Randi H. Goldman

Predicting the likelihood of live birth for elective oocyte cryopreservation: a counseling tool for physicians and patients

Study question Can a counseling tool be developed for women desiring elective oocyte cryopreservation to predict the likelihood of live birth based on age and number of oocytes frozen? Summary answer Using data from ICSI cycles of a population of women with uncompromised ovarian reserve, an evidence-based counseling tool was created to guide women and their physicians regarding the number of oocytes needed to freeze for future family-building goals. What is known already Elective oocyte cryopreservation is increasing in popularity as more women delay family building. By undertaking elective oocyte freezing at a younger age, women hope to optimize their likelihood of successful live birth(s) using their thawed oocytes at a future date. Questions often arise in clinical practice regarding the number of cryopreserved oocytes sufficient to achieve live birth(s) and whether or not additional stimulation cycles are likely to result in a meaningful increase in the likelihood of live birth. As relatively few women who have electively cryopreserved oocytes have returned to use them, available data for counseling patients wishing to undergo fertility preservation are limited. Study design, size, duration A model was developed to determine the proportion of mature oocytes that fertilize and then form blastocysts as a function of age, using women with presumably normal ovarian reserve based on standard testing who underwent ICSI cycles in our program from January, 2011 through March, 2015 (n = 520). These included couples diagnosed exclusively with male-factor and/or tubal-factor infertility, as well as cycles utilizing egg donation. Age-specific probabilities of euploidy were estimated from 14 500 PGS embryo results from an external testing laboratory. Assuming survival of thawed oocytes at 95% for women <36 y and for egg donors, and 85% for women ≥36 y, and 60% live birth rate per transferred euploid blastocyst, probabilities of having at least one, two or three live birth(s) were calculated. Participants/materials, setting, method First fresh male-factor and/or tubal-factor only autologous ICSI cycles (n = 466) were analyzed using Poisson regression to calculate the probability that a mature oocyte will become a blastocyst based on age. Egg donation cycles (n = 54) were analyzed and incorporated into the model separately. The proportion of blastocysts expected to be euploid was determined using PGS results of embryos analyzed via array comparative genomic hybridization. A counseling tool was developed to predict the likelihood of live birth, based on individual patient age and number of mature oocytes. Main results and the role of chance This study provides an evidence-based model to predict the probability of a woman having at least one, two or three live birth(s) based on her age at egg retrieval and the number of mature oocytes frozen. The model is derived from a surrogate population of ICSI patients with uncompromised ovarian reserve. A user-friendly counseling tool was designed using the model to help guide physicians and patients. LIMITATIONS, REASONS FOR CAUTION The data used to develop the prediction model are, of necessity, retrospective and not based on patients who have returned to use their cryopreserved oocytes. The assumptions used to create the model, albeit reasonable and data-driven, vary by study and will likely vary by center. Centers are therefore encouraged to consider their own blastocyst formation and thaw survival rates when counseling patients. Limitations, reasons for caution Our model will provide a counseling resource that may help inform women desiring elective fertility preservation regarding their likelihood of live birth(s), how many cycles to undergo, and when additional cycles would bring diminishing returns. Study funding/competing interests None. Trial registration number Not applicable.

Outcomes after intrauterine insemination are independent of provider type

OBJECTIVE We sought to determine whether the success of intrauterine insemination (IUI) varies based on the type of health care provider performing the procedure. STUDY DESIGN This was a retrospective cohort study set at an infertility clinic at an academic institution. The patients who comprised this study were 1575 women who underwent 3475 IUI cycles from late 2003 through early 2012. Cycles were stratified into 3 groups according to the type of provider who performed the procedure: attending physician, fellow physician, or registered nurse (RN). The primary outcome was live birth. Additional outcomes of interest included positive pregnancy test and clinical pregnancy. Repeated measures log binomial regression was used to estimate the risk ratios (RR) and 95% confidence intervals (CI) for the outcomes and to evaluate the effect of potential confounders. All tests were 2-sided, and P values < .05 were considered statistically significant. RESULTS Of the 3475 IUI cycles, 2030 (58.4%) were gonadotropin stimulated, 929 (26.7%) were clomiphene citrate stimulated, and 516 (14.9%) were natural. The incidences of clinical pregnancy and live birth among all cycles were 11.8% and 8.8%, respectively. After adjusting for female age, male partner age, and cycle type, the incidence of live birth was similar for RNs compared with attending physicians (RR, 0.80; 95% CI, 0.58-1.1) and fellow physicians compared with attending physicians (RR, 0.84; 95% CI, 0.58-1.2). Similar results were seen for positive pregnancy test and clinical pregnancy. CONCLUSION There was no significant difference in live birth following IUI cycles in which the procedure was performed by a fellow physician or RN compared with an attending physician.

