Low-dose aspirin in reproductive health: effects on menstrual cycle characteristics

Objective To estimate the effect of daily 81 mg low-dose aspirin (LDA) on menstrual cycle length and hormone profiles. Design Secondary analysis of a trial evaluating the effect of daily LDA or placebo on live birth among women with one or two previous pregnancy losses. Setting University medical centers. Patient(s) A total of 915 regularly menstruating women who had at least one menstrual cycle (3,190 total cycles) in which pregnancy did not occur. Intervention(s) Randomized allocation to LDA versus placebo. Main Outcome Measure(s) Menstrual cycle length and follicular and luteal phases were measured. Urinary pregnanediol glucuronide, follicle-stimulating hormone, luteinizing hormone, and estrone-3-glucuronide were assessed up to six times during the first two cycles. Generalized estimating equations estimated relative risk of short (<25th percentile: 75th percentile: ≥ days) versus normal cycle length. Linear mixed models estimated mean hormone level differences with weights used to account for multiple cycles contributed per participant. Result(s) There were no significant differences in total menstrual cycle, follicular phase, or luteal phase length between LDA and placebo groups. LDA posed no greater risk of having a short versus normal-length or long versus normal-length follicular phase, or having a short versus normal-length or long versus normal-length luteal phase. There were no significant differences in hormone levels across the menstrual cycle between the LDA and placebo groups. Conclusion(s) Daily LDA use did not result in any changes to menstrual cycle, follicular phase, or luteal phase length or hormone levels across the menstrual cycle compared with placebo. Clinical Trial Registration Number

Low dose aspirin, menstrual cycle length, menstrual cycle hormones Low-dose aspirin (LDA) is increasingly used in women’s health for the prevention of a wide range of diseases. One study suggested that more than one-half of middle aged people in the United States take daily LDA (1). Potential benefits of daily LDA include decreased risks of stroke and cardiovascular disease, colon cancer, and obstetrical complications such as preeclampsia, fetal growth restriction, pregnancy loss, and preterm birth (2–11). The mechanism of action for many of these benefits is thought to be via decreased blood viscosity and increased blood flow, secondary to inhibition of cyclooxygenase-1 and decreased production of thromboxane-2 (12, 13). LDA may have further benefits in human reproduction by improving uterine and ovarian blood flow, enhancing embryo implantation and placental invasion, and sustaining early pregnancy (5–8, 14–18). Indeed, we previously observed in the Effects of Aspirin in Gestation and Reproduction (EAGeR) trial that preconception LDA in women with a single recent pregnancy loss was associated with higher rate of live birth compared with placebo (5, 8). While LDA use is increasing, one study suggested that nearly one-half of women surveyed feel uninformed when making decisions about factors that may influence their reproductive health (19). As LDA use is more commonly recommended and used by women of reproductive age for a wide variety of benefits, it is of interest to consider the potential effects of LDA on the menstrual cycle. For example, if a women is taking LDA and experiences irregular bleeding patterns, understanding whether or not this is an expected side-effect may be of interest to the patient. In addition, potential changes in menstrual cycle characteristics may elicit concern that could be addressed by her provider before taking LDA. LDA may have menstrual bleeding effects via direct vascular actions in the endometrium and myometrium, but its use may also affect the hormonal profile of the menstrual cycle. It has been hypothesized that LDA could alter hormonal profiles by increasing ovarian blood flow and the delivery of steroid hormone precursors (20). Indeed, some studies have shown that aspirin use is associated with both higher (20) and lower (21) progesterone levels, although aspirin use was not randomized and its use was self-reported in those studies. LDA could also potentially alter follicular and luteal phase lengths in addition to overall menstrual cycle length owing to its effects on prostaglandins, which are involved in steroidogenesis, ovulation, and luteolysis (21). However, current evidence examining aspirin use and the menstrual cycle is limited to only a few small studies (21–23) that examined much higher doses of aspirin (650–3,000 mg daily), relied on nonrandomized study designs, and were based on self-reported aspirin use. Therefore, our objective was to evaluate the effect of LDA on menstrual cycle length and hormonal profiles in a secondary analysis of a large randomized controlled trial. We hypothesized that daily administration of LDA would have no overall effect on these evaluated objectives.

