Abstract
Purpose:
The objective was to evaluate whether the mid-luteal progesterone (MLP) assessment in same-sex female couples benefits clinical outcomes in natural cycles using donor sperm intrauterine insemination (IUI).
Methods:
This retrospective cohort study included same-sex female couples undergoing donor sperm IUI from January 2004 to April 2022. Cases included patients with MLP evaluation; controls included those without. MLP was obtained ∼7 days after ovulation. Only natural cycles with human chorionic gonadotropin trigger for timing were included. Primary outcome was clinical pregnancy rate; secondary outcomes were ongoing pregnancy (OP) and spontaneous abortion rates.
Results:
The study included 912 cycles, in 56 of which MLP was assessed. There were no demographic differences between groups. Among those assessed, the mean MLP was 10.96 ng/mL and 19/56 (33.9%) subsequently received supplemental progesterone. In an unadjusted analysis, there were no significant differences in clinical outcomes between groups. After adjusting for age, body mass index, day 3 follicle-stimulating hormone, and endometrial thickness at the time of ovulation, having MLP evaluated did not predict clinical (odds ratio [OR]: 3.34, confidence interval [CI]: 0.194–57.510, p = 0.406) or OP rate (OR 5.056, CI 0.24–106.62, p = 0.298). A subanalysis found no difference in clinical and OP rates when comparing patients who had received supplemental progesterone versus those who had not.
Conclusion:
Clinicians may reconsider the MLP assessment within same-sex female couples who use donor sperm IUI, as it does not appear to enhance treatment outcome. Prospective studies may delineate the cost–benefit analysis of the MLP assessment in this cohort.
Introduction
Heterosexual couples typically present for fertility treatment due to “medical infertility,” which is defined as the inability to conceive after 1 year or longer of unprotected sex. 1 Same-sex female couples can experience “medical infertility,” yet most seek fertility care out of the need to access donor sperm, which is commonly referred to as “social infertility.” 2 One could hypothesize that same-sex female couples would have similar or even higher pregnancy rates compared with heterosexual couples seeking infertility treatment. 3 There are limited data on specific considerations for same-sex female couples using assisted reproductive technologies.4,5 As a result, there is ongoing debate about how to tailor clinical protocols for this population (i.e., whether to do natural intrauterine insemination [IUI] timed to luteinizing hormone surge or use oral medications or injectable gonadotropins for ovulation induction [OI]). 4 Providers often use the same clinical assessments within same-sex female couples without diagnoses of medical infertility that are routinely used in heterosexual couples experiencing medical infertility. As a result, same-sex female couples are frequently subjected to costly, time-consuming, or invasive testing and procedures that are not evidence-based within the socially infertile patient population. For instance, the mid-luteal progesterone (MLP) assessment is a serum evaluation of a patient’s progesterone levels seven days after ovulation to confirm ovulation has occurred. 2 Although the MLP assessment may be useful for indicating ovulation in medically infertile heterosexual couples, it is not widely understood whether the assessment is clinically valuable within same-sex female couples. 6
Adequate luteal phase production of progesterone is needed to induce endometrial changes required for successful implantation of an embryo. 7 An MLP concentration of 4 ng/mL or greater typically indicates ovulation; however, the clinical value of assessing MLP levels has been widely disputed.6–8 The American Society for Reproductive Medicine does not have guidelines regarding the value of the MLP assessment but has noted that luteal phase deficiency—a clinical diagnosis potentially caused by inadequate progesterone duration or levels that is associated with an abnormal luteal phase length ≤10 days—is not proven to independently cause infertility or recurrent pregnancy loss. 9
A number of studies have presented mixed results about the relationship between MLP concentrations and pregnancy outcomes in patients undergoing OI, OI followed by IUI (OI-IUI), IUI without exogenous hormones, and in vitro fertilization (IVF).8,10–18 While some studies have demonstrated a significant relationship between MLP levels, pregnancy, and live birth rates, others have found that MLP levels do not enhance treatment outcomes.8,10–18
Research evaluating the clinical significance of MLP has thus far focused on infertile heterosexual couples; however, there is a paucity of data evaluating the utility of MLP in same-sex female couples. 19 Therefore, our objective was to determine whether there is an association between MLP levels and clinical outcomes in same-sex female couples undergoing donor sperm IUI (dsIUI).
