- Email: [email protected]
Does empiric superovulation improve fecundity in healthy women undergoing therapeutic donor insemination without a male partner?
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
Does empiric superovulation improve fecundity in healthy women undergoing therapeutic donor insemination without a male partner? Leah Hawkins Bressler, M.D., M.P.H., Brittany Papworth, M.D., Sarah Moustafa, M.D., Audrey Chang, B.A., and Jennifer E. Mersereau, M.D., M.S.C.I. Division of Reproductive Epidemiology and Infertility, Department of Obstetrics and Gynecology, University of North Carolina, Chapel Hill, North Carolina
Objective: To evaluate whether superovulation improves fecundity in women undergoing therapeutic donor insemination (TDI). Design: Retrospective cohort study. Setting: University-affiliated fertility clinic. Patient(s): Healthy women aged 23–45 years with no history of or risk factors for infertility who underwent 152 medicated and 104 unmedicated TDI cycles from 2013 to 2018. Intervention: Unmedicated TDI versus use of medication in a TDI cycle (clomiphene citrate or letrozole). Main Outcome Measure(s): Cumulative probability of pregnancy in six TDI cycles. Result(s): In adjusted all-cycle analysis, medicated TDI cycles were less likely to result in pregnancy compared with unmedicated cycles. The incidence of twins was 23% in the medicated group and 0% in the unmedicated group. Medicated cycles were less likely to result in pregnancy in women younger than 40 years or with an antim€ ullerian hormone (AMH) level >1.2. After three cycles not resulting in pregnancy, the only women who conceived were those who crossed over from an unmedicated to a medicated cycle (12% vs. 0%). Conclusion(s): Patients undergoing unmedicated TDI cycles had higher fecundity and no incidence of twin gestations. Older women, those with low AMH, and those who fail to conceive after three unmedicated cycles may benefit from medication. (Fertil SterilÒ 2019;-:-–-. Ó2019 by American Society for Reproductive Medicine.) Key Words: Donor insemination, IUI, superovulation, clomiphene citrate, letrozole Discuss: You can discuss this article with its authors and other readers at https://www.fertstertdialog.com/users/16110-fertilityand-sterility/posts/52893-27957
W
omen without a male partner or those in a same-sex relationship commonly seek conception through therapeutic donor insemination (TDI) (1). According to a survey study of assisted reproductive technology clinics, 74%–79% offer services to women without a male partner (2). Recently, the American Society for Reproductive Medicine
issued a report affirming the right of these individuals to fertility services (3). The rising need to provide services to meet this population's fertility goals implies the need for an increase in the study of women without male partners who are seeking pregnancy. Numerous factors influence the success of TDI cycles. Age of the female partner is inversely associated with
Received March 14, 2019; revised July 12, 2019; accepted August 29, 2019. L.H.B. has nothing to disclose. B.P. has nothing to disclose. S.M. has nothing to disclose. A.C. has nothing to disclose. J.E.M. has nothing to disclose. Supported by the National Center for Advancing Translational Sciences, National Institutes of Health (NIH), through grant award no. UL1TR001111. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Reprint requests: Leah Hawkins Bressler, M.D., M.P.H., 4001 Old Clinic Building, CB#7570, Chapel Hill, NC 27599-7570 (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2019 0015-0282/$36.00 Copyright ©2019 American Society for Reproductive Medicine, Published by Elsevier Inc. https://doi.org/10.1016/j.fertnstert.2019.08.101 VOL. - NO. - / - 2019
success of TDI (4). Studies show that intrauterine insemination (IUI) is superior to intracervical insemination (5, 6), double insemination does not yield higher pregnancy rates than single (7), and triggering ovulation with injection of hCG does not confer an increase in pregnancy rate (8). Ovulatory women undergoing TDI can undergo natural-cycle insemination or empiric superovulation (9). In couples with unexplained infertility, combining superovulation with IUI synergistically improves cycle fecundity (10). Medications for superovulation are often empirically administered to ovulatory patients undergoing TDI in attempts to limit the monetary, emotional, and 1
ORIGINAL ARTICLE: ASSISTED REPRODUCTION time-intensive investment of TDI (11). The primary potential risk of oral medications for superovulation is multiple gestation (12). Whether empiric superovulation in healthy women without a male partner undergoing TDI increases fecundity, fecundability (cycle fecundity), or risk of multiple gestation has not been well characterized. Two investigations evaluated the impact of empiric superovulation on TDI outcome. A retrospective study by Ferrera et al. from 1993 to 1997 compared fecundity among 261 women undergoing TDI in unmedicated, clomiphene citrate (CC), and injectable gonadotropin cycles and found no differences (13%, 7.2%, and 11%, respectively) (1). In 2016, El Hachem et al. compared the impact of CC or letrozole in 257 women undergoing up to three TDI cycles each and identified no differences in live birth rate (16.5% vs. 11.5%, respectively) or multiple gestation (11.6% vs. 8.7%, respectively) (13). These investigations did not assess the impact of crossover between treatment arms and did not compare the three most common approaches encountered in current clinical practice: unmedicated, CC, and letrozole. The present study evaluates the impact of empiric medication for superovulation on cumulative pregnancy rate (fecundity), cycle-specific pregnancy rate (fecundability), and multiple gestations among ovulatory women without male partners undergoing TDI. We hypothesized that among healthy women undergoing TDI, fecundity and fecundability would be no different but that the incidence of multiple gestation would be lower in unmedicated than in medicated cycles.