The EAGeR trial was a block-randomized, double-blind, placebo-controlled trial conducted from 2006 to 2012 at four medical centers across the United States. As previously described (9), women aged 18–40 years who were attempting pregnancy and had one of two documented previous pregnancy losses and no more than two previous live births were enrolled. Exclusion criteria included: history of irregular menstrual cycles, self-reported menstrual cycle length 42 days, history of infertility, major medical problems (including polycystic ovary syndrome), or medical contraindication to aspirin. A total of 1,228 participants were randomized to LDA or placebo on days 2–4 of their first menstrual cycle of follow-up and were followed for six menstrual cycles and throughout pregnancy for those who became pregnant. Participants completed demographic and medical history questionnaires, provided blood and urine samples, and had anthropometric measures taken at enrollment. In addition, participants provided daily first morning urine samples during the first two cycles of follow-up for measurement of hormone levels. Participants were instructed in the use of a fertility monitor (Clearblue Easy Fertility Monitor; Inverness Medical Innovations) to plan timing of study visits to specific days of their menstrual cycle. The protocol was approved by the institutional review board at each participating clinical center, and all participants provided written informed consent prior to data collection. The trial was registered at clinicaltrials.gov: identifier NCT00467363. Trial Design Participants were randomized to either 81 mg LDA plus 400 mg folic acid or placebo plus 400 μg folic acid in a 1:1 ratio by means of a computerized algorithm based on a permuted-block design consisting of six or eight random blocks in random order. Participants took either LDA or placebo daily while attempting to conceive for up to 6 menstrual cycles and up to 36 weeks of gestation if they became pregnant. Participants, study personnel, and investigators were blinded to treatment assignment throughout the trial, and the aspirin and placebo tablets were manufactured to be identical in appearance and weight. Folic acid was provided as a separate supplement. Adherence was both self-reported and assessed through study medication bottle weights. Menstrual Cycle Length and Reproductive Hormone Levels The beginning of menses and ovulation date were recorded through self-report and fertility monitor data. The date of ovulation was determined to occur the day after the peak luteinizing hormone (LH) reading, as determined by an internal proprietary algorithm of the fertility monitor. Menstrual cycle length was calculated as the number of days from the beginning of menses to the beginning of menses in the next cycle. Follicular phase length was calculated as the number of days from menses start to ovulation, and luteal phase as the number of days from ovulation to the beginning of menses in the next cycle. Cycle length was defined as short (<25th percentile: 75th percentile: >31 days). Short, normal, and long follicular lengths were similarly defined by quartiles as 17 days, respectively. And short, normal, and long luteal lengths were defined as 14 days, respectively. Urinary reproductive hormones were measured up to six times per cycle during the first two cycles of follow-up and included pregnanediol glucuronide (PdG), follicle-stimulating hormone (FSH), LH, and estrone-3-glucuronide (E1G). Samples were timed to the phases of the menstrual cycle: during menses (approximately cycle day 2), on the day of ovulation (1 day after LH peak), peri-implantation (9 days after LH peak), and a random luteal phase sample (range 4 days after LH peak to 2 days before date of next menses). Urinary E1G and PdG were measured by means of competitive chemiluminescence duplex assay, with limits of detection of 0.40 ng/mL for E1G and 45 ng/mL for PdG (Quansys Biosciences). Using an in-house control, interassay coefficients of variation (CVs) were 16.9% at 36.3 ng/mL and 20.2% at 1.9 ng/mL for E1G and 23.2% at 40,60 ng/mL and 20.2% at 1,604 ng/mL for PdG. Urinary LH and FSH were measured by means of reagent/sandwich immunoassay (Roche Diagnostics) with interassay CVs of 1.6% and 1.8%, respectively. Participant