Materials and Methods
This single-center, retrospective study included all same-sex female couples from 25 to 44 years of age who underwent human chorionic gonadotropin (hCG)-triggered natural cycle dsIUI at an academic, private fertility practice between January 2004, when the center began using an electronic medical record, and April 2022. Same-sex female couples who pursued dsIUI were identified in an electronic medical record database and included in the study. Cases included all patients who had an evaluation of MLP. Controls included all patients without MLP testing. Only hCG-triggered natural cycles were included; medicated cycles for OI or super ovulation were excluded. This study was approved by Mount Sinai’s Institutional Review Board with a waiver of patient consent (STUDY-18-00441).
Patient selection
Women who self-identified as “lesbian” or as a “same-sex couple” were identified through natural language processing of electronic medical records and were included in the study. Women who had a known male partner in the database, who identified as “single,” a “single mother by choice,” or those who indicated a desire for “single parenting” were excluded from the analysis. In cases where a patient’s familial status (e.g., “single” or “in a same-sex couple”) was not recorded anywhere in the electronic record, the patient was excluded from analysis.
Total group protocol
Cycles followed one of two of the following protocols. In the first protocol, patients undergoing natural cycle preparation for IUI were monitored for the presence of a dominant follicle. On day 12, patients presented for a follicle evaluation with transvaginal ultrasound. Once a dominant follicle (≥18 mm) was observed, ovulation was triggered with 250 µg of recombinant hCG (Ovidrel; EMD Serono, Inc., Rockland, MA). Endometrial thickness (EnT) was also recorded at this cycle time point. dsIUI was performed 36 hours after hCG administration. If preferred, patients were assigned to a second protocol, in which patients utilized daily ovulation predictor kits (OPK). When the OPK yielded a positive result indicating ovulation, patients presented to the clinic that same day for dsIUI. Given these protocols, MLP was defined as a progesterone level obtained seven days after either the ovulation trigger or first positive OPK.
Of note, select patients received supplemental progesterone, either in oral or in vaginal formulation, as luteal phase support. For the patients who received supplemental progesterone, the median MLP was 7.04 ± 4.61 ng/mL (range: 2.5–24.6 ng/mL). Those who did not receive supplemental progesterone had a higher median MLP (13.10 ± 4.60 ng/mL) with a range of 0.16–25.1 ng/mL. On average, patients with an MLP of <7 ng/mL received supplemental progesterone, but patient clinical characteristics (e.g., past successful pregnancy with use of progesterone supplementation) also contributed to decision making.
Intrauterine insemination
As referenced in Nazem et al., previously cryopreserved sperm samples were thawed in a 37°C incubator for 15 minutes on the morning of the scheduled dsIUI. 20 Samples were then thoroughly homogenized with a large volume pipette. The volume of the sample and spermatozoa count were recorded. Sperm wash (Irvine Scientific, Santa Ana, CA) was then added to twice the volume of the sample, and the sample was mixed. The sample was centrifuged at 300 times gravity, ∼1500 revolutions per minute for 10 minutes. The supernatant was removed, and the pellet was resuspended in 0.3 mL of sperm wash media and mixed.
Outcome measures
Baseline patient characteristics and demographic data were obtained, including age, body mass index (BMI), anti-Müllerian hormone (AMH) level, day 3 follicle-stimulating hormone (D3FSH) level, gravidity, parity, and EnT at time of ovulation.
The primary study outcome was clinical pregnancy (CP) rate, which was determined by the presence of a gestational sac visualized on transvaginal ultrasound ∼7–10 days following a positive serum β-hCG. Secondary outcomes included ongoing pregnancy (OP) and spontaneous abortion rate. An OP was defined as a viable intrauterine gestation at the time of discharge from the practice, which was no earlier than 8 weeks of gestation. A spontaneous abortion was considered a loss after a visualized intrauterine gestational sac on transvaginal ultrasound.