MATERIALS AND METHODS All women presenting to the University of North Carolina at Chapel Hill Fertility Center from July 2013 through April 2018 were screened for inclusion in this retrospective cohort. Eligible participants were identified with the use of a prospectively maintained clinical database. Healthy regularly cycling women aged 23–45 years without male partners who presented for a TDI cycle were eligible for inclusion. Exclusion criteria included a history of endometriosis, pelvic infection, oligo-ovulation, malignancy, or a TDI sample with <5 million total motile sperm. Demographic and clinical data were abstracted from the clinical database and electronic medical records and maintained securely with the use of Redcap (14). The protocol was approved by the Institutional Review Board. The use of medication (CC or letrozole) during a TDI cycle was defined as the exposure. Women in an unmedicated or natural TDI cycle constituted the unexposed group. The decision to use medication for superovulation or to proceed with TDI without medication was made by the primary physician; the clinic does not have a standard policy regarding this treatment. Superovulation was defined as identification of more than one mature (R14 mm) follicle at the time of midcycle ultrasound. Not all women underwent midcycle ultrasound; women able to detect native ovulations with the use of ovulation predictor kits did not undergo routine midcycle ultrasound monitoring nor did those with a previously established response to medications. The primary outcome 2
was conception, defined as positive urine or serum hCG that was either reported or documented. Nonparametric bivariate analyses were used to compare demographic and clinical variables according to medication status. Continuous variables are presented as mean SD and range. All models were adjusted for Society for Assisted Reproductive Technology age category (<35, 35–37, 37–39, 39–41, or >42 y) as an a priori confounder as well as for covariates identified as potential confounders in univariate analyses (15). Results were analyzed by means of all-cycle and first cycle–only analysis to account for the propensity of the former model to bias against early success and the latter model to select for it (16). In addition, first cycle–only analysis addressed potential heterogeneity among subjects, whereby women with more than one cycle may experience suboptimal outcomes. Conception was assessed as the cumulative probability of conception in all-cycle and crossover analyses and as cycle-specific fecundability in analysis restricted to a woman's first-ever TDI cycle. All-cycle analysis was accomplished with the use of a Cox proportional hazards model with time defined as menstrual cycles culminating in TDI. Because time was measured by menstrual cycles, hazard ratios are referred to as fecundability ratios (FRs), which are the relative probability of pregnancy in a given menstrual cycle for the exposed compared with unexposed group, as previously described (17). Proportional hazard models were used to calculate the cumulative probability of conception after six cycles of attempt. Kaplan-Meier curves with 95% confidence intervals (CIs) were also constructed. Multivariate-adjusted logistic regression was used to generate effect estimates in analyses restricted to the first cycle only. Sensitivity analyses were performed to identify whether threshold values for age, body mass index (BMI), and antim€ ullerian hormone (AMH) levels existed such that the direction of the association between exposure and outcome changed. All testing was two sided. Statistical significance was assumed for P< .05. SAS 15.1 (SAS Institute) software was used for all data analysis. A convenience sample of all TDI cycles performed in the available time period was used. Given the sample size of 256 cycles, we had the power to detect a 12% difference in fecundability between the two groups (17).
RESULTS A total of 101 women met the inclusion criteria and underwent a total of 256 (152 unmedicated and 104 medicated) cycles that were eligible for analysis. The majority of patients were white, nulligravid, and overweight (Table 1). Compared with women who underwent unmedicated cycles, those whose first cycles were medicated cycles were older (34.3 vs. 38.0 y, respectively; P< .01, Table 1). In a multivariateadjusted analysis restricted to the first cycles only, the use of medications did not significantly increase the odds of pregnancy (adjusted odds ratio [aOR] 1.07, 95% CI 0.32–3.60; P¼.92). Comparing all cycles, women on medication were older (35.1 vs. 34.2 years; P< .01) and underwent more cycles VOL. - NO. - / - 2019
Fertility and Sterility®
TABLE 1 Patient demographic and clinical characteristics (first cycle; n [ 101 patients). Characteristic Age (y) Mean SD Range SART age category <35 y 35–37 y 38–40 y 41–42 y >42 y Race White Black Asian Other Unknown BMI (kg/m2) Mean SD Range €llerian hormone (ng/mL) Antimu Mean SD Range Gravidity 0 1 2 3 4 Parity 0 1 2 3 4 Mature follicles (>14 mm) Mean SD Range Endometrial thickness (mm) Mean SD Range Trigger shot use Total motile sperm (millions) Mean SD Range Cycle outcome Pregnant Not pregnant Missing Pregnancy outcome Live birth Spontaneous abortion Unknown
Unmedicated (n [ 81; 80%)
Medicated (n [ 20; 20%)
n ¼ 81 (100%) 33.4 4.3 23.3–42.4 n ¼ 81 (100%) 55 (68%) 14 (17%) 7 (9%) 5 (6%) 0 (0%) n ¼ 81 (100%) 61 (75%) 7 (9%) 1 (1%) 1 (1%) 11 (14%) n ¼ 81 (100%) 28.1 7.2 17.6–52.8 n ¼ 57 (70%) 3.1 2.1 0.08–8.83 n ¼ 81 (100%) 63 (78%) 14 (17%) 2 (2%) 1 (1%) 1 (1%) n ¼ 81 (100%) 72 (89%) 8 (10%) 0 (0%) 0 (0%) 1 (1%) n ¼ 19 (23%) 1.0 0.3 0.0–2.0 n ¼ 19 (23%) 9.4 2.0 5.8–14.4 4 (5%) n ¼ 79 (98%) 18.5 8.8 7.1–54.5 n ¼ 81 (100%) 22 (27.2%) 58 (71.6%) 1 (1.2%) n ¼ 19 (23.4%) 13 (68.0%) 1 (5.2%) 5 (26.3%)
n ¼ 20 (100%) 36.7 5.0 26.1–44.0 n ¼ 20 (100%) 6 (30%) 4 (20%) 7 (35%) 1 (5%) 2 (10%) n ¼ 20 (100%) 17 (85%) 0 (0%) 0 (0%) 0 (0%) 3 (15%) n ¼ 20 (100%) 27.8 5.4 20.8–40.2 n ¼ 15 (75%) 3.3 3.4 0.7–14.4 n ¼ 20 (100%) 14 (70%) 3 (15%) 2 (10%) 0 (0%) 1 (5%) n ¼ 20 (100%) 16 (80%) 3 (15%) 0 (0%) 1 (5%) 0 (0%) n ¼ 16 (80%) 2.1 1.1 0.0–4.0 n ¼ 15 (75%) 8.5 1.3 6.6–10.5 2 (10%) n ¼ 20 (100%) 18.2 11.7 7.5–64.5 n ¼ 20 (100%) 8 (40%) 12 (60%) 0 (0%) n ¼ 8 (40%) 6 (75%) 0 (0%) 2 (25%)
P value .004 < .001
.61
.76 .79 .22
.29
< .001 .12 .22 .9 .28
.94
Note: Mantel-Haentzel odds and Wilcoxon rank sum were used as appropriate. BMI ¼ body mass index; SART ¼ Society for Assisted Reproductive Technologies. Hawkins Bressler. Empiric superovulation for donor insemination. Fertil Steril 2019.
than unmedicated women (four vs. two cycles; P< .001) but had similar AMH levels and BMI (Supplemental Table 1, available online at www.fertstert.org). Cumulative pregnancy rate was highest among women in unmedicated cycles (23.0%) compared with medicated cycles (20.6% CC, 11.8% letrozole). In multivariate-adjusted all-cycle analysis, medicated TDI cycles were less likely to result in pregnancy than unmedicated cycles (adjusted FR [aFR] 0.48, 95% CI 0.26– 0.88; P¼.02; Fig. 1; Table 2). Cycle fecundability in medicated cycles was 5.3% (6.9% CC, 2.6% letrozole), compared with VOL. - NO. - / - 2019
12.8% in unmedicated cycles. Compared with women in unmedicated cycles, women on CC did not experience significantly different fecundability (aFR 0.52, 95% CI 0.27–1.01; P¼.05), but women on letrozole had 81% lower fecundability (aFR 0.19, 95% CI 0.06–0.66; P¼.009; Table 2). Among women who conceived, women in medicated cycles experienced a higher incidence of twin gestation (23.1%, 3/13) compared with those in unmedicated cycles (0%, 0/18; P¼.03; Supplemental Table 1). Subgroup analysis in the 102 cycles in which midcycle ultrasound was performed revealed 3
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
1.0
TABLE 2 Fecundability ratios in therapeutic donor insemination cycles by medication.
