ± 0.35, P = 0.081) and Groups 3 and 4 (1.08 ± 0.82 vs. 1.09 ± 0.94, P = 0.899) were similar. The OSI in the order from the highest to the lowest were 3.01 ± 1.46 in Group 1, 2.28 ± 1.09 in Group 2, 1.54 ± 1.04 in Group 3, and 1.34 ± 0.96 in Group 4 (P < 0.001). The trend in the ORPI values was consistent with that in the OSI. FORT, OSI, ORPI, and FOI complemented each other and offered excellent effectiveness in reflecting ovarian reserve and response, but they were not good predictors of clinical pregnancy rate (CPR) from IVF.

Ovarian sensitivity index can be used as a more sensitive indicator than follicular output rate to predict IVF/ICSI outcomes in patients of normal expected ovarian response stimulated with GnRH antagonist protocol

This retrospective study was performed to investigate the predictive power of the Ovarian Sensitivity Index (OSI) for IVF/ICSI outcomes in infertile patients who were of normal expected ovarian response. A total of 912 infertile patients who underwent GnRH antagonist protocol between January 2017 to August 2019 at the Medical Center for Human Reproduction, Beijing Chao-Yang Hospital were included. All patients completed the full oocyte retrieval cycle and either had a live birth or had no embryos left. OSI was significantly lower in patients with a live birth (196.0 ± 120.4 in the live birth group vs 276.4 ± 235.7 in the non-live birth group, p < 0.001) while follicular output rate (FORT, defined as the ratio of pre-ovulatory follicle count on hCG day x 100/small antral follicle count at baseline) showed no significant difference. Patients were divided into low, average and high OSI groups and analysed in tertiles. From the low to the high OSI group, the cumulative live birth rate (CLBR) decreased dramatically (72.7 vs 67.2 vs 54.8%, p < 0.001). Multivariate regression analysis showed that OSI was an independent factor affecting CLBR (OR: 0.996, 95%CI: 0.995-0.998, p < 0.001) in our study population. In conclusion, OSI can be used as an independent indicator to distinguish fecundity in infertile patients with normal expected ovarian response and is probably more sensitive than FORT.

Race, ovarian responsiveness, and live birth after in vitro fertilization

OBJECTIVE

To determine if ovarian responsiveness to gonadotropin stimulation differs by race/ethnicity and whether this predicts live birth rates (LBRs) in non-White patients undergoing in vitro fertilization (IVF).

DESIGN

Retrospective cohort study.

SETTING

Academic infertility center.

PATIENT(S)

White, Asian, Black, and Hispanic patients undergoing ovarian stimulation for IVF.

INTERVENTION(S)

Self-reported race and ethnicity.

MAIN OUTCOME MEASURE(S)

The primary outcome was ovarian sensitivity index (OSI), defined as (the number of oocytes retrieved ÷ total gonadotropin dose) × 1,000 as a measure of ovarian responsiveness, adjusting for age, body mass index, infertility diagnosis, and cycle number. Secondary outcomes included live birth and clinical pregnancy after first retrievals, adjusting for age, infertility diagnosis, and history of fibroids, as well as miscarriage rate per clinical pregnancy, adjusting for age, body mass index, infertility diagnosis, duration of infertility, history of fibroids, and use of preimplantation genetic testing for aneuploidy.

RESULT(S)

The primary analysis of OSI included 3,360 (70.2%) retrievals from White patients, 704 (14.7%) retrievals from Asian patients, 553 (11.6%) retrievals from Black patients, and 168 (3.5%) retrievals from Hispanic patients. Black and Hispanic patients had higher OSIs than White patients after accounting for those with multiple retrievals and adjusting for confounders (6.08 in Black and 6.27 in Hispanic, compared with 5.25 in White). There was no difference in OSI between Asian and White patients. The pregnancy outcomes analyses included 2,299 retrievals. Despite greater ovarian responsiveness, Black and Hispanic patients had lower LBRs compared with White patients, although these differences were not statistically significant after adjusting for confounders (adjusted odds ratio, 0.83; 95% confidence interval [CI], 0.63-1.09, for Black; adjusted odds ratio, 0.93; 95% CI, 0.61-1.43, for Hispanic). Ovarian sensitivity index was modestly predictive of live birth in White and Asian patients but not in Black (area under the curve, 0.51; 95% CI, 0.38-0.64) and Hispanic (area under the curve, 0.50; 95% CI, 0.37-0.63) patients.