Statistical analyses
Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC). Continuous data were reported as mean ± standard deviation or median (interquartile range) as appropriate. Comparative statistics were performed using chi-square tests for categorical data and either the Student’s t-test or Mann–Whitney U test for continuous data. To assess differences in clinical outcomes and adjust for potential confounders, a multivariable logistic regression was performed for each outcome. Likelihood of clinical outcomes was presented as odds ratios (OR) with 95% confidence intervals (CIs). In addition, a subanalysis was performed evaluating the impact of progesterone supplementation on the aforementioned outcomes. All p-values were two sided with a clinical significance level determined at p < 0.05.
Results
A total of 912 hCG-triggered natural dsIUI cycles completed by 364 unique same-sex female couples were included in the study. If a couple underwent >1 cycle at the center, all cycles that fit inclusion criteria were included. In 56 cycles, which included 24 distinct patients, the MLP assessment was performed. The median MLP level was 7.04 ± 4.71 ng/mL. The decision to assess MLP was based on individual provider or patient preference (e.g., in cases where the patient had had an MLP assessment in a prior cycle where they conceived). Progesterone was supplemented in 19 of 56 cycles involving 14 unique patients. In the remaining 37 cycles including 31 unique patients, the median MLP was 10 ± 4.60 ng/mL, and progesterone was not supplemented (p < 0.001). Progesterone was not supplemented in any of the cycles where MLP was not assessed.
Demographic and cycle characteristics of the patients are included in Table 1. There were no statistically significant differences between patients who underwent MLP assessment and those who did not with respect to age, BMI, AMH, D3FSH, EnT at time of ovulation, and gravidity (Table 1). The only significant difference between the two groups was parity: patients who underwent an MLP evaluation had significantly higher parity compared to those who did not (p = 0.0067), though the majority of patients in both groups had a parity of zero (85.71% in the MLP group vs. 91.25% in the no MLP group). The mean MLP in those assessed was 10.96 ng/mL.
Demographic and Cycle Characteristics of Same-Sex Female Couples Undergoing Human Chorionic Gonadotropin-Triggered Natural Cycles Using Donor Sperm Intrauterine Insemination
Bolded p-value denotes statistical significance (p < 0.05).
AMH, anti-Müllerian hormone; BMI, body mass index; D3FSH, day 3 follicle-stimulating hormone; EnT, endometrial thickness; IQR, interquartile range; MLP, mid-luteal progesterone; SD, standard deviation.
In an unadjusted analysis, there were no significant differences in clinical outcomes, including CP rate (p = 0.72), OP rate (p = 0.72), and spontaneous abortion rate (p = 0.19), between the two groups (Table 2). After adjusting for age, BMI, D3FSH, and EnT at time of ovulation with logistic regression, the presence of the MLP evaluation was not associated with CP rate (OR 3.34, CI 0.19–57.51, p = 0.41) or OP rate (OR 5.06, CI 0.24–106.62, p = 0.30) (Table 3).
Pregnancy Outcomes of Same-Sex Couples Undergoing Human Chorionic Gonadotropin-Triggered Natural Cycles Using Donor Sperm Intrauterine Insemination (Unadjusted)
Denominator is the number of patients who achieved biochemical pregnancy.
Denominator is the number of patients who achieved clinical pregnancy.
MLP, mid-luteal progesterone.
Pregnancy Outcomes of Same-Sex Couples (Adjusted)
CI, confidence interval; OR, odds ratio.
A subanalysis was performed to assess whether the supplementation of progesterone influenced clinical outcomes (Table 4). No significant differences in CP and OP were found when comparing patients who had received supplemental progesterone versus those who had not.
Supplemental Luteal Progesterone in Same-Sex Couples Undergoing Human Chorionic Gonadotropin-Triggered Natural Cycles Using Donor Sperm Intrauterine Insemination
Suppl. Prog.: supplemental progesterone.