0.8
FIGURE 1
Type of cycle
Cumulative Probability of Pregnancy
0.4
0.6
Unmedicated cycle Medicated cycle Clomiphene Letrozole
Fecundability ratioa (95% CI%)
P value
Referent 0.48 (0.26–0.88%) 0.52 (0.27–1.10%) 0.19 (0.06–0.66%)
1.0 .02 .05 .009
0.2
Note: CI ¼ confidence interval. a Adjusted for Society for Assisted Reproductive Technologies age category.
0
Hawkins Bressler. Empiric superovulation for donor insemination. Fertil Steril 2019.
0
1
2
95% CI Unmedicated
3 Cycle Number 95% CI Medicated
4
5
6
95% CI Crossover
Cumulative likelihood of pregnancy by treatment group, with 95% confidence intervals (CIs). Hawkins Bressler. Empiric superovulation for donor insemination. Fertil Steril 2019.
that patients who achieved superovulation had a nonsignificantly higher cumulative pregnancy rate than those that did not (25.8% vs. 10%; P¼.05). Sensitivity analyses were performed to evaluate for the presence of a threshold age, BMI level, or AMH level at which medicated cycles resulted in a higher fecundability ratio compared with unmedicated cycles. Findings demonstrated that at age R40 years, medication significantly improved fecundability (P¼.01). AMH <1.2 ng/mL was associated with a lower likelihood of conception (aFR 0.09, CI: 0.01–0.82; P¼.01), and medication improved fecundability in these women (P¼.01). In women with a BMI R44 kg/m2, medication improved fecundability but this affected just four women in the cohort and did not achieve significance (P>.05). Time to conception was analyzed via Kaplan-Meier curves (Fig. 1). Findings demonstrated a nonsignificantly lower FR in women in medicated compared with unmedicated cycles (FR 0.56, 95% CI 0.23–1.36; P¼.20), evidenced by overlapping 95% CIs (Fig. 1). Given that some clinicians recommend initiating medication after three unsuccessful TDI cycles, we performed additional analyses evaluating whether medication was associated with higher pregnancy rate after this clinical time point. After three cycles not resulting in pregnancy, crossing over from unmedicated to medicated cycles (n ¼ 79 cycles) compared to staying in an unmedicated cycle (n ¼ 10) was associated with improved cumulative pregnancy rate (12% vs. 0%). As such, statistical comparisons perfectly predicted pregnancy in women who crossed over to medications after their third unsuccessful TDI cycle (Fig. 1).
DISCUSSION The main results showed that unmedicated cycles conferred the highest cumulative and cycle-specific odds of conception and the lowest incidence of multiple gestations. Sensitivity analyses highlighted age and AMH cutoffs at which empiric 4
initiation of medicated TDI cycles may be of benefit. At age R40 years and AMH level <1.2 ng/mL, medication was associated with higher fecundability. The findings of the present work suggest that a reasonable strategy may be a trial of three unmedicated TDI cycles in young women with adequate ovarian reserve and then adding medication for those who have not conceived. Although the only women to conceive at age R40 years or with AMH level <1.2 ng/mL were in medicated cycles, no such significant trend was identified for BMI. Considering the synergistic adverse maternal and fetal outcomes associated with multiple gestation in women with class III and class IV obesity, we propose that avoiding superovulation may be preferred in that population to avoid added risks associated with multiple gestation. Our results demonstrate that women with AMH <1.2 ng/mL were more likely to conceive in a medicated than a natural cycle. Steiner et al. compared categoric AMH levels (<0.7, 0.7–8.4, or R8.5 ng/mL) as a predictor of time to conception and found that it was not associated with time to conception in healthy women (17). In contrast, the present study assessed the impact of oral medications on fecundity and fecundability in healthy women and identified that AMH, evaluated as a dichotomous variable, was associated with time to conception. The disparity in our results may reflect a difference in exposure ascertainment or a difference in the impact of AMH on spontaneous conception compared with conception in a fertility-treatment cycle. Crossover analysis was limited by low sample size but showed that the only conceptions occurring in women undergoing more than three unmedicated TDI cycles were in women who crossed over to medications. Although it was outside the scope of this project, it may be hypothesized that women who do not conceive after three medicated cycles conceive most efficiently via more aggressive treatments, such as gonadotropin/IUI or in vitro fertilization (Fig. 1) (18). Our findings are inconsistent with those from Ferrera et al., and in contrast to their null findings, we identified higher cycle-specific fecundability and fecundity rates among women in unmedicated TDI cycles. A key difference between studies is Ferrera et al.'s inclusion of women on injectable gonadotropins in the exposed group and our inclusion of women on letrozole (1). El Hachem et al. compared live birth rates between women undergoing TDI after letrozole vs. CC treatment and found no difference (13). Our study was not powered to assess live birth rate, but our study design VOL. - NO. - / - 2019
Fertility and Sterility® included an unmedicated group that permitted evaluation of the impact of medication on fecundability overall and by type of medication. We observed lower fecundability in women on letrozole than in women on CC but were not powered to compare directly between groups. Our finding is in keeping with observations made by Diamond et al., who identified a lower pregnancy rate in women with unexplained infertility undergoing IUI after letrozole compared with CC, though the difference did not achieve significance in a direct comparison (19). Other main results demonstrated a significantly higher incidence of twin gestation in medicated cycles compared with unmedicated cycles (23% vs. 0%). The incidence of twin gestation in our cohort is higher than the 11.6% prevalence observed by Hachem et al. in women on CC undergoing TDI (13) and much higher than the 3% national prevalence of twin births in the United States (20). The former may be partly explained by low sample size, whereby 62% of women in our study who conceived in a medicated TDI cycle had an early obstetrical ultrasound. Strengths of the present study include strict inclusion criteria to closely mimic a growing population of healthy women desiring TDI. Unlike previous investigations, this was the first to compare an unmedicated group with a group receiving CC or letrozole and the first to evaluate the impact of oral medications in women undergoing TDI for more than three attempts. In addition, the inclusion of crossover cycles, in which women who failed to conceive in unmedicated cycles crossed over to medicated cycles, mirrors common clinical practice and provides useful foundational literature. Access to the complete medical record permitted a low rate of loss to follow-up, whereby <3% of patients had an unknown cycle outcome. Our final analysis included 256 cycles, which was an adequate sample according to our calculation which demonstrated that a sample size of 256 cycles could