Patient-specific predictions of outcome after gonadotropin ovulation induction/intrauterine insemination.

OBJECTIVE To use patient-specific and cycle-specific characteristics to predict clinical pregnancy, multiple pregnancy, and spontaneous abortion rates after gonadotropin ovulation induction (OI)/IUI. DESIGN Retrospective chart review. SETTING Academic fertility center. PATIENT(S) A total of 1,438 women who underwent 3,375 gonadotropin OI/IUI cycles. INTERVENTION(S) Individual and cycle-specific characteristics were evaluated to determine predictors of the rates of clinical pregnancy, multiple pregnancy, and spontaneous abortion. Logistic regression using individual parameters was used to create predictive models. MAIN OUTCOME MEASURE(S) Clinical pregnancy (CPR), multiple pregnancy (MPR), and spontaneous abortion rates (SABR). RESULT(S) Multiple predictors were identified for CPR, MPR, and SABR. The presence of at least two follicles ≥ 13 mm at ovulation trigger significantly increased CPR (odds ratio [OR], 95% confidence interval [CI] = 1.45, 1.18-1.78) and MPR (OR, 95% CI = 5.17, 2.16-12.41). An E2 level >400 pg/mL significantly increased MPR (OR, 95% CI = 9.54, 2.31-39.42). Logistic regression models were developed for individualized predictions of outcome. CONCLUSION(S) Regression analysis reveals the patient and cycle-specific characteristics that are significant predictors of CPR, MPR, and SABR after OI/IUI. Logistic models using significant or nearly significant predictors for CPR, MPR, and SABR offer improved predictive power relative to simpler models, and allow for the development of a risk calculator for personalized patient counseling.

Uterine Transplantation: A Survey of Perceptions and Attitudes of American Reproductive Endocrinologists and Gynecologic Surgeons

OBJECTIVE To determine whether reproductive endocrinologists and minimally invasive surgeons support uterine transplantation as a treatment option for absolute uterine factor infertility (AUFI). DESIGN A cross-sectional study (Canadian Task Force classification II-2). SETTING A Web-based survey. PATIENTS Physician members of the American Society of Reproductive Medicine (ASRM) and the American Association of Gynecologic Laparoscopists (AAGL). INTERVENTIONS A Web-based questionnaire administered between January and February 2017. MEASUREMENTS AND MAIN RESULTS Support for (strongly agree or agree) or opposition to (strongly disagree or disagree) various aspects of uterine transplantation were described using descriptive statistics and analyzed using chi-square tests. A total of 414 physicians (ASRM: 49.5%, AAGL: 50.5%) responded to the Web-based survey; 43.7% were female, 52.4% were between the ages of 45 and 65 years, and 73.4% were white. Nearly fifty-six percent supported women being allowed to donate or receive a transplanted uterus. Fifty-four percent strongly agreed or agreed that uterine transplantation carried an acceptable risk for donors, 28.0% for the recipient and 21.0% for the infant. Forty-two percent agreed that uterine transplantation should be considered a therapeutic option for women with AUFI, whereas 19.6% felt it should be covered by insurance. Nearly 45% of respondents felt uterine transplantation to be ethical. The most common ethical concerns regarding uterine transplantation were related to medical or surgical complications to the recipient (48.8%). CONCLUSION Just under half of the reproductive endocrinologists and minimally invasive surgeons surveyed find uterine transplantation to be an ethical option for patients with AUFI. Important concerns remain regarding the risk to donors, recipients, and resulting infants, all contributing to only a minority currently recommending it as a therapeutic option.