Discussion
This study evaluated whether the assessment of MLP levels in same-sex female couples influences pregnancy-related outcomes in hCG triggered natural cycles using dsIUI. In both adjusted and unadjusted analyses, there were no significant differences in clinical outcomes between patients who had versus those who did not have MLP evaluation. Having MLP evaluated did not appear to be associated with the odds of implantation, OP, or spontaneous abortion. However, given the small number of cycles (n = 56) in which the MLP assessment was performed, the study lacked power to determine a definitive difference in outcomes between the two groups. Nevertheless, this study suggests no compelling justification for performing the MLP assessment in a patient without a specific indication and without medical infertility. Hence, our results suggest foregoing MLP testing except when specifically indicated or until evidence is generated to support its use in same-sex female couples.
Studies have presented mixed findings regarding the relationship between MLP levels and pregnancy-related outcomes in OI, IUI, and IVF.8,10–18 Only two of these studies investigated the clinical value of the MLP assessment in patients undergoing natural IUI cycles specifically. Fukuda et al. found that lower MLP levels were associated with a higher pregnancy rate in the subsequent IUI treatment cycle, especially when ovulation occurred in the same ovary for two consecutive cycles. 21 Those findings suggest that mid-luteal hormone profiles may impact whether conception will occur in the following menstrual cycle. Takaya et al. demonstrated a positive association between high MLP concentrations and cycles that achieved a pregnancy in patients utilizing IUI without human menopausal gonadotropin stimulation or timed intercourse. 22
With regard to luteal phase progesterone supplementation, existing research has focused on OI-IUI rather than hCG-triggered natural cycles using dsIUI.23–34 A 2017 meta-analysis found that luteal phase support improved CP and live birth rates in patients undergoing OI using injectable gonadotropins but not with either clomiphene citrate alone or clomiphene citrate plus injectable gonadotropins. 23 In our subanalysis, this study found no significant differences in clinical outcomes between those who received versus did not receive supplemental progesterone. However, the subanalysis included 56 cycles (in 19 of which progesterone supplementation was performed) and did not reach power to detect a difference between groups.
Even though the use of assisted reproductive technology (ART) is significantly increasing among same-sex female couples, there is a paucity of data evaluating the utility of tests and interventions specifically in this patient population. Data of this nature would be germane to generating evidence-based, personalized infertility treatment protocols for same-sex female couples. This is the first study to investigate the clinical significance of the MLP assessment in same-sex female couples undergoing hCG-triggered natural cycle dsIUI. 23 The study findings suggest that clinicians should reconsider the MLP evaluation within same-sex female couples who utilize dsIUI. Our findings showed MLP assessment does not appear to enhance treatment outcomes in same-sex female couples.
The evaluation of MLP could represent an example of over-medicalization in the same-sex female couple patient population. Medicalization is the process by which human conditions and problems become defined and treated as medical conditions. Medicalization can not only be stigmatizing but it can also subject patients to unnecessary, costly, time-consuming, and invasive testing and procedures. An Australian study found that clinical protocols that do not differ between women with and without male partners were seen by clinicians as responsible practice. 2 However, providers should be cognizant of patients such as single women and women in same-sex relationships who are seeking fertility treatment due to social infertility. Socially infertile patients may, but often do not, have concomitant medical infertility and therefore likely require significantly less testing and intervention. Fertility treatment for socially infertile women should be personalized and distinct from treatment for women with a medical infertility diagnosis. Although the MLP assessment is not a particularly costly intervention, it is time consuming without providing clinical value in same-sex female couples undergoing dsIUI. Practitioners might consider reexamining the tests and interventions that they utilize in this patient population rather than adopting a “one-size-fits-all” approach to patients seeking assisted reproduction. Personalized treatment for all patients will allow providers to offer the best possible care while avoiding unnecessary treatment.
Study strengths and limitations
This study had several strengths. Our study was performed at a single, high-volume academic medical center. This reduces the inherent variability around protocols and management that may arise from multicenter studies. Study participants received relatively uniform management due to the streamlined protocols at the single study center. In addition, this study was conducted in New York State, where same-sex marriage was legalized in 2011. 35 Subsequently, in 2021, New York State began requiring that private insurers cover services for the diagnosis and treatment of infertility for those unable to conceive due to their sexual orientation or gender identity. 36 Therefore, this study was conducted in an environment that is amenable and safe for same-sex female couples to grow their families.