detect a 12% difference in fecundity between groups with 80% power at a P value of 0.05.18 In the present investigation, there was a <3% difference in the cumulative 6-month pregnancy rate between groups (23% unmedicated vs. 20.6% medicated), indicating that a larger-scale investigation is merited to definitively rule out type II error. Evaluation of the impact of letrozole compared with CC would require evaluation of nearly 1,000 cycles. Although we sought to evaluate factors associated with TDI success in the first six consecutive treatment cycles, it is possible that women underwent at-home inseminations before presenting to care. Our investigation is limited by retrospective design at a single center. Evaluation of crossover between medicated and unmedicated treatments was limited by low sample size. Inclusion of women over the age of 40 years may have introduced a risk factor for infertility (21), and although we accounted for this with multivariable-adjusted regression models, residual confounding is possible. Similarly, AMH levels were not checked for all women without male partners seeking TDI. Data from Steiner et al.'s Time to Conceive Study shows that AMH is a poor indicator of six-cycle fecundity or time to conception, and therefore, in a non–insurance-mandated state, it is not our clinic practice to recommend routine AMH testing for VOL. - NO. - / - 2019
women without risk factors for infertility (17). The study population was >75% white, so results may not be readily generalized to clinics that serve a more ethnically diverse population. Future investigations may include a cost analysis, because medication for superovulation incurs cost of both medication and, typically, ultrasound monitoring. The concept of cost is particularly relevant for same-sex couples in light of research demonstrating that the medicalization of childbearing in fertility clinics is undesirable (20). The present study suggests that adding medication does not improve fecundability until after three failed TDI cycles, or potentially sooner in women aged R40 years or with an AMH level <1.2 ng/mL. In contrast, some women present to care with the assumption that empiric medication improves fecundability. In an era of focus on personalized medicine and emphasis on patient satisfaction, patient preferences about treatment options can be significant. Our findings demonstrated an increased risk of multiple gestation with medication and suggests that medication should be used judiciously in goodprognosis patients seeking TDI. This study evaluated a population that is likely to grow in the coming years: single women and same-sex partners that desire pregnancy. Key results suggest that in younger women with adequate ovarian reserve, medication does not improve TDI cycle fecundability. Empiric medication may be considered in older women, those with an AMH level <1.2 ng/mL, and those who have not conceived after three unmedicated TDI cycles. These findings may be used to improve counseling, standardize practice models, and potentially mitigate financial and emotional cost associated with repeated or more invasive treatments. Reproduction of results in a prospective randomized investigation would lend credence to the results presented here.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
Ferrara I, Balet R, Grudzinskas JG. Intrauterine donor insemination in single women and lesbian couples: a comparative study of pregnancy rates. Hum Reprod 2000;15:621–5. Stern JE, Cramer CP, Garrod A, Green RM. Access to services at assisted reproductive technology clinics: a survey of policies and practices. Am J Obstet Gynecol 2001;184:591–7. Ethics Committee, American Society for Reproductive Medicine. Access to fertility treatment by gays, lesbians, and unmarried persons: a committee opinion. Fertil Steril 2013;100:1524–7. Brzechffa PR, Daneshmand S, Buyalos RP. Sequential clomiphene citrate and human menopausal gonadotrophin with intrauterine insemination: the effect of patient age on clinical outcome. Hum Reprod 1998;13:2110–4. Wainer R, Merlet F, Ducot B, Bailly M, Tribalat S, Lombroso R. Prospective randomized comparison of intrauterine and intracervical insemination with donor spermatozoa. Hum Reprod 1995;10:2919–22. Guzick DS, Carson SA, Coutifaris C, Overstreet JW, Factor-Litvak P, Steinkampf MP, et al. National Cooperative Reproductive Medicine Network. Efficacy of superovulation and intrauterine insemination in the treatment of infertility. N Engl J Med 1999;340:177–83. Khalifa Y, Redgment CJ, Tsirigotis M, Grudzinskas JG, Craft IL. The value of single versus repeated insemination in intra-uterine donor insemination cycles. Hum Reprod 1995;10:153–4. Awonuga A, Govindbhai J. Is waiting for an endogenous luteinizing hormone surge and/or administration of human chorionic gonadotrophin of benefit in intrauterine insemination? Hum Reprod 1999;14:1765–70. 5
ORIGINAL ARTICLE: ASSISTED REPRODUCTION 9.
10. 11. 12.
13.
14.
15.
6
Kyrou D, Kolibianakis EM, Fatemi HM, Grimbizis GF, Theodoridis TD, Camus M, et al. Spontaneous triggering of ovulation versus HCG administration in patients undergoing IUI: a prospective randomized study. Reprod Biomed Online 2012;25:278–83. Cohlen B, Ombelet W. Intra-uterine Insemination: Evidence Based Guidelines for Daily Practice. CRC Press; 2013. Wallach EE. Gonadotropin treatment for the ovulatory patient—the pros and cons of empiric therapy for infertility. Fertil Steril 1991;55:478–80. Liu A, Zheng C, Lang J, Chen W. Letrozole versus clomiphene citrate for unexplained infertility: a systematic review and meta-analysis. J Obstet Gynaecol Res 2014;40:1205–16. El Hachem H, Antaki R, Sylvestre C, Kadoch IJ, Lapensee L, Bouet PE. Clomiphene citrate versus letrozole for ovarian stimulation in therapeutic donor sperm insemination. Gynecol Obstet Invest 2017;82:481–6. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989;79:340–9.
16.
17.
18.
19.
20. 21.
Missmer SA, Pearson KR, Ryan LM, Meeker JD, Cramer DW, Hauser R. Analysis of multiple-cycle data from couples undergoing in vitro fertilization: methodologic issues and statistical approaches. Epidemiology 2011;22: 497–504. Steiner AZ, Pritchard D, Stanczyk FZ, Kesner JS, Meadows JW, Herring AH, et al. Association between biomarkers of ovarian reserve and infertility among older women of reproductive age. JAMA 2017;318:1367–76. Reindollar RH, Regan MM, Neumann PJ, Levine BS, Thornton KL, Alper MM, et al. A randomized clinical trial to evaluate optimal treatment for unexplained infertility: the Fast Track and Standard Treatment (FASTT) trial. Fertil Steril 2010;94:888–99. Diamond MP, Legro RS, Coutifaris C, Alvero R, Robinson RD, Casson P, et al. Letrozole, gonadotropin, or clomiphene for unexplained infertility. N Engl J Med 2015;373:1230–40. Hamilton BE, Martin JA, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for 2014. Natl Vital Stat Rep 2015;64:1–64. Franasiak JM, Forman EJ, Hong KH, Werner MD, Upham KM, Treff NR, et al. The nature of aneuploidy with increasing age of the female partner: a review of 15,169 consecutive trophectoderm biopsies evaluated with comprehensive chromosomal screening. Fertil Steril 2014;101:656–63.e1.