CONCLUSION(S)

Black and Hispanic patients have higher ovarian responsiveness to stimulation during IVF but do not experience a consequent increase in LBR. Factors beyond differences in responsiveness to ovarian stimulation need to be explored to address the racial/ethnic disparity established in prior literature.

Impact of Insulin Resistance on Ovarian Sensitivity and Pregnancy Outcomes in Patients with Polycystic Ovary Syndrome Undergoing IVF

BACKGROUND

Ovarian sensitivity index (OSI) is an accurate index to reflect the ovarian sensitivity to exogenous gonadotropins in in vitro fertilization (IVF). How insulin resistance (IR) affects OSI and pregnancy outcomes during IVF remains unclear.

METHODS

This was a large retrospective, cohort study. A total of 2055 women with polycystic ovary syndrome (PCOS) undergoing the first fresh IVF cycle were enrolled. They were grouped into terciles based on the homeostasis model assessment of insulin resistance (HOMA-IR) values as control, medium and IR group for comparison. Multivariate regression analysis was also conducted.

RESULTS

HOMA-IR had a significantly negative impact on OSI (adjusted β = -0.24; 95% CI, -0.35 to -0.13), especially in lean patients with an adjusted β of -0.33 (95% CI, -0.51 to -0.16). The interaction analysis revealed an interactive association between HOMA-IR and body mass index (BMI) (p = 0.017). IR was related to an increased early miscarriage risk independently with an odds ratio (OR) of 2.21 (95% CI, 1.13 to 4.33), without significant impact on pregnancy and live birth rate.

CONCLUSION

IR decreased the ovarian response in PCOS patients undergoing IVF, especially in the lean subgroup. IR may result in a higher risk of early miscarriage, but did not impair pregnancy and live birth rate.

Association between HOMA-IR and ovarian sensitivity index in women with PCOS undergoing ART: A retrospective cohort study

INTRODUCTION

Insulin resistance (IR) may play a central role in the pathophysiology of polycystic ovary syndrome (PCOS). Controlled ovarian stimulation (COS) in PCOS women in the setting of assisted reproductive technology (ART) is always a challenge for clinicians. However, it remains unclear whether IR in women with PCOS correlates with reduced ovarian sensitivity to exogenous gonadotropin (Gn). This study aimed to explore the association between homeostasis model assessment of insulin resistance (HOMA-IR) and ovarian sensitivity index (OSI).

METHODS

In this retrospective cohort study, we explored the association between Ln HOMA-IR and Ln OSI based on smoothing splines generated by generalized additive model (GAM). Then the correlation between HOMA-IR and OSI was further tested with a multivariable linear regression model and subgroup analysis.

RESULTS

1508 women with PCOS aged 20-39 years undergoing their first oocyte retrieval cycle were included consecutively between 2018 until 2022. We observed a negative association between Ln HOMA-IR and Ln OSI by using smoothing splines. In multivariable linear regression analysis, the inverse association between Ln HOMA-IR and Ln OSI was still found in PCOS women after adjustment for potential confounders (β = -0.18, 95% CI -0.25, -0.11). Compared with patients with the lowest tertile of HOMA-IR, those who had the highest tertile of HOMA-IR had lower OSI values (β = -0.25, 95% CI -0.36, -0.15).

DISCUSSION

Our study provided evidence for the inverse correlation between IR and the ovarian sensitivity during COS in PCOS women. Herein, we proposed new insights for individualized manipulation in PCOS patients with IR undergoing ART.

Association between chemical mixtures and female fertility in women undergoing assisted reproduction in Sweden and Estonia

OBJECTIVE

Women of reproductive age are exposed to ubiquitous chemicals such as phthalates, parabens, and per- and polyfluoroalkyl substances (PFAS), which have potential endocrine disrupting properties and might affect fertility. Our objective was to investigate associations between potential endocrine-disrupting chemicals (EDCs) and female fertility in two cohorts of women attending fertility clinics.

METHODS

In a total population of 333 women in Sweden and Estonia, we studied the associations between chemicals and female fertility, evaluating ovarian sensitivity index (OSI) as an indicator of ovarian response, as well as clinical pregnancy and live birth from fresh and frozen embryo transfers. We measured 59 chemicals in follicular fluid samples and detected 3 phthalate metabolites, di-2-ethylhexyl phthalate (DEHP) metabolites, 1 paraben, and 6 PFAS in >90% of the women. Associations were evaluated using multivariable-adjusted linear or logistic regression, categorizing EDCs into quartiles of their distributions, as well as with Bayesian Kernel Machine Regression.