Public support for intergenerational oocyte donation in the United States

OBJECTIVE To determine whether the general public supports intergenerational oocyte donation. DESIGN Cross-sectional study. SETTING Not applicable. PATIENT(S) A nationally representative sample based on age distribution of United States residents. INTERVENTIONS(S) Not applicable. MAIN OUTCOME MEASURE(S) Characteristics of respondents who supported (strongly agree and agree) various oocyte donation practices were compared with participants who did not support them (disagree and strongly disagree) using log binomial regression to calculate risk ratios (RRs) and 95% confidence intervals of support (95% CIs). Models were adjusted for age, gender, and religion to yield adjusted risk ratios (aRR). RESULT(S) A total of 1,915 people responded to the Web-based survey; 53% were female, and 24% were racial/ethnic minorities. Eighty-five percent had prior knowledge of oocyte donation, and 74% felt that a woman should be able to donate oocytes to a family member. The desire to help a family member was the most commonly perceived motivation for donors (79%). Christian-Catholics compared with Christian-non-Catholics (aRR 0.91, 95% CI 0.86-0.98), African Americans compared with non-Hispanic Caucasians (aRR 0.86, 95% CI 0.76-0.97), and Republicans compared with Democrats (RR 0.93, 95% CI 0.88-0.98) were less likely to support intergenerational oocyte donation. Respondents with three or more biological children (RR 1.06, 95% CI 1.00-1.11) compared with those with no children were less likely to support this practice. Eight percent of participants disapproved of donation to any family member. The most common reason for disapproval was the potential negative impact on the child (53%). CONCLUSION(S) A majority of Americans support the practice of intergenerational oocyte donation; however, support varies according to demographic characteristics.

Response from the authors re: Letter to the Editor for our manuscript “The cost of a euploid embryo identified from preimplantation genetic testing for aneuploidy (PGT-A): a counseling tool”

We appreciate the letter writer’s interest in our manuscript and agree that their two referenced articles add to the growing literature on both the efficacy and cost-effectiveness of PGTA. The goals and theoretical models implemented in each paper are distinct, and it is therefore challenging to draw direct comparisons amongst their conclusions. In our study, we developed a novel model to determine the expected out-ofpocket costs of IVF with PGT-A based on individual age and AMH. However, we strongly disagree that PGT-A should be confined to study protocols. On the contrary, we feel that PGT should be available and offered, and that counseling patients regarding this technology should include a discussion of estimated costs. We believe that, following a discussion of the benefits, limitations, and costs of PGT-A, well-informed patients should be able to decide what is right for them.

The association between quality of supernumerary embryos in a cohort and implantation potential of the transferred blastocyst

PurposeDespite studies focused on the association between embryo morphology and implantation potential, it is unknown how the collective quality of the supernumerary embryos in a cohort is associated with the implantation rate (IR) of the transferred embryo. This study tested the hypothesis that a relationship exists between the quality of the supernumerary cohort and IR.MethodsA retrospective cohort study of first fresh autologous IVF cycles from 05/2012 to 09/2016, with ≥ 3 blastocysts, resulting in a single blastocyst transfer (n = 819) was performed. Cohorts were grouped in two ways: by mean priority score (PS; 1 being best) of supernumerary embryos and by percent supernumerary embryos with low implantation potential. The relationship between cohort quality and IR was assessed using logistic regression.ResultsAs mean cohort PS increased, IR of the transferred embryo decreased (test for linear trend, p = 0.05). When ≥ 75% of the supernumerary cohort was predicted to have low implantation potential, IR of the transferred embryo was significantly lower compared to when < 75% of the cohort was predicted to have low implantation potential (OR 0.71; 95% CI (0.53–0.94)). All associations were attenuated when adjusting for PS of the transferred embryo.ConclusionsOur findings suggest that quality of supernumerary embryos is associated with IR of the transferred embryo, among patients with ≥ 3 blastocysts available on day 5. As cohort quality declines and the proportion of low implantation potential embryos increases, the IR of the transferred embryo declines. These associations are attenuated when controlling for quality of the transferred embryo, suggesting that the relationship between embryo cohort quality and implantation is not independent of the transferred embryo quality.

Are women with obesity and infertility willing to attempt weight loss prior to fertility treatment

OBJECTIVE To assess attitudes towards weight loss interventions in patients seeking infertility treatment. METHODS We evaluated prior weight loss experiences, attitudes towards future interventions by body mass index (BMI), and willingness to delay fertility treatment for weight loss interventions stratified by BMI using logistic regression amongst women ≤45years old with infertility over three months or recurrent pregnancy loss. RESULTS The average age of our convenience sample of respondents (148 of 794 eligible women, 19%) was 34.5 years old, with a mean BMI of 26.7±7.4kg/m2, including 37 with a BMI >30kg/m2 (25%). Most women had attempted conception over 1year. The majority of women with overweight or obesity were attempting weight loss at the time of survey completion (69%). While 47% of these women reported interest in a supervised medical weight loss program, 92% of overweight women and 84% of women with obesity were not willing to delay fertility treatment more than 3 months to attempt weight loss. CONCLUSION Most women with obesity and infertility in our population are unwilling to postpone fertility treatment for weight loss interventions.