VOL. - NO. - / - 2019
Does empiric superovulation improve fecundity in healthy women undergoing therapeutic donor insemination without a male partner? Leah Hawkins Bressler, M.D., M.P.H., Brittany Papworth, M.D., Sarah Moustafa, M.D., Audrey Chang, B.A., and Jennifer E. Mersereau, M.D., M.S.C.I. Division of Reproductive Epidemiology and Infertility, Department of Obstetrics and Gynecology, University of North Carolina, Chapel Hill, North Carolina
Objective: To evaluate whether superovulation improves fecundity in women undergoing therapeutic donor insemination (TDI). Design: Retrospective cohort study. Setting: University-affiliated fertility clinic. Patient(s): Healthy women aged 23–45 years with no history of or risk factors for infertility who underwent 152 medicated and 104 unmedicated TDI cycles from 2013 to 2018. Intervention: Unmedicated TDI versus use of medication in a TDI cycle (clomiphene citrate or letrozole). Main Outcome Measure(s): Cumulative probability of pregnancy in six TDI cycles. Result(s): In adjusted all-cycle analysis, medicated TDI cycles were less likely to result in pregnancy compared with unmedicated cycles. The incidence of twins was 23% in the medicated group and 0% in the unmedicated group. Medicated cycles were less likely to result in pregnancy in women younger than 40 years or with an antim€ ullerian hormone (AMH) level >1.2. After three cycles not resulting in pregnancy, the only women who conceived were those who crossed over from an unmedicated to a medicated cycle (12% vs. 0%). Conclusion(s): Patients undergoing unmedicated TDI cycles had higher fecundity and no incidence of twin gestations. Older women, those with low AMH, and those who fail to conceive after three unmedicated cycles may benefit from medication. (Fertil SterilÒ 2019;-:-–-. Ó2019 by American Society for Reproductive Medicine.) Key Words: Donor insemination, IUI, superovulation, clomiphene citrate, letrozole Discuss: You can discuss this article with its authors and other readers at https://www.fertstertdialog.com/users/16110-fertilityand-sterility/posts/52893-27957
W
omen without a male partner or those in a same-sex relationship commonly seek conception through therapeutic donor insemination (TDI) (1). According to a survey study of assisted reproductive technology clinics, 74%–79% offer services to women without a male partner (2). Recently, the American Society for Reproductive Medicine
issued a report affirming the right of these individuals to fertility services (3). The rising need to provide services to meet this population's fertility goals implies the need for an increase in the study of women without male partners who are seeking pregnancy. Numerous factors influence the success of TDI cycles. Age of the female partner is inversely associated with
Received March 14, 2019; revised July 12, 2019; accepted August 29, 2019. L.H.B. has nothing to disclose. B.P. has nothing to disclose. S.M. has nothing to disclose. A.C. has nothing to disclose. J.E.M. has nothing to disclose. Supported by the National Center for Advancing Translational Sciences, National Institutes of Health (NIH), through grant award no. UL1TR001111. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Reprint requests: Leah Hawkins Bressler, M.D., M.P.H., 4001 Old Clinic Building, CB#7570, Chapel Hill, NC 27599-7570 (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2019 0015-0282/$36.00 Copyright ©2019 American Society for Reproductive Medicine, Published by Elsevier Inc. https://doi.org/10.1016/j.fertnstert.2019.08.101 VOL. - NO. - / - 2019
success of TDI (4). Studies show that intrauterine insemination (IUI) is superior to intracervical insemination (5, 6), double insemination does not yield higher pregnancy rates than single (7), and triggering ovulation with injection of hCG does not confer an increase in pregnancy rate (8). Ovulatory women undergoing TDI can undergo natural-cycle insemination or empiric superovulation (9). In couples with unexplained infertility, combining superovulation with IUI synergistically improves cycle fecundity (10). Medications for superovulation are often empirically administered to ovulatory patients undergoing TDI in attempts to limit the monetary, emotional, and 1
ORIGINAL ARTICLE: ASSISTED REPRODUCTION time-intensive investment of TDI (11). The primary potential risk of oral medications for superovulation is multiple gestation (12). Whether empiric superovulation in healthy women without a male partner undergoing TDI increases fecundity, fecundability (cycle fecundity), or risk of multiple gestation has not been well characterized. Two investigations evaluated the impact of empiric superovulation on TDI outcome. A retrospective study by Ferrera et al. from 1993 to 1997 compared fecundity among 261 women undergoing TDI in unmedicated, clomiphene citrate (CC), and injectable gonadotropin cycles and found no differences (13%, 7.2%, and 11%, respectively) (1). In 2016, El Hachem et al. compared the impact of CC or letrozole in 257 women undergoing up to three TDI cycles each and identified no differences in live birth rate (16.5% vs. 11.5%, respectively) or multiple gestation (11.6% vs. 8.7%, respectively) (13). These investigations did not assess the impact of crossover between treatment arms and did not compare the three most common approaches encountered in current clinical practice: unmedicated, CC, and letrozole. The present study evaluates the impact of empiric medication for superovulation on cumulative pregnancy rate (fecundity), cycle-specific pregnancy rate (fecundability), and multiple gestations among ovulatory women without male partners undergoing TDI. We hypothesized that among healthy women undergoing TDI, fecundity and fecundability would be no different but that the incidence of multiple gestation would be lower in unmedicated than in medicated cycles.