RESULTS

We observed statistically significant lower OSI at higher concentrations of the sum of DEHP metabolites in the Swedish cohort (Q4 vs Q1, β = -0.21, 95% CI: -0.38, -0.05) and methylparaben in the Estonian cohort (Q3 vs Q1, β = -0.22, 95% CI: -0.44, -0.01). Signals of potential associations were also observed at higher concentrations of PFUnDA in both the combined population (Q2 vs. Q1, β = -0.16, 95% CI -0.31, -0.02) and the Estonian population (Q2 vs. Q1, β = -0.27, 95% CI -0.45, -0.08), and for PFOA in the Estonian population (Q4 vs. Q1, β = -0.31, 95% CI -0.61, -0.01). Associations of chemicals with clinical pregnancy and live birth presented wide confidence intervals.

CONCLUSIONS

Within a large chemical mixture, we observed significant inverse associations levels of DEHP metabolites and methylparaben, and possibly PFUnDA and PFOA, with OSI, suggesting that these chemicals may contribute to altered ovarian function and infertility in women.

Mixtures of persistent organic pollutants and ovarian function in women undergoing IVF

RESEARCH QUESTION

Do internal levels of persistent organic pollutants (POP) in serum and follicular fluid affect ovarian function of women attending IVF?

DESIGN

This cohort study included 136 women undergoing IVF in the assisted reproductive technology (ART) service of University Hospital from Nantes (France). Representative POP were measured using gas and liquid chromatography coupled with tandem mass spectrometry. Polyfluoroalkylated and perfluoroalkylated substances were measured in serum and polychlorinated biphenyls and organochlorinated pesticides in follicular fluid. Statistical associations between POP and ovarian reserve markers (anti-Müllerian Hormone [AMH] and antral follicle count [AFC], and ovarian responsiveness markers (Ovarian Sensitivity Index [OSI] and Follicular Output RaTe [FORT]), were explored in single and multipollutant regression models.

RESULTS

Twenty-seven out of 53 POP congeners were frequently detected in almost all women attending IVF. Adjusted models did not show statistically significant associations between POP and ovarian reserve markers. Positive associations were found between some POP, i.e. hexachlorobenzene with FORT (β 0.42, 95% CI 0.13 to 0.71, P = 0.005) or PCB52 with Ovarian Sensitivity Index (β 0.22; 95% CI, 0.07 to 0.38, P = 0.005). Negative associations between some polyfluoroalkylated and perfluoroalkylated substances, PCB189 and trans-nonachlor with AFC and AMH were found among current smokers.

CONCLUSIONS

Globally, associations between POP and the markers of ovarian function or responsiveness were lacking. Nonetheless, the stratification analysis suggested that current smoking could be a risk modifier, and extension of the study to a larger population sample size is needed.

Association of HOMA-IR with Ovarian Sensitivity Index in Women Undergoing IVF/ICSI: A Retrospective Cohort Study

BACKGROUND

Insulin resistance (IR) may adversely impact the in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) outcomes. However, the association of IR with ovarian sensitivity during controlled ovarian hyperstimulation (COH) remains controversial.

PURPOSE

We aimed to explore the association between homeostasis model assessment of insulin resistance (HOMA-IR) and ovarian sensitivity index (OSI).

PATIENTS AND METHODS

In this retrospective cohort study, 7836 patients aged between 20 and 39 years with good ovarian reserve were included consecutively between June 2018 and May 2022. HOMA-IR was calculated to evaluate the degree of IR and OSI was used to measure ovarian sensitivity. Owing to skewed distribution, HOMA-IR and OSI were log e transformed to Ln HOMA-IR and Ln OSI. Smoothing splines were generated by generalized additive models to explore the correlation between Ln HOMA-IR and Ln OSI. Then, the relationship between Ln HOMA-IR and Ln OSI was tested with multivariable linear regression model after adjustments for the potential confounders.

RESULTS

We observed negative association between Ln HOMA-IR and Ln OSI after adjustment for potential confounders by using smoothing spline fitting curves in generalized additive model. In the multivariable linear regression model, after full adjustments, each one unit increase in Ln HOMA-IR was related to reduced Ln OSI values (β = -0.12, 95% CI, -0.15, -0.09). As sensitivity analysis, those who had HOMA-IR of quartile 3 (HOMA-IR 2.33-3.43) and quartile 4 (HOMA-IR ≥3.43) had decreases in Ln OSI values (β = -0.07, 95% CI -0.11, -0.03 and β = -0.18, 95% CI -0.22, -0.13, respectively) when compared with patients in the quartile 1 (HOMA-IR <1.63). Furthermore, stratified and interaction analysis showed a strong inverse association of Ln HOMA-IR with Ln OSI in subgroups of PCOS patients and overweight/obese populations (P-value for interaction <0.0001).