Charictaristics Participant demographics, including age (years), race (white, nonwhite), marital status (married, living with partner, other), education (up to high school, beyond high school), annual household income (%$19,999, $20,000–$39,999, $40,000-$74,999, $75,000–$99,999, R$100,000), and current smoking (yes, no), were collected through self-report at enrollment. Participants also self-reported reproductive history, including parity (nulliparous, parous), ever use of contraceptives (yes, no), time since last use of hormonal contraceptives (months), age at menarche (years), and history of menstrual function, including usual menstrual bleeding (days), number of periods in last 6 months, and whether periods were regular (yes, no). Body mass index (BMI) was calculated from weight and height (kg/m2) measured by trained study personnel at enrollment. Statistical Analysis Of the 1,228 women randomized, 243 became pregnant during the first cycle and 70 were missing information on cycle length. A total of 915 women contributed 3,190 cycles for this analysis (Fig. 1). Of these, 141 women had missing information on the date of ovulation, so analyses of follicular and luteal phase lengths were based on 774 women contributing 1,986 cycles. Participant characteristics were summarized as counts and percentages or means and standard deviations. We compared the distribution of participant characteristics between those assigned to LDA and those assigned to placebo by means of Pearson chi-square test or Student t test. Using intention-to-treat analyses, the association of LDA with short versus normal and long versus normal menstrual cycle length was modeled using generalized estimating equations with log-binomial regression. Generalized estimating equations accounted for dependence among cycles in the same women, weighting by total number of cycles per woman to adjust for fecundability differences among women contributing multiple cycles. We log-transformed menstrual cycle hormone levels to approximate normality. We evaluated the relationship of LDA with mean difference in log(PdG), log(LH), log(FSH), and log(E1G) across the menstrual cycle and within the follicular and luteal phases with the use of linear mixed models. Eighteen percent of the cycles had missing data regarding ovulation status such that we were unable to determine lengths of follicular and luteal phases. Reasons for missing data included infrequent or lack of proper use of the fertility monitor (10% of cycles), testing malfunction (1% of cycles), or missing tests around the anticipated LH surge day (7% of cycles). Missing data were accounted for with multiple imputation using chained equations to create ten datasets (24, 25). We conducted a thorough sensitivity analysis evaluating whether excluding participants with missing data (missing cycle length or unobserved cycle lengths for cycles in which a pregnancy occurred) may have affected our estimates for total cycle length (n = 265 cycles in the placebo group and n = 277 cycles in the LDA group) by imputing cycle lengths from various scenarios. Specifically, missing cycle length was imputed by means of two approaches (10 iterations each): randomly assigned cycle length categories to maintain observed prevalences of cycle lengths in the population; and multiple imputation approach with the use of information on total cycle length as well as age, BMI, race/ethnicity, number of previous pregnancy losses, smoking, parity, and time from last pregnancy loss to randomization. All analyses were conducted in SAS 9.4.

Demonstrating successful randomization of the treatment intervention, participant characteristics did not differ between the LDA and placebo groups, including in age (29.1 vs. 28.9 y; P=.55), BMI (26.4 vs. 26.6 kg/m2; P=.73), race (white vs. nonwhite; P=.11), and menstrual regularity (P=.72; Table 1). Women in the LDA group were more likely to self-report at baseline longer menstrual bleeding before the study (5.2 days vs. 4.9 days; P=.01), although the history of other cycle characteristics (including past use of hormonal contraception, age at menarche, and years before periods become regular) were similar by treatment assignment. In our prospective data, during the first menstrual cycle of