MATERIALS AND METHODS All women presenting to the University of North Carolina at Chapel Hill Fertility Center from July 2013 through April 2018 were screened for inclusion in this retrospective cohort. Eligible participants were identified with the use of a prospectively maintained clinical database. Healthy regularly cycling women aged 23–45 years without male partners who presented for a TDI cycle were eligible for inclusion. Exclusion criteria included a history of endometriosis, pelvic infection, oligo-ovulation, malignancy, or a TDI sample with <5 million total motile sperm. Demographic and clinical data were abstracted from the clinical database and electronic medical records and maintained securely with the use of Redcap (14). The protocol was approved by the Institutional Review Board. The use of medication (CC or letrozole) during a TDI cycle was defined as the exposure. Women in an unmedicated or natural TDI cycle constituted the unexposed group. The decision to use medication for superovulation or to proceed with TDI without medication was made by the primary physician; the clinic does not have a standard policy regarding this treatment. Superovulation was defined as identification of more than one mature (R14 mm) follicle at the time of midcycle ultrasound. Not all women underwent midcycle ultrasound; women able to detect native ovulations with the use of ovulation predictor kits did not undergo routine midcycle ultrasound monitoring nor did those with a previously established response to medications. The primary outcome 2
was conception, defined as positive urine or serum hCG that was either reported or documented. Nonparametric bivariate analyses were used to compare demographic and clinical variables according to medication status. Continuous variables are presented as mean SD and range. All models were adjusted for Society for Assisted Reproductive Technology age category (<35, 35–37, 37–39, 39–41, or >42 y) as an a priori confounder as well as for covariates identified as potential confounders in univariate analyses (15). Results were analyzed by means of all-cycle and first cycle–only analysis to account for the propensity of the former model to bias against early success and the latter model to select for it (16). In addition, first cycle–only analysis addressed potential heterogeneity among subjects, whereby women with more than one cycle may experience suboptimal outcomes. Conception was assessed as the cumulative probability of conception in all-cycle and crossover analyses and as cycle-specific fecundability in analysis restricted to a woman's first-ever TDI cycle. All-cycle analysis was accomplished with the use of a Cox proportional hazards model with time defined as menstrual cycles culminating in TDI. Because time was measured by menstrual cycles, hazard ratios are referred to as fecundability ratios (FRs), which are the relative probability of pregnancy in a given menstrual cycle for the exposed compared with unexposed group, as previously described (17). Proportional hazard models were used to calculate the cumulative probability of conception after six cycles of attempt. Kaplan-Meier curves with 95% confidence intervals (CIs) were also constructed. Multivariate-adjusted logistic regression was used to generate effect estimates in analyses restricted to the first cycle only. Sensitivity analyses were performed to identify whether threshold values for age, body mass index (BMI), and antim€ ullerian hormone (AMH) levels existed such that the direction of the association between exposure and outcome changed. All testing was two sided. Statistical significance was assumed for P< .05. SAS 15.1 (SAS Institute) software was used for all data analysis. A convenience sample of all TDI cycles performed in the available time period was used. Given the sample size of 256 cycles, we had the power to detect a 12% difference in fecundability between the two groups (17).
RESULTS A total of 101 women met the inclusion criteria and underwent a total of 256 (152 unmedicated and 104 medicated) cycles that were eligible for analysis. The majority of patients were white, nulligravid, and overweight (Table 1). Compared with women who underwent unmedicated cycles, those whose first cycles were medicated cycles were older (34.3 vs. 38.0 y, respectively; P< .01, Table 1). In a multivariateadjusted analysis restricted to the first cycles only, the use of medications did not significantly increase the odds of pregnancy (adjusted odds ratio [aOR] 1.07, 95% CI 0.32–3.60; P¼.92). Comparing all cycles, women on medication were older (35.1 vs. 34.2 years; P< .01) and underwent more cycles VOL. - NO. - / - 2019
Fertility and Sterility®
TABLE 1 Patient demographic and clinical characteristics (first cycle; n [ 101 patients). Characteristic Age (y) Mean SD Range SART age category <35 y 35–37 y 38–40 y 41–42 y >42 y Race White Black Asian Other Unknown BMI (kg/m2) Mean SD Range €llerian hormone (ng/mL) Antimu Mean SD Range Gravidity 0 1 2 3 4 Parity 0 1 2 3 4 Mature follicles (>14 mm) Mean SD Range Endometrial thickness (mm) Mean SD Range Trigger shot use Total motile sperm (millions) Mean SD Range Cycle outcome Pregnant Not pregnant Missing Pregnancy outcome Live birth Spontaneous abortion Unknown
Unmedicated (n [ 81; 80%)
Medicated (n [ 20; 20%)
n ¼ 81 (100%) 33.4 4.3 23.3–42.4 n ¼ 81 (100%) 55 (68%) 14 (17%) 7 (9%) 5 (6%) 0 (0%) n ¼ 81 (100%) 61 (75%) 7 (9%) 1 (1%) 1 (1%) 11 (14%) n ¼ 81 (100%) 28.1 7.2 17.6–52.8 n ¼ 57 (70%) 3.1 2.1 0.08–8.83 n ¼ 81 (100%) 63 (78%) 14 (17%) 2 (2%) 1 (1%) 1 (1%) n ¼ 81 (100%) 72 (89%) 8 (10%) 0 (0%) 0 (0%) 1 (1%) n ¼ 19 (23%) 1.0 0.3 0.0–2.0 n ¼ 19 (23%) 9.4 2.0 5.8–14.4 4 (5%) n ¼ 79 (98%) 18.5 8.8 7.1–54.5 n ¼ 81 (100%) 22 (27.2%) 58 (71.6%) 1 (1.2%) n ¼ 19 (23.4%) 13 (68.0%) 1 (5.2%) 5 (26.3%)
n ¼ 20 (100%) 36.7 5.0 26.1–44.0 n ¼ 20 (100%) 6 (30%) 4 (20%) 7 (35%) 1 (5%) 2 (10%) n ¼ 20 (100%) 17 (85%) 0 (0%) 0 (0%) 0 (0%) 3 (15%) n ¼ 20 (100%) 27.8 5.4 20.8–40.2 n ¼ 15 (75%) 3.3 3.4 0.7–14.4 n ¼ 20 (100%) 14 (70%) 3 (15%) 2 (10%) 0 (0%) 1 (5%) n ¼ 20 (100%) 16 (80%) 3 (15%) 0 (0%) 1 (5%) 0 (0%) n ¼ 16 (80%) 2.1 1.1 0.0–4.0 n ¼ 15 (75%) 8.5 1.3 6.6–10.5 2 (10%) n ¼ 20 (100%) 18.2 11.7 7.5–64.5 n ¼ 20 (100%) 8 (40%) 12 (60%) 0 (0%) n ¼ 8 (40%) 6 (75%) 0 (0%) 2 (25%)
P value .004 < .001
.61
.76 .79 .22
.29
< .001 .12 .22 .9 .28
.94
Note: Mantel-Haentzel odds and Wilcoxon rank sum were used as appropriate. BMI ¼ body mass index; SART ¼ Society for Assisted Reproductive Technologies. Hawkins Bressler. Empiric superovulation for donor insemination. Fertil Steril 2019.
than unmedicated women (four vs. two cycles; P< .001) but had similar AMH levels and BMI (Supplemental Table 1, available online at www.fertstert.org). Cumulative pregnancy rate was highest among women in unmedicated cycles (23.0%) compared with medicated cycles (20.6% CC, 11.8% letrozole). In multivariate-adjusted all-cycle analysis, medicated TDI cycles were less likely to result in pregnancy than unmedicated cycles (adjusted FR [aFR] 0.48, 95% CI 0.26– 0.88; P¼.02; Fig. 1; Table 2). Cycle fecundability in medicated cycles was 5.3% (6.9% CC, 2.6% letrozole), compared with VOL. - NO. - / - 2019
12.8% in unmedicated cycles. Compared with women in unmedicated cycles, women on CC did not experience significantly different fecundability (aFR 0.52, 95% CI 0.27–1.01; P¼.05), but women on letrozole had 81% lower fecundability (aFR 0.19, 95% CI 0.06–0.66; P¼.009; Table 2). Among women who conceived, women in medicated cycles experienced a higher incidence of twin gestation (23.1%, 3/13) compared with those in unmedicated cycles (0%, 0/18; P¼.03; Supplemental Table 1). Subgroup analysis in the 102 cycles in which midcycle ultrasound was performed revealed 3
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
1.0
TABLE 2 Fecundability ratios in therapeutic donor insemination cycles by medication.