CONCLUSION

HOMA-IR value was negatively associated with the ovarian sensitivity assessed by OSI. Among the PCOS and overweight/obese populations, this inverse relationship may be strong. These findings will increase the understanding of the contribution of IR to the development of decreased ovarian sensitivity within the assisted reproductive technology (ART) setting. It may have implications for optimizing gonadotropin dose manipulation in patients with IR undergoing IVF/ICSI.

Evaluation of correlation between antral follicle diameters with Follicular Output Rate (FORT) in women under controlled ovarian hyperstimulation for assisted reproductive techniques

Background

This study aimed to determine the association of diametrical antral follicles with the ovarian response by Follicular Output Rate (FORT) ratio in 100 females undergoing in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI). This study selected 100 women undergoing controlled ovarian hyperstimulation (COH). The number of antral follicles (diameter of 3–10 mm) was calculated with two-dimensional vaginal sonography on days 1–3 of the cycle. Then, on a triggering day with human chorionic gonadotropin (HCG), the number of follicles (with a diameter of 16–22 mm) and the ratio of FORT were determined. The correlation among FORT with age, antral follicle count (AFC), AFC ≤ 5, AFC > 5, number of preovulatory follicles (16–20 mm), number of metaphase II (MII) oocytes, body mass index (BMI), infertility period, and anti-Mullerian hormone (AMH) was assessed.

Results

There was a significant correlation between FORT and total AFC, AFC > 5, number of preovulatory follicles (16–20 mm), and number of MII oocytes retrieved. There is no significant relationship between FORT and examined two variables (AMH and AFC ≤ 5). Multiple linear regression analysis showed no significant relationship between FORT and examined two variables (AMH and AFC > 5). There was a significant correlation between MII oocytes retrieved and age, total AFC, AFC ≤ 5, AFC > 5, number of preovulatory follicles (16–20 mm), and AMH. A significant positive relationship existed between MII oocytes retrieved and examined two variables (AFC ≤ 5 and AFC > 5).

Conclusion

There was not much difference in the correlation between the AFC ≤ 5 and AFC > 5, and both positively correlated with the number of MII oocytes retrieved.

Prediction model of gonadotropin starting dose and its clinical application in controlled ovarian stimulation

Background

Selecting an appropriate and personalized Gn starting dose (GSD) is an essential procedure for determining the quality and quantity of oocytes in the controlled ovarian stimulation (COS) process of the in-vitro fertilization (IVF) treatment cycle. The current approach for determining the GSD is mainly based on the experience of a clinician, lacking unified and scientific standards. This study aims to establish a prediction model of GSD, based on which good COS outcomes can be achieved with the influencing factors comprehensively evaluated quantitatively.

Material and methods

We collected a total of 1555 patients undergoing the first oocytes retrieving cycle and conducted correlation analysis to find the significant factors related to the GSD. Two GSD models are built based on two popular machine learning approaches, and the one with better model performance is selected as the final model. Finally, clinical application and validation were conducted to verify the effectiveness of the proposed model.

Results

(1) Age, duration of infertility, type of infertility, body mass index (BMI), antral follicle count (AFC), basal follicle stimulating hormone (bFSH), estradiol (E₂), luteinizing hormone (LH), anti-Müllerian hormone (AMH) and COS treatment regimen were closely related to the GSD (P < 0.05). (2) The selected model has good modeling performance in terms of both root mean square error (RMSE) (29.87 ~ 34.21) and regression coefficient R (0.947 ~ 0.953). (3) A comprehensive evaluation of influencing factors for GSD is conducted and shows that the top four most significant factors are age, AMH, AFC, and BMI. (4) The proposed GSD can approximate the actual value well in the clinical application, with the mean absolute error of only 11.26 units, and the recommended results can prompt the number of oocytes retrieved (NOR) close to the optimal number.

Conclusion

Modeling the GSD value with machine learning approaches is feasible and effective, and the proposed model has good clinical application for determining the GSD in the IVF treatment cycle.

Are There Ovarian Responsive Indexes That Predict Cumulative Live Birth Rates in Women over 39 Years?