The study is not without its limitations. For one, it was calculated that in order to have 80% power, we would need 199 patients per group to detect a 10% difference in pregnancy rate. As we did not have a sufficient number of patients with MLP, this study was underpowered. Therefore, we were unable to draw a decisive conclusion regarding the utility of the MLP assessment in same-sex female couples. A study with a larger sample size of patients who received the MLP assessment may be able to generate this evidence; however, there is no compelling argument for the clinical utility of performing the MLP assessment in a patient without medical infertility or a specific indication for the test. Another limitation was the study’s retrospective design. To overcome this limitation, an adjusted multivariable logistic regression analysis was performed to minimize selection bias.
In addition, a third of the patients who underwent the MLP assessment and were found to have an MLP level of <7 ng/mL on average received supplemental progesterone. The MLP levels of the patients who were supplemented ranged from 2.5 to 24.6 ng/mL, illustrating that patient characteristics (e.g., past successful pregnancy with progesterone supplementation) as well as provider judgment impact clinical decision making. To account for the fact that the information obtained from the MLP assessment informed subsequent clinical decision making, we performed a subanalysis to compare patients who had received supplemental progesterone with those who had not. The subanalysis found no differences in terms of clinical outcomes between the two groups. It should also be noted that data regarding the time interval between the prior delivery date and date of presentation to the fertility center were not available.
Future directions
Large, prospective, multicenter studies could further delineate the risk–benefit analysis of MLP assessment in same-sex female couples undergoing natural cycle dsIUI. The generation of more robust datasets will help providers deliver evidence-based, personalized treatment to same-sex couples to optimize clinical outcomes without administering unnecessary tests or treatments.
Conclusions
The experience of same-sex female couples differs from that of heterosexual couples undergoing fertility treatment. Many same-sex female couples suffer from social rather than medical infertility, making their sole barrier to conception access to viable sperm. This is the first study to investigate the clinical value of the MLP assessment in same-sex female couples undergoing hCG-triggered natural cycles with dsIUI. Our results show that MLP assessment does not appear to be associated with differences or improvements in CP outcomes in this population. Clinicians should reconsider the evaluation of the MLP within same-sex female couples who use dsIUI, as there is no compelling justification for its use in patients without medical infertility, it may also be an example of over-medicalization of social infertility. The MLP assessment is just one example of an intervention that may be used in same-sex female couples without benefit. Future research should continue to evaluate specific considerations of testing and interventions in LGBT patients such that protocols can be patient centered and evidence based.
It has been suggested that medicalization adds to healthcare costs without improving treatment. 37 While the medicalization of infertility and the advent of ART have allowed for the creation of alternative family structures (i.e., with single or same-sex parents), treatment protocols should be tailored to same-sex female couples such that they are not unnecessarily subjected to clinically invaluable, costly, or time-consuming tests and interventions.
Footnotes
Authors’ Contributions
I.C.B.: Writing—original draft preparation (lead), project administration (lead), visualization (equal), formal analysis (equal), and writing—review and editing (equal). S.L.E.: Formal analysis (equal), investigation (equal), and writing—review and editing (equal). J.A.L.: Conceptualization, methodology, resources (lead), and writing—review and editing (equal). M.B.: Data curation (lead), formal analysis (equal), investigation (equal), and visualization (equal). N.C.: Writing—original draft preparation. D.S.: Conceptualization, methodology, and writing—review and editing (equal). T.M.: Conceptualization, methodology, and writing—review and editing (equal). A.B.C.: Conceptualization, methodology, writing—review and editing (equal), and supervision. J.F.: Conceptualization (lead), methodology (equal), writing—review and editing, and supervision (lead). All authors agreed to the submission of this article to the Journal of Women’s Health.
Data Availability
Data regarding any of the subjects in the study have not been previously published unless specified. Data will be made available to the editors of the journal for review or query upon request.
Author Disclosure Statement
A.B.C. is currently a company officer and direct stockholder for Progyny. D.S. is the Medical Director of WINFertility. None of the remaining authors have personal, professional, or financial interest in any of the products, devices, or drugs mentioned in this article.
Funding Information
The authors did not receive support from any organization for the submitted work.