follow-up, median cycle length was 30 (interquartile range [IQR] 27–32) days, while median follicular phase length was 16 (14–18) days and median luteal phase length was 13 (12–14) days. TABLE 1. | Characteristic | Low-dose aspirin (n = 452) | Placebo (n = 463) | P value | |---|---|---|---| | Age, y | 29.1 ± 5.0 | 28.9 ± 4.7 | .55 | | BMI, kg/m2 | 26.4 ± 6.9 | 26.6 ± 6.5 | .73 | | Race | .11 | || | White | 421 (93.1) | 443 (95.7) | | | Nonwhite | 31 (6.9) | 20 (4.3) | | | Marital status | .27 | || | Married | 421 (93.1) | 419 (90.5) | | | Living with partner | 25 (5.5) | 32 (6.9) | | | Other | 6 (1.3) | 12 (2.6) | | | Education | .50 | || | Beyond high school | 385 (85.2) | 403 (87.0) | | | Up to high school | 66 (14.6) | 60 (13.0) | | | Missing data | 1 (0.2) | 0 (0) | | | Annual household income, US$ | .12 | || | ≥100,000 | 180 (39.9) | 193 (41.7) | | | 75,000–$99,999 | 64 (14.2) | 43 (9.3) | | | 40,000–$74,999 | 67 (14.9) | 73 (15.8) | | | 20,000–$39,999 | 104 (23.1) | 125 (27.0) | | | ≤19,999 | 36 (8.0) | 29 (6.3) | | | Current smoker | .42 | || | Yes | 60 (13.4) | 53 (11.5) | | | No | 388 (86.6) | 406 (88.5) | | | Parity | .55 | || | Nulliparous | 219 (48.5) | 215 (46.4) | | | Parous (1 or 2 prior live births) | 233 (51.5) | 248 (53.6) | | | Ever used hormonal contraception/meds, any reason | .73 | || | Yes | 349 (81.2) | 362 (80.1) | | | No | 81 (18.8) | 90 (19.9) | | | Age at menarche, y | 12.7 ± 1.5 | 12.7 ± 1.5 | .51 | | Years for periods to become regulara | 1.8 ± 4.8 | 1.5 ± 4.0 | .29 | | Usual menstrual bleeding, d | 5.2 ± 1.9 | 4.9 ± 1.4 | .01 | | No. of periods in past 6 mo | 4.6 ± 1.6 | 4.7 ± 1.6 | .30 | | Are your periods regular? | .72 | || | Yes | 367 (83.8) | 369 (82.0) | | | No | 52 (11.9) | 57 (12.7) | | | Don’t know | 19 (4.3) | 24 (5.3) | Note: Values presented as mean ± standard deviation or n (%), unless stated otherwise. BMI = body mass index; LDA = low-dose aspirin. Among women who report regular periods. Evans. Effect of aspirin on the menstrual cycle. Fertil Steril 2020. No significant differences were found in total menstrual cycle, follicular phase, or luteal phase length between the LDA and placebo groups (Table 2). For example, women assigned to LDA had no greater risk of having a short follicular phase length (relative risk [RR] 1.02, 95% confidence interval [CI] 0.76–1.39) or long follicular phase length (RR 1.01, 95% CI 0.70–1.47) versus normal-length follicular phase. TABLE 2. | Cycle length | RR (95% CI) | |---|---| | Total cycle length | | | Short (31 d) | 0.97 (0.75–1.26) | | Follicular phase length, d | | | Short (17) | 1.01 (0.70–1.47) | | Luteal phase length, d | | | Short (14) | 1.03 (0.77–1.36) | Note: Weighted for number of menstrual cycles contributed by participant. Short and long cycle lengths represent 75th percentiles, respectively. CI = confidence interval; Ref. = reference; RR = relative risk. Evans. Effect of aspirin on the menstrual cycle. Fertil Steril 2020. Average urinary hormone concentrations were analyzed throughout the menstrual cycle and as well as characterized by follicular and luteal phases (Table 3). When averaged over the entire menstrual cycle, women in the LDA group had a 2%–5% increase in urinary hormones compared with the placebo group, although estimates were imprecise and nonsignificant. For example, women assigned to LDA had only a 5% greater urinary PdG level than the placebo group (95% CI 4.0% to 15.0%). When subcategorized into follicular and luteal phases, there were no differences in hormonal levels between the LDA and placebo groups (Table 3). TABLE 3. | Low-dose aspirin | Placebo | ||||| |---|---|---|---|---|---|---| | Cycle | Hormone | Mean (SD) | Range | Mean (SD) | Range | Percentage difference (95% CI) | | Total cycle | PdG (μg/mL) | 13.14 (18.99) | 0.29–185.24 | 12.95 (18.28) | 0.09–240.7 | 5.0 (−4.0 to 15.0) | | FSH (mIU/mL) | 3.67 (5.07) | 0.27–51.3 | 3.8 (6.09) | 0.34–75.62 | 2.0 (−8.0 to 12.0) | | | LH (mIU/mL) | 2.16 (4.9) | 0.01–59.26 | 2.35 (5.85) | 0.01, 74.8 | 6.0 (−8.0 to 19.0) | | | E1G (ng/mL) | 51.65 (51.19) | 0.21–510.6 | 54.27 (56.13) | 0.06–547.5 | 2.0 (−7.0 to 12.0) | | | Follicular phase | PdG (μg/mL) | 3.93 (4.56) | 0.29–32.71 | 4.87 (6.88) | 0.12–52.04 | 5.0 (−10.0 to 20.0) | | FSH (mIU/mL) | 4.03 (4.09) | 0.32–24.28 | 3.96 (4.86) | 0.37–52.85 | −1.0 (−17.0 to 14.0) | | | LH (mIU/mL) | 1.2 (1.39) | 0.07–6.45 | 1.18 (1.45) | 0.07–8.28 | 5.0 (−12.0 to 23.0) | | | E1G (ng/mL) | 30.42 (28.14) | 0.44–191.36 | 32.88 (30.59) | 0.06–228.85 | −3.0 (−17.0 to 11.0) | | | Luteal phase | PdG (μg/mL) | 20.32 (22.55) | 1.1–185.24 | 20.27 (22.93) | 0.09–240.7 | 2.0 (−10.0 to 14.0) | | FSH (mIU/mL) | 1.81 (1.91) | 0.27–15.38 | 2.03 (2.19) | 0.34–15.4 | −5.0 (−15.0 to 6.0) | | | LH (mIU/mL) | 0.73 (1.05) | 0.01–9.88 | 0.94 (1.54) | 0.01–12.29 | −5.0 (−21.0 to 10.0) | | | E1G (ng/mL) | 41.64 (33.4) | 0.21–215.1 | 43.89 (37.23) | 0.48–350.25 | −2.0 (−14.0 to 10.0) | Note: Weighted for number of menstrual cycles contributed by participant. CI = confidence interval; E1G = estrone-3-glucuronide; FSH = follicle-stimulating hormone; LH = luteinizing hormone; PdG = pregnanediol glucuronide; SD = standard deviation. Evans. Effect of aspirin on the menstrual cycle. Fertil Steril 2020. In a sensitivity analysis evaluating the impact of excluding cycles with missing information on total cycle length, we imputed missing total cycle length based on two missing data scenarios and observed similar findings to the overall results suggesting that missing data was not driving the conclusions (Supplemental Table 1, available online at www.fertstert.org). In scenario 1 (imputing with the observed fixed prevalence), the RRs ranged from 1.01 (95% CI 0.79–1.28) to 1.04 (95% CI 0.82–1.32) and from 0.98 (95% CI 0.78–1.24) to 1.02 (95% CI 0.81–1.29) for effect estimates for LDA versus placebo on short versus normal and long versus normal total cycle lengths, respectively. Similarly, no meaningful differences in estimates compared with our main analyses were observed in scenario 2 (imputing based on covariate information), where effect estimates for LDA versus placebo on short versus normal total cycle length ranged from RR 0.99 (95% CI 0.79–1.24) to 1.09 (95% CI 0.88–1.37) and from RR 0.90 (95% CI 0.73–1.12) to 1.07 (95% CI 0.86–1.32) for long versus normal total cycle lengths, respectively.

In this randomized controlled trial of women attempting pregnancy, daily use of LDA had no effects on menstrual cycle patterns or reproductive hormone concentrations. Although other doses of aspirin have been evaluated (22, 23, 26), to our knowledge there are few other studies that evaluate the effect of LDA on menstrual cycle characteristics. Aspirin has been increasingly recommended for a variety of preventive indications, but the potential effects of LDA on menstrual cycle characteristics have not been well studied. These data are reassuring for women of reproductive age using LDA, because all women studied in the EAGeR trial were attempting to become pregnant. Previous data vary and are limited regarding the dose and duration of aspirin use, as well as the menstrual characteristics studied. Our results are consistent with one study evaluating serum hormone concentrations that also found no association between over-the-counter aspirin taken for one or two menstrual cycles and E2, luteal progesterone, FSH, and LH (22, 27). However, our findings regarding menstrual cycle length differ from several other studies, likely owing to differences in aspirin dose and duration of use. Specifically, high doses (3,000 mg daily for 20 days starting cycle day 5) have been associated with a shortening of the menstrual cycle and luteal phase by 2 or more days, in conjunction with lower peak urine PdG and LH (23). Additional studies reveal that over-the-counter analgesics such as ibuprofen and acetaminophen, used in the follicular phase, may reduce the risk of anovulation and therefore shorten total menstrual cycle length (5). In addition, a study of 650 mg aspirin taken 4 times daily during the first 3 days for two cycles did not alter the days of menstrual flow compared with placebo, although other menstrual cycle parameters were not assessed (22). Overall, our work expands upon previous studies by evaluating the role of daily LDA on not only menstrual cycle length, but also reproductive hormone concentrations throughout the menstrual cycle. The present study used 81 mg aspirin for daily use for a longer duration of time than other studies, which best reflects the increasing practice of preconception