Objective: Ovarian response indexes have been proposed in assisted reproductive technology (ART) in order to optimize live birth rates (LBR), adjusting ovarian stimulation (OS), and minimizing risks. Gonadotropin doses are commonly adjusted according to ovarian reserve parameters, including antral follicle count (AFC), anti-Mullerian hormone (AMH), and basal follicle stimulating hormone (FSH) levels. The retrospective assessment of ovarian responses allows one to identify three primary indexes: (i) follicular output rate (FORT), the ratio of the number of pre-ovulatory follicles obtained at OS completion over AFC; (ii) follicle oocyte index (FOI), the ratio of oocytes retrieved over AFC; (iii) ovarian sensitivity index (OSI), the ratio of oocytes retrieved over the total gonadotropin dose administered. In recent publications, these indexes were reported to predict ART outcome. In the present study, we assessed the ability of these indexes to predict cumulative ART outcome in women ≥39 years. Materials and Methods: Retrospective cohort study. All patients ≥39 years who performed their first ART cycle with an antagonist protocol in our center between 01/2018 and 04/2020 were included. Patients with basal FSH > 20 IU/l, AMH < 0.1 ng/mL and severe male factors (azoospermia with testicular biopsy) were excluded. All patients received both recombinant FSH and human menopausal gonadotropin (hMG). Cumulative live birth rate (cLBR) was the primary outcome. Secondary outcomes included: the number of MII oocytes, cumulative implantation (cIR), and usable blastulation rates. Logistic regressions were performed to assess the predictive values of FORT, FOI, and OSI in cLBR and embryo culture success. For each parameter, the ability of the logistic regression models to predict embryo culture success was quantified by the area under the ROC curve (AUC). Only the significant findings related to FORT, FOI, and OSI were included in the multiple logistic regression model. Linear regression models were performed between cIR, cLB, FORT, FOI, and OSI. Each statistic model was adjusted for age. Concerning OR for OSI, values were multiplied *100 due to the very low value. Results: 429 patients met the inclusion criteria. There were 298 obtained usable blastocysts after ART treatment. Age-adjusted OSI was significantly associated with cLBR [OR = 17.58 95% CI (5.48−56.40), AUC = 0.707 95% CI (0.651−0.758)) and cIR (beta = 30.22 (SE: 7.88), p < 0.001, R2= 0.060). Both FOI (OR = 6.33 95% CI (3.27−12.25), AUC = 0.725 95% CI (0.675−0.771), R2 = 0.090, p < 0.001) and OSI (OSI*100; OR = 1808.93 95% CI (159.24−19,335.13), AUC = 0.790 95% CI (0.747−0.833), R2 = 0.156, p < 0.001) were independently, when age adjusted, associated with embryo culture success. OSI showed a main performance to explain successful embryo culture than FOI (R2 = 0.156 vs. R2 = 0.090, p < 0.001). In the age-adjusted linear regression model, FOI (R2 = 0.159, p < 0.001), OSI (R2 = 0.606, p < 0.001), and FORT (r2 = 0.030, p < 0.001) were predictive of the number of MII oocytes collected. Furthermore, for OSI (r2 = 0.759, p < 0.001) and FOI (r2 = 0.297, p < 0.001), the correlation with the number of metaphase II oocytes collected was significantly higher in the non-linear regression model. Conclusions: Our findings suggest that the best index, among those analyzed, to predict cIR and cLBR, is OSI. Both OSI and FOI predict embryo culture with success, but OSI is more accurate. OSI, FOI, and FORT are significantly related to the number of MII oocytes obtained.

Ovarian Sensitivity Decreased Significantly in Patients With Insulin Resistance Undergoing in vitro Fertilization and Embryo Transfer

Ovarian sensitivity could affect the outcome of in vitro fertilization and embryo transfer (IVF-ET). The objective of this study was to explore the relationship between the ovarian sensitivity index (OSI) and traditional ovarian response makers and observe the relationship between OSI and insulin resistance (IR). The patients enrolled in this study included 131 patients with polycystic ovary syndrome (PCOS) with IR (PCOS-IR), 52 patients with PCOS without IR (PCOS-N), 164 patients with control with IR (control-IR), 133 patients with control without IR (control-N), 295 patients with IR, 184 patients with non-IR, 183 patients with PCOS, and 297 patients with control (patients with non-PCOS). All patients received standard long protocol or the gonadotropin-releasing hormone (GnRH) antagonist protocol to induce follicular development. The two protocols downregulated the pituitary function or blocked the pituitary luteinizing hormone (LH) secretion with a GnRH antagonist. Both protocols can block premature LH surges because premature luteinization is not conducive to follicular development. All patients underwent IVF or intracytoplasmic sperm injection (ICSI). Embryo transfer was carried out according to the specific situation of each patient. The OSI was significantly reduced in patients with IR. The OSI had a significant positive relationship with anti-Müllerian hormone (AMH), antral follicle count (AFC), basal LH/follicle-stimulating hormone (FSH), dominant follicle number on trigger day, retrieved oocytes, embryo number, and high-quality embryo number. OSI had a significant negative relationship with age, body mass index (BMI), basal FSH, initial dose of Gn, and total dose of Gn. The receiver operating characteristic (ROC) curve of OSI demonstrated a better accuracy in distinguishing patients with positive pregnancy and clinical pregnancy, with an area under the curve (AUC) of 0.662 (95% CI, 0.598–0.727) and 0.636 (95% CI, 0.577–0.695), respectively. Patients could get a higher rate of dominant follicle count (p &lt; 0.0001) through the treatment of standard long protocol when compared with GnRH antagonist protocol. The OSI has a significant correlation with traditional ovarian response markers and could be a good predictor of positive pregnancy and clinical pregnancy for patients with IR.