0.8
FIGURE 1
Type of cycle
Cumulative Probability of Pregnancy
0.4
0.6
Unmedicated cycle Medicated cycle Clomiphene Letrozole
Fecundability ratioa (95% CI%)
P value
Referent 0.48 (0.26–0.88%) 0.52 (0.27–1.10%) 0.19 (0.06–0.66%)
1.0 .02 .05 .009
0.2
Note: CI ¼ confidence interval. a Adjusted for Society for Assisted Reproductive Technologies age category.
0
Hawkins Bressler. Empiric superovulation for donor insemination. Fertil Steril 2019.
0
1
2
95% CI Unmedicated
3 Cycle Number 95% CI Medicated
4
5
6
95% CI Crossover
Cumulative likelihood of pregnancy by treatment group, with 95% confidence intervals (CIs). Hawkins Bressler. Empiric superovulation for donor insemination. Fertil Steril 2019.
that patients who achieved superovulation had a nonsignificantly higher cumulative pregnancy rate than those that did not (25.8% vs. 10%; P¼.05). Sensitivity analyses were performed to evaluate for the presence of a threshold age, BMI level, or AMH level at which medicated cycles resulted in a higher fecundability ratio compared with unmedicated cycles. Findings demonstrated that at age R40 years, medication significantly improved fecundability (P¼.01). AMH <1.2 ng/mL was associated with a lower likelihood of conception (aFR 0.09, CI: 0.01–0.82; P¼.01), and medication improved fecundability in these women (P¼.01). In women with a BMI R44 kg/m2, medication improved fecundability but this affected just four women in the cohort and did not achieve significance (P>.05). Time to conception was analyzed via Kaplan-Meier curves (Fig. 1). Findings demonstrated a nonsignificantly lower FR in women in medicated compared with unmedicated cycles (FR 0.56, 95% CI 0.23–1.36; P¼.20), evidenced by overlapping 95% CIs (Fig. 1). Given that some clinicians recommend initiating medication after three unsuccessful TDI cycles, we performed additional analyses evaluating whether medication was associated with higher pregnancy rate after this clinical time point. After three cycles not resulting in pregnancy, crossing over from unmedicated to medicated cycles (n ¼ 79 cycles) compared to staying in an unmedicated cycle (n ¼ 10) was associated with improved cumulative pregnancy rate (12% vs. 0%). As such, statistical comparisons perfectly predicted pregnancy in women who crossed over to medications after their third unsuccessful TDI cycle (Fig. 1).
DISCUSSION The main results showed that unmedicated cycles conferred the highest cumulative and cycle-specific odds of conception and the lowest incidence of multiple gestations. Sensitivity analyses highlighted age and AMH cutoffs at which empiric 4
initiation of medicated TDI cycles may be of benefit. At age R40 years and AMH level <1.2 ng/mL, medication was associated with higher fecundability. The findings of the present work suggest that a reasonable strategy may be a trial of three unmedicated TDI cycles in young women with adequate ovarian reserve and then adding medication for those who have not conceived. Although the only women to conceive at age R40 years or with AMH level <1.2 ng/mL were in medicated cycles, no such significant trend was identified for BMI. Considering the synergistic adverse maternal and fetal outcomes associated with multiple gestation in women with class III and class IV obesity, we propose that avoiding superovulation may be preferred in that population to avoid added risks associated with multiple gestation. Our results demonstrate that women with AMH <1.2 ng/mL were more likely to conceive in a medicated than a natural cycle. Steiner et al. compared categoric AMH levels (<0.7, 0.7–8.4, or R8.5 ng/mL) as a predictor of time to conception and found that it was not associated with time to conception in healthy women (17). In contrast, the present study assessed the impact of oral medications on fecundity and fecundability in healthy women and identified that AMH, evaluated as a dichotomous variable, was associated with time to conception. The disparity in our results may reflect a difference in exposure ascertainment or a difference in the impact of AMH on spontaneous conception compared with conception in a fertility-treatment cycle. Crossover analysis was limited by low sample size but showed that the only conceptions occurring in women undergoing more than three unmedicated TDI cycles were in women who crossed over to medications. Although it was outside the scope of this project, it may be hypothesized that women who do not conceive after three medicated cycles conceive most efficiently via more aggressive treatments, such as gonadotropin/IUI or in vitro fertilization (Fig. 1) (18). Our findings are inconsistent with those from Ferrera et al., and in contrast to their null findings, we identified higher cycle-specific fecundability and fecundity rates among women in unmedicated TDI cycles. A key difference between studies is Ferrera et al.'s inclusion of women on injectable gonadotropins in the exposed group and our inclusion of women on letrozole (1). El Hachem et al. compared live birth rates between women undergoing TDI after letrozole vs. CC treatment and found no difference (13). Our study was not powered to assess live birth rate, but our study design VOL. - NO. - / - 2019
Fertility and Sterility® included an unmedicated group that permitted evaluation of the impact of medication on fecundability overall and by type of medication. We observed lower fecundability in women on letrozole than in women on CC but were not powered to compare directly between groups. Our finding is in keeping with observations made by Diamond et al., who identified a lower pregnancy rate in women with unexplained infertility undergoing IUI after letrozole compared with CC, though the difference did not achieve significance in a direct comparison (19). Other main results demonstrated a significantly higher incidence of twin gestation in medicated cycles compared with unmedicated cycles (23% vs. 0%). The incidence of twin gestation in our cohort is higher than the 11.6% prevalence observed by Hachem et al. in women on CC undergoing TDI (13) and much higher than the 3% national prevalence of twin births in the United States (20). The former may be partly explained by low sample size, whereby 62% of women in our study who conceived in a medicated TDI cycle had an early obstetrical ultrasound. Strengths of the present study include strict inclusion criteria to closely mimic a growing population of healthy women desiring TDI. Unlike previous investigations, this was the first to compare an unmedicated group with a group receiving CC or letrozole and the first to evaluate the impact of oral medications in women undergoing TDI for more than three attempts. In addition, the inclusion of crossover cycles, in which women who failed to conceive in unmedicated cycles crossed over to medicated cycles, mirrors common clinical practice and provides useful foundational literature. Access to the complete medical record permitted a low rate of loss to follow-up, whereby <3% of patients had an unknown cycle outcome. Our final analysis included 256 cycles, which was an adequate sample according to our calculation which demonstrated that a sample size of 256 cycles could detect a 