LDA use by women. Another strength of the present study is the large number of participants (n = 915) and cycles (>3,000), multiple measures of hormone concentrations per cycle, multiple cycles per participant, and prospective assessment of cycle length, as well as the randomized placebo-controlled design. Previous studies have studied much higher doses of aspirin in a nonrandomized fashion, which may increase the risk of biases. Adherence to randomization of the study participants was 94%, and 82% of mensutral cycles had recorded data with the use of the fertility monitors (8). A limitation of the study is the inclusion of only normally menstruating women; therefore we could not assess different effects among women with a history of abnormal cycle duration. Study participants also had a history of a pregnancy loss, and it is unknown if LDA would have an effect in different populations, although 10% of clinically recognized pregnancies result in pregnancy loss (28–31). Another limitation of the study is the use of urinary instead of serum hormones. Using urinary instead of serum samples was mainly due to practical reasons, because participants stored daily urine samples in home freezers and it was not feasible to schedule multiple well timed serum samples. Slight discrepancies between serum and urinary hormones in relation to ovulation have been reported (20, 21, 32, 33), and some studies have reported poor reliability in detecting peak progesterone levels when taken in tandem with serum LH levels (20). However, the overall similar associations between serum and urinary progesterone would still allow the participants to be analyzed and ranked in a similar fashion, even with discrepancies in their specific concentrations. In addition, although menstrual cycle length was not its prespecified study aim, the EAGeR trial provided a unique opportunity to estimate the impact of randomization to LDA on menstrual cycle length, finding little evidence of a clinically meaningful difference in cycle length. We also presented our results with corresponding confidence intervals to allow for an assessment of the imprecision around our point estimate, and we conducted several sensitivity analyses to assess the robustness of our findings. Finally, although compliance in EAGeR was high for the LDA and placebo groups (15% of the the LDA group and 13% of the placebo group stopped taking the study drug, and 6% of the LDA group and 8% of the placebo group temporarily stopped taking the study drug) (8), it is possible that noncompliance may have biased our findings toward the null. However, the modest effect sizes observed suggest that this is unlikely to have greatly influenced our findings. Although per-protocol effects are also of interest, they are outside of the scope of this paper owing to the complex methods that are used (34).

LDA is increasingly prescribed for various indications in women attempting pregnancy. The present data demonstrate that LDA has no adverse effects on menstrual cycle length and hormone patterns in women trying to conceive. The physiologic effects of aspirin seen in previous studies are likely dose related, because the low doses of aspirin used in our study based on a randomized placebo-controlled trial did not affect menstrual cycle characteristics. Women desiring pregnancy can be reassured that LDA is unlikely to adversely affect their menstrual cycles. Supplementary Material Acknowledgments The authors thank all of the EAGeR participants for their commitment to the study; the EAGeR investigators and staff; and the members of the data safety monitoring board. Supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland (contract nos. HHSN267200603423, HHSN267200603424, and HHSN267200603426). Please see http://grants.nih.gov/grants/policy/data_sharing/ for National Institutes of Health data sharing policy. Footnotes M.B.E. has nothing to disclose. C.J.N. has nothing to disclose. K.K. has nothing to disclose. M.J.H. reports personal fees from Ohana Biosciences as member of advisory board, outside of the submitted work. A.H.D. has nothing to disclose. R.M.S. has nothing to disclose. S.L.M. has nothing to disclose. L.A.S. has nothing to disclose. N.J.P. has nothing to disclose. E.F.S. has nothing to disclose.