Machine Learning-Based Modeling of Ovarian Response and the Quantitative Evaluation of Comprehensive Impact Features

Appropriate ovarian responses to the controlled ovarian stimulation strategy is the premise for a good outcome of the in vitro fertilization cycle. With the booming of artificial intelligence, machine learning is becoming a popular and promising approach for tailoring a controlled ovarian stimulation strategy. Nowadays, most machine learning-based tailoring strategies aim to generally classify the controlled ovarian stimulation outcome, lacking the capacity to precisely predict the outcome and evaluate the impact features. Based on a clinical cohort composed of 1365 women and two machine learning methods of artificial neural network and supporting vector regression, a regression prediction model of the number of oocytes retrieved is trained, validated, and selected. Given the proposed model, an index called the normalized mean impact value is defined and calculated to reflect the importance of each impact feature. The proposed models can estimate the number of oocytes retrieved with high precision, with the regression coefficient being 0.882% and 89.84% of the instances having the prediction number ≤ 5. Among the impact features, the antral follicle count has the highest importance, followed by the E2 level on the human chorionic gonadotropin day, the age, and the Anti-Müllerian hormone, with their normalized mean impact value > 0.3. Based on the proposed model, the prognostic results for ovarian response can be predicted, which enables scientific clinical decision support for the customized controlled ovarian stimulation strategies for women, and eventually helps yield better in vitro fertilization outcomes.

Women in Advanced Reproductive Age: Are the Follicular Output Rate, the Follicle-Oocyte Index and the Ovarian Sensitivity Index Predictors of Live Birth in an IVF Cycle?

(1) Background: Several researchers have investigated alternative markers related to ovarian responsiveness in order to better predict IVF outcomes, particularly in advanced reproductive-aged women. The follicular output rate (FORT), the follicle-oocyte index (FOI) and the ovarian sensitivity index (OSI) are among the most promising. However, these three metrics have not been investigated as independent predictors of live birth in women of advanced reproductive age; neither have they been compared to the two ‘component’ characteristics that are used to calculate them. (2) Methods: A logistic regression model containing all relevant predictors of ovarian reserve or response was used to evaluate the potential of FORT, FOI and OSI as predictors of live birth. After, the non-linear associations between FORT, FOI and OSI and the probability of live birth were evaluated. Finally, we fitted multiple logistic regression models to compare whether FORT, FOI and OSI were more informative predictors than their components. (3) Results: 590 couples received a total of 740 IVF cycles, after which, 127 (17.5%) obtained a live birth. None of FORT, FOI and OSI showed a strength of association or a p-value even close to female age (odds ratio for live birth (95% confidence interval) 1.00 (0.99–1.01), 1.00 (0.99–1.01), 0.98 (0.88–1.11) and 0.58 (0.48–0.72), respectively). The three models comparing FORT, FOI and OSI with the number of oocytes retrieved, the AFC, the number of preovulatory follicles and the FSH total dose were not more informative. (4) Conclusions: In a population of women of advanced age with unexplained infertility, none of FORT, FOI and OSI were predictive of live birth or more predictive than the two ‘component’ characteristics that were used to calculate them. We suggest clinicians and researchers still use female age as the most reliable predictor of an IVF treatment.

Association of follicle-to-oocyte index and clinical pregnancy in IVF treatment: A retrospective study of 4,323 fresh embryo transfer cycles

OBJECTIVE

The aim of this study is to investigate whether the follicle-to-oocyte index [FOI: (number of retrieved oocytes/antral follicle count) × 100] was associated with clinical pregnancy after fresh cleavage transfer.

DESIGN

The framework used to organize this study is retrospective cohort analysis.