12% difference in fecundity between groups with 80% power at a P value of 0.05.18 In the present investigation, there was a <3% difference in the cumulative 6-month pregnancy rate between groups (23% unmedicated vs. 20.6% medicated), indicating that a larger-scale investigation is merited to definitively rule out type II error. Evaluation of the impact of letrozole compared with CC would require evaluation of nearly 1,000 cycles. Although we sought to evaluate factors associated with TDI success in the first six consecutive treatment cycles, it is possible that women underwent at-home inseminations before presenting to care. Our investigation is limited by retrospective design at a single center. Evaluation of crossover between medicated and unmedicated treatments was limited by low sample size. Inclusion of women over the age of 40 years may have introduced a risk factor for infertility (21), and although we accounted for this with multivariable-adjusted regression models, residual confounding is possible. Similarly, AMH levels were not checked for all women without male partners seeking TDI. Data from Steiner et al.'s Time to Conceive Study shows that AMH is a poor indicator of six-cycle fecundity or time to conception, and therefore, in a non–insurance-mandated state, it is not our clinic practice to recommend routine AMH testing for VOL. - NO. - / - 2019
women without risk factors for infertility (17). The study population was >75% white, so results may not be readily generalized to clinics that serve a more ethnically diverse population. Future investigations may include a cost analysis, because medication for superovulation incurs cost of both medication and, typically, ultrasound monitoring. The concept of cost is particularly relevant for same-sex couples in light of research demonstrating that the medicalization of childbearing in fertility clinics is undesirable (20). The present study suggests that adding medication does not improve fecundability until after three failed TDI cycles, or potentially sooner in women aged R40 years or with an AMH level <1.2 ng/mL. In contrast, some women present to care with the assumption that empiric medication improves fecundability. In an era of focus on personalized medicine and emphasis on patient satisfaction, patient preferences about treatment options can be significant. Our findings demonstrated an increased risk of multiple gestation with medication and suggests that medication should be used judiciously in goodprognosis patients seeking TDI. This study evaluated a population that is likely to grow in the coming years: single women and same-sex partners that desire pregnancy. Key results suggest that in younger women with adequate ovarian reserve, medication does not improve TDI cycle fecundability. Empiric medication may be considered in older women, those with an AMH level <1.2 ng/mL, and those who have not conceived after three unmedicated TDI cycles. These findings may be used to improve counseling, standardize practice models, and potentially mitigate financial and emotional cost associated with repeated or more invasive treatments. Reproduction of results in a prospective randomized investigation would lend credence to the results presented here.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
Ferrara I, Balet R, Grudzinskas JG. Intrauterine donor insemination in single women and lesbian couples: a comparative study of pregnancy rates. Hum Reprod 2000;15:621–5. Stern JE, Cramer CP, Garrod A, Green RM. Access to services at assisted reproductive technology clinics: a survey of policies and practices. Am J Obstet Gynecol 2001;184:591–7. Ethics Committee, American Society for Reproductive Medicine. Access to fertility treatment by gays, lesbians, and unmarried persons: a committee opinion. Fertil Steril 2013;100:1524–7. Brzechffa PR, Daneshmand S, Buyalos RP. Sequential clomiphene citrate and human menopausal gonadotrophin with intrauterine insemination: the effect of patient age on clinical outcome. Hum Reprod 1998;13:2110–4. Wainer R, Merlet F, Ducot B, Bailly M, Tribalat S, Lombroso R. Prospective randomized comparison of intrauterine and intracervical insemination with donor spermatozoa. Hum Reprod 1995;10:2919–22. Guzick DS, Carson SA, Coutifaris C, Overstreet JW, Factor-Litvak P, Steinkampf MP, et al. National Cooperative Reproductive Medicine Network. Efficacy of superovulation and intrauterine insemination in the treatment of infertility. N Engl J Med 1999;340:177–83. Khalifa Y, Redgment CJ, Tsirigotis M, Grudzinskas JG, Craft IL. The value of single versus repeated insemination in intra-uterine donor insemination cycles. Hum Reprod 1995;10:153–4. Awonuga A, Govindbhai J. Is waiting for an endogenous luteinizing hormone surge and/or administration of human chorionic gonadotrophin of benefit in intrauterine insemination? Hum Reprod 1999;14:1765–70. 5
ORIGINAL ARTICLE: ASSISTED REPRODUCTION 9.
10. 11. 12.
13.
14.
15.
6
Kyrou D, Kolibianakis EM, Fatemi HM, Grimbizis GF, Theodoridis TD, Camus M, et al. Spontaneous triggering of ovulation versus HCG administration in patients undergoing IUI: a prospective randomized study. Reprod Biomed Online 2012;25:278–83. Cohlen B, Ombelet W. Intra-uterine Insemination: Evidence Based Guidelines for Daily Practice. CRC Press; 2013. Wallach EE. Gonadotropin treatment for the ovulatory patient—the pros and cons of empiric therapy for infertility. Fertil Steril 1991;55:478–80. Liu A, Zheng C, Lang J, Chen W. Letrozole versus clomiphene citrate for unexplained infertility: a systematic review and meta-analysis. J Obstet Gynaecol Res 2014;40:1205–16. El Hachem H, Antaki R, Sylvestre C, Kadoch IJ, Lapensee L, Bouet PE. Clomiphene citrate versus letrozole for ovarian stimulation in therapeutic donor sperm insemination. Gynecol Obstet Invest 2017;82:481–6. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989;79:340–9.
16.
17.
18.
19.
20. 21.
Missmer SA, Pearson KR, Ryan LM, Meeker JD, Cramer DW, Hauser R. Analysis of multiple-cycle data from couples undergoing in vitro fertilization: methodologic issues and statistical approaches. Epidemiology 2011;22: 497–504. Steiner AZ, Pritchard D, Stanczyk FZ, Kesner JS, Meadows JW, Herring AH, et al. Association between biomarkers of ovarian reserve and infertility among older women of reproductive age. JAMA 2017;318:1367–76. Reindollar RH, Regan MM, Neumann PJ, Levine BS, Thornton KL, Alper MM, et al. A randomized clinical trial to evaluate optimal treatment for unexplained infertility: the Fast Track and Standard Treatment (FASTT) trial. Fertil Steril 2010;94:888–99. Diamond MP, Legro RS, Coutifaris C, Alvero R, Robinson RD, Casson P, et al. Letrozole, gonadotropin, or clomiphene for unexplained infertility. N Engl J Med 2015;373:1230–40. Hamilton BE, Martin JA, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for 2014. Natl Vital Stat Rep 2015;64:1–64. Franasiak JM, Forman EJ, Hong KH, Werner MD, Upham KM, Treff NR, et al. The nature of aneuploidy with increasing age of the female partner: a review of 15,169 consecutive trophectoderm biopsies evaluated with comprehensive chromosomal screening. Fertil Steril 2014;101:656–63.e1.
VOL. - NO. - / - 2019