- 1.Williams CD, Chan AT, Elman MR, Kristensen AH, Miser WF, Pignone MP, et al. Aspirin use among adults in the U.S.: results of a national survey. Am J Prev Med 2015;48:501–8. [DOI] [PubMed] [Google Scholar] - 2.American College of Obstetricians and Gynecologists. ACOG committee opinion no. 743: Low-dose aspirin use during pregnancy. Obstet Gynecol 2018;132:e44–52. [DOI] [PubMed] [Google Scholar] - 3.Henderson JT, Whitlock EP, O’Connor E, Senger CA, Thompson JH, Rowland MG. Low-dose aspirin for prevention of morbidity and mortality from preeclampsia: a systematic evidence review for the U.S. Preventive Services Task Force. Ann Intern Med 2014;160:695–703. [DOI] [PubMed] [Google Scholar] - 4.Mumford SL, Silver RM, Sjaarda LA, Wactawski-Wende J, Townsend JM, Lynch AM, et al. Expanded findings from a randomized controlled trial of preconception low-dose aspirin and pregnancy loss. Hum Reprod 2016; 31:657–65. [DOI] [PMC free article] [PubMed] [Google Scholar] - 5.Radin RG, Sjaarda LA, Perkins NJ, Silver RM, Chen Z, Lesher LL, et al. Low-dose aspirin and sporadic anovulation in the EAGeR randomized trial. J Clin Endocrinol Metab 2017;102:86–92. [DOI] [PMC free article] [PubMed] [Google Scholar] - 6.Renaud SJ, Cotechini T, Quirt JS, Macdonald-Goodfellow SK, Othman M, Graham CH. Spontaneous pregnancy loss mediated by abnormal maternal inflammation in rats is linked to deficient uteroplacental perfusion. J Immunol 2011;186:1799–808. [DOI] [PubMed] [Google Scholar] - 7.Rubinstein M, Marazzi A, Polak de Fried E. Low-dose aspirin treatment improves ovarian responsiveness, uterine and ovarian blood flow velocity, implantation, and pregnancy rates in patients undergoing in vitro fertilization: a prospective, randomized, double-blind placebo-controlled assay. Fertil Steril 1999;71:825–9. [DOI] [PubMed] [Google Scholar] - 8.Schisterman EF, Silver RM, Lesher LL, Faraggi D, Wactawski-Wende J, Townsend JM, et al. Preconception low-dose aspirin and pregnancy outcomes: results from the EAGeR randomised trial. Lancet 2014;384:29–36. [DOI] [PMC free article] [PubMed] [Google Scholar] - 9.Schisterman EF, Silver RM, Perkins NJ, Mumford SL, Whitcomb BW, Stanford JB, et al. A randomised trial to evaluate the effects of low-dose aspirin in gestation and reproduction: design and baseline characteristics. Paediatr Perinat Epidemiol 2013;27:598–609. [DOI] [PMC free article] [PubMed] [Google Scholar] - 10.Thaler I, Manor D, Itskovitz J, Rottem S, Levit N, Timor-Tritsch I, et al. Changes in uterine blood flow during human pregnancy. Am J Obstet Gynecol 1990;162:121–5. [DOI] [PubMed] [Google Scholar] - 11.Zhu BP, Rolfs RT, Nangle BE, Horan JM. Effect of the interval between pregnancies on perinatal outcomes. N Engl J Med 1999;340:589–94. [DOI] [PubMed] [Google Scholar] - 12.Clarke RJ, Mayo G, Price P, FitzGerald GA. Suppression of thromboxane A2 but not of systemic prostacyclin by controlled-release aspirin. N Engl J Med 1991;325:1137–41. [DOI] [PubMed] [Google Scholar] - 13.Patrono C. Aspirin as an antiplatelet drug. N Engl J Med 1994;330:1287–94. [DOI] [PubMed] [Google Scholar] - 14.Cacciatore B, Simberg N, Fusaro P, Tiitinen A. Transvaginal Doppler study of uterine artery blood flow in in vitro fertilization–embryo transfer cycles. Fertil Steril 1996;66:130–4. [DOI] [PubMed] [Google Scholar] - 15.Haapsamo M, Martikainen H, Rasanen J. Low-dose aspirin and uterine haemodynamics on the day of embryo transfer in women undergoing IVF/ICSI: a randomized, placebo-controlled, double-blind study. Hum Reprod 2009;24: 861–6. [DOI] [PubMed] [Google Scholar] - 16.Isaksson R, Tiitinen A, Cacciatore B. Uterine artery impedance to blood flow on the day of embryo transfer does not predict obstetric outcome. Ultrasound Obstet Gynecol 