SETTING

The study was performed in a single in vitro fertilization center in a public hospital.

PATIENTS

In total, 4,323 fresh embryo transfer cycles from 1 August 2011 to 31 January 2022 were retrospectively analyzed. Data were designated into three groups according to FOI tertile values.

INTERVENTIONS

There are no interventions in this study.

MAIN OUTCOME MEASURE

The primary outcome measure is the clinical pregnancy rate (CPR).

RESULTS

A total of 4,323 patients were included in the study. According to their FOI, patients were divided into low (FOI ≤ 0.70, n = 1,434), medium (FOI = 0.71-0.95, n = 1,070), and high (FOI = 0.96-1.00, n = 1,819) tertile groups. A significant statistical increase in the CPR from the lowest to the highest tertile FOI group was detected (47.28%, 51.78%, and 51.57%; P =0.026). After adjusted for potential confounders, multivariate logistic regression analysis revealed a positive association between FOI and CPR [odds ratio (OR) = 1.57; 95% confidence interval (CI): 1.18-2.11]. Each standard deviation increments in FOI (SD = 0.24) corresponded to a 20% increase in the CPR. Trend analysis also showed that FOI tertile groups were positively associated with CPR (P for trend = 0.010). Smooth curve fitting indicated the existence of a linear relationship across the entire range of FOI. No optimal cutoff value of FOI for prognosing CPR was found in smooth curve fitting analysis. Moreover, subgroup analyses suggested that the association was significantly stronger in the single cleavage transfer cycle (OR = 2.04; 95% CI: 1.14-3.65).

CONCLUSIONS

FOI is an independent variable in prediction for CPR in fresh embryo transfer cycle, especially in the single cleavage transfer cycle.

Single-cell RNA-seq analysis and cell-cluster deconvolution of the human preovulatory follicular fluid cells provide insights into the pathophysiology of ovarian hyporesponse

Reduction in responsiveness to gonadotropins or hyporesponsiveness may lead to the failure of in vitro fertilization (IVF), due to a low number of retrieved oocytes. The ovarian sensitivity index (OSI) is used to reflect the ovarian responsiveness to gonadotropin stimulation before IVF. Although introduced to clinical practice already years ago, its usefulness to predict clinical outcomes requires further research. Nevertheless, pathophysiological mechanisms of ovarian hyporesponse, along with advanced maternal age and in younger women, have not been fully elucidated. Follicles consist of multiple cell types responsible for a repertoire of biological processes including responding to pituitary gonadotropins necessary for follicle growth and oocyte maturation as well as ovulation. Encouraging evidence suggests that hyporesponse could be influenced by many contributing factors, therefore, investigating the variability of ovarian follicular cell types and their gene expression in hyporesponders is highly informative for increasing their prognosis for IVF live birth. Due to advancements in single-cell analysis technologies, the role of somatic cell populations in the development of infertility of ovarian etiology can be clarified. Here, somatic cells were collected from the fluid of preovulatory ovarian follicles of patients undergoing IVF, and RNA-seq was performed to study the associations between OSI and gene expression. We identified 12 molecular pathways differentially regulated between hypo- and normoresponder patient groups (FDR<0.05) from which extracellular matrix organization, post-translational protein phosphorylation, and regulation of Insulin-like Growth Factor (IGF) transport and uptake by IGF Binding Proteins were regulated age-independently. We then generated single-cell RNA-seq data from matching follicles revealing 14 distinct cell clusters. Using cell cluster-specific deconvolution from the bulk RNA-seq data of 18 IVF patients we integrated the datasets as a novel approach and discovered that the abundance of three cell clusters significantly varied between hypo- and normoresponder groups suggesting their role in contributing to the deviations from normal ovarian response to gonadotropin stimulation. Our work uncovers new information regarding the differences in the follicular gene expression between hypo- and normoresponders. In addition, the current study fills the gap in understanding the inter-patient variability of cell types in human preovulatory follicles, as revealed by single-cell analysis of follicular fluid cells.

Expectant Management Before In vitro Fertilization in Women Aged 39 or Above and Unexplained Infertility Does Not Decrease Live Birth Rates Compared to Immediate Treatment

Unexplained infertile couples can have further expectant management before starting assisted reproductive treatments. However, ovarian reserve and in vitro fertilization (IVF) outcomes rapidly decline after 39 years or more. It is thus important to clarify whether a waiting policy is also appropriate for women of advanced age. Couples who had access to a waiting list for approximately 1 year before receiving reimbursed public IVF were compared with those paying for access to immediate treatment. To allow for comparisons between these two strategies, we followed up couples who opted to pay for 1 year after the last embryo transfer from their first cycle. We calculated the proportion of live births in both groups and compared these using logistic regression models and a two-sample Z test for equality of proportions. Six hundred thirty-five couples were evaluated. Out of 359 couples in the immediate group, 70 (19.5%) had a live birth of which 11 after natural conception and 59 after IVF. Out of 276 couples in the waiting group, 57 (20.7%) had a live birth of which 37 after natural conception and 20 after IVF. There was no statistically significant difference between the two strategies in terms of the crude cumulative live birth rate (cLBR). The adjusted odds ratio of 0.69 (95%CI:0.39–1.22) did not change this conclusion as our sensitivity analyses. The cLBR for the ‘waiting before IVF’ and the ‘immediate’ strategies were similar. Further studies are needed to better characterize couples affected by unexplained infertility in order to individualize treatment strategies.

Does Ovarian Endometriosis Increase Oocyte Sensitivity to ICSI-Induced Mechanical Damage?

Some studies have shown that ICSI obtains poorer results than conventional IVF in women with ovarian endometriosis, suggesting that oocytes could be sensitive to ICSI-induced mechanical damage. The aims of this study were to clarify (a) whether ovarian endometriosis could induce peculiar fragility in the oocyte, so that ICSI would finally result harmful, and (b) whether endometrioma removal before IVF could be advisable in order to avoid any hypothetical detrimental effect. We retrospectively studied 368 women, 203 with in situ endometrioma (128 of which underwent ICSI, 75 conventional IVF) and 164 who received laparoscopic stripping of endometrioma before ICSI. For women with in situ endometrioma, cIVF and ICSI outcome was comparable for all parameters studied, including the clinical pregnancy rate per embryo transfer (PR/ET: 31.8% vs. 39.5% in the cIVF and ICSI groups) and cumulative live birth rate per ovum pick-up (CLBR/OPU: 24.4% vs. 27.7%). ICSI outcome was similar comparing women with in situ endometrioma and women previously submitted to laparoscopic stripping of cysts (CLPR/OPU 27.7% vs. 25.3%). Our findings suggest that (a) in women with in situ endometrioma ICSI may be performed, when needed, without harming oocytes and compromising the outcome and (b) that there is no advantage in removing endometrioma before ICSI.

Luteal Support with very Low Daily Dose of Human Chorionic Gonadotropin after Fresh Embryo Transfer as an Alternative to Cycle Segmentation for High Responders Patients Undergoing Gonadotropin-Releasing Hormone Agonist-Triggered IVF

The segmentation of the in vitro fertilization (IVF) cycle, consisting of the freezing of all embryos and the postponement of embryo transfer (ET), has become popular in recent years, with the main purpose of preventing ovarian hyperstimulation syndrome (OHSS) in patients with high response to controlled ovarian stimulation (COS). Indeed cycle segmentation (CS), especially when coupled to a GnRH-agonist trigger, was shown to reduce the incidence of OHSS in high-risk patients. However, CS increases the economic costs and the work amount for IVF laboratories. An alternative strategy is to perform a fresh ET in association with intensive luteal phase pharmacological support, able to overcome the negative effects of the GnRH-agonist trigger on the luteal phase and on endometrial receptivity. In order to compare these two strategies, we performed a retrospective, real-life cohort study including 240 non-polycystic ovarian syndrome (PCO) women with expected high responsiveness to COS (AMH >2.5 ng/mL), who received either fresh ET plus 100 IU daily human chorionic gonadotropin (hCG) as luteal support (FRESH group, n = 133), or cycle segmentation with freezing of all embryos and postponed ET (CS group, n = 107). The primary outcomes were: implantation rate (IR), live birth rate (LBR) after the first ET, and incidence of OHSS. Overall, significantly higher IR and LBR were observed in the CS group than in the FRESH group (42.9% vs. 27.8%, p < 0.05 and 32.7% vs. 19.5%, p < 0.05, respectively); the superiority of CS strategy was particularly evident when 16-19 oocytes were retrieved (LBR 42.2% vs. 9.5%, p = 0.01). Mild OHSS appeared with the same incidence in the two groups, whereas moderate and severe OHSS forms were observed only in the FRESH group (1.5% and 0.8%, respectively). In conclusion, in non-PCO women, high responders submitted to COS with the GnRH-antagonist protocol and GnRH-agonist trigger, CS strategy was associated with higher IR and LBR than the strategy including fresh ET followed by luteal phase support with a low daily hCG dose. CS appears to be advisable, especially when >15 oocytes are retrieved.