2000;15:527–30. [DOI] [PubMed] [Google Scholar] - 17.Zaidi J, Campbell S, Pittrof R, Tan SL. Endometrial thickness, morphology, vascular penetration and velocimetry in predicting implantation in an in vitro fertilization program. Ultrasound Obstet Gynecol 1995;6:191–8. [DOI] [PubMed] [Google Scholar] - 18.Kuo HC, Hsu CC, Wang ST, Huang KE. Aspirin improves uterine blood flow in the peri-implantation period. J Formos Med Assoc 1997;96:253–7. [PubMed] [Google Scholar] - 19.Gervis Z. Nearly half of young women feel uninformed when making reproductive health decisions New York Post, October 2, 2019. Available at: https://nypost.com/2019/10/02/nearly-half-of-young-women-feel-uninformed-when-making-reproductive-health-decisions/2019. [Google Scholar] - 20.Merlo AB, Farinati Z, Quiroga S, Nagle CA, Carril LM, Lahoz M, et al. Direct assay of urinary steroid glucuronides for monitoring the approach of ovulation. Int J Fertil 1984;29:189–93. [PubMed] [Google Scholar] - 21.Sauer MV, Paulson RJ. Utility and predictive value of a rapid measurement of urinary pregnanediol glucuronide by enzyme immunoassay in an infertility practice. Fertil Steril 1991;56:823–6. [DOI] [PubMed] [Google Scholar] - 22.Pendergrass PB, Scott JN, Ream LJ, Agna MA. Effect of small doses of aspirin and acetaminophen on total menstrual loss and pain of cramps and headache. Gynecol Obstet Invest 1985;19:32–7. [DOI] [PubMed] [Google Scholar] - 23.Souka AR, Medhat M,Rahman HA, Osman M, El Sokkary H. Effect of aspirinon the luteal phase of human menstrual cycle. Contraception 1984;29:181–8. [DOI] [PubMed] [Google Scholar] - 24.Schafer J, Reid N, Cox D, Keiding N, Louis T, Tong H, Isham V. Analysis of incomplete multivariate data. New York: Chapman and Hall/CRC; 1997. [Google Scholar] - 25.van Buuren S. Multiple imputation of discrete and continuous data by fully conditional specification. Stat Methods Med Res 2007;16:219–42. [DOI] [PubMed] [Google Scholar] - 26.Pendergrass PB, Ream LJ, Scott JN, Agna MA. Do aspirin and acetaminophen affect total menstrual loss? Gynecol Obstet Invest 1984;18:129–33. [DOI] [PubMed] [Google Scholar] - 27.Matyas RA, Mumford SL, Schliep KC, Ahrens KA, Sjaarda LA, Perkins NJ, et al. Effects ofover-the-counter analgesicuse onreproductivehormonesand ovulation in healthy, premenopausal women. Hum Reprod 2015;30:1714–23. [DOI] [PMC free article] [PubMed] [Google Scholar] - 28.Wilcox AJ, Weinberg CR, O’Connor JF, Baird DD, Schlatterer JP, Canfield RE, et al. Incidence of early loss of pregnancy. N Engl J Med 1988;319:189–94. [DOI] [PubMed] [Google Scholar] - 29.Wang X, Chen C, Wang L, Chen D, Guang W, French J. Conception, early pregnancy loss, and time to clinical pregnancy: a population-based prospective study. Fertil Steril 2003;79:577–84. [DOI] [PubMed] [Google Scholar] - 30.Zinaman MJ, Clegg ED, Brown CC, O’Connor J, Selevan SG. Estimates of human fertility and pregnancy loss. Fertil Steril 1996;65:503–9. [PubMed] [Google Scholar] - 31.American College of Obstetricians and Gynecologists. ACOG practice bulletin no. 200: Early pregnancy loss. Obstet Gynecol 2018;132:e197–207. [DOI] [PubMed] [Google Scholar] - 32.Stanczyk FZ, Miyakawa I, Goebelsmann U. Direct radioimmunoassay of urinary estrogen and pregnanediol glucuronides during the menstrual cycle. Am J Obstet Gynecol 1980;137:443–50. [DOI] [PubMed] [Google Scholar] - 33.Maskarinec G, Beckford F, Morimoto Y, Franke AA, Stanczyk FZ. Association of estrogen measurements in serum and urine of premenopausal women. Biomark Med 2015;9:417–24. [DOI] [PMC free article] [PubMed] [Google Scholar] - 34.Hernan MA, Robins JM. Per-protocol analyses of pragmatic trials. N Engl J Med 2017;377:1391–8. [DOI] [PubMed] [Google Scholar] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article.