Abstract
Context
Ovarian stimulation (OS) increases pregnancy rates but can cause multiple folliculogenesis and multiple pregnancy.
Objective
To determine whether the probability of pregnancy differs in OS cycles with mono- vs multifolliculogenesis in women with unexplained infertility (UI).
Design
Secondary analysis of a multicenter, randomized controlled trial: Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation with 3 treatment arms: gonadotropins, clomiphene, or letrozole, combined with intrauterine insemination. Women were categorized as having either 1 or ≥ 2 mature follicles (≥ 16 mm). Relative risk (RR) and 95% CIs for clinical pregnancy and live birth by number of follicles were estimated using generalized linear models adjusted for age, body mass index, years of infertility, and history of prior live birth.
Setting
12 US-based clinical sites.
Participants
Normally cycling women aged 18 to 40 years with a normal uterine cavity and at least 1 patent fallopian tube. Male partners with ≥ 5 million total motile sperm.
Interventions
Gonadotropins, clomiphene, or letrozole with insemination
Main Outcome Measure(s)
Clinical pregnancy rates (CPR) and live birth rates (LBR).
Results
A single mature follicle > 16 mm resulted in lower CPR (RR, 0.70; 95% CI, 0.54-0.90) and LBR (RR, 0.67; 95% CI, 0.51-0.89) compared with ≥ 2 mature follicles. When stratified by treatment modality, no association of follicle number with CPR or LBR was observed for letrozole or clomiphene, but women using gonadotropins had lower CPR and LBR with monofolliculogenesis.
Conclusion
In couples undergoing gonadotropin treatment for UI, monofolliculogenesis following OS is related to a lower rate of live birth.
Keywords: unexplained infertility, gonadotropin, folliculogenesis, single dominant follicle
Infertility is a disease in which a pregnancy is not achieved after 12 (or more) months of regular, unprotected intercourse [1]. Typically, an infertility evaluation is started after 12 months in women 34 years old or younger but begins after 6 months without success if the female partner is aged 35 or years older. Unfortunately, infertility is common, affecting up to 15% of American women [2].
An infertility investigation involves determining if the female partner is ovulating, at least 1 fallopian tube is patent, and using a semen analysis to investigate if the male partner has an adequate number of motile sperm [3]. If no cause is identified after a thorough evaluation searching for an etiology, approximately 30% of couples with infertility will be diagnosed with unexplained infertility (UI) [3]. Various treatment options for UI have been proposed including intrauterine insemination (IUI), ovulation stimulation (OS) (with either oral or injectable medications), or a combination of OS and IUI and assisted reproductive technologies [3]. Currently, UI is often treated with OS and intrauterine insemination as a first-line therapy [4].
Recently, a large multicenter randomized controlled trial, the Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation (AMIGOS), tested the hypothesis that the rate of multiple gestations in women with UI who had been treated with OS with letrozole, gonadotropin, or clomiphene would be at least as low with letrozole as with the other 2 agents [5]. Compared with women undergoing gonadotropin stimulation, women who were stimulated with letrozole had a significantly lower rate of multiple gestation as well as a lower frequency of live birth. Overall, the live birth rate (LBR) among the 3 groups was 32.2% (gonadotropin), 23.3% (clomiphene), and 18.7% (letrozole) and the multiple gestation rates were 22%, 9%, and 13%, respectively.
OS can result in multiple follicles and a higher pregnancy rate but unfortunately can lead to a higher risk of multiple pregnancies. A hyperresponse to OS is often defined as a cycle in which there are 4 or more follicles measuring > 12 mm [6]. It is widely believed that the more mature the follicles present, the greater the probability of pregnancy. There is concern that OS treatment may be futile if multiple follicles are not created. Indeed, some clinicians do not recommend that patients with UI who have only a single mature follicle undergo human chorionic gonadotropin (hCG) trigger and insemination because the chances of pregnancy may not be better than what they would be in an unstimulated cycle. A meta-analysis of OS cycles indicated that there is no further benefit to more than 1 follicle in terms of overall pregnancy rate, and the accompanying risk of multiple pregnancy increased substantially [7]; however, the same author performed another meta-analysis 2 years later that found multifollicular growth did improve pregnancy rates [8]. A more recent retrospective cohort analyzed more than 50 000 OS cycles in a single private practice and concluded that pregnancy rates were not substantially improved when women have more than 2 mature follicles, but clinical pregnancy was more likely with 2 mature follicles compared with 1 [9]. Given previous inconsistent results, the purpose of this study was to determine if there was a difference in the probability of pregnancy with mono- vs multiple folliculogenesis among women with UI undergoing OS in the AMIGOS study clinical trial, in which a rigorous OS algorithm was followed by all participating sites.
Materials and Methods
Study Design and Population
This is a secondary analysis from the AMIGOS clinical trial. The trial design has been previously published elsewhere [10]. In brief, this multicenter, randomized clinical trial enrolled 900 couples with UI. Female participants were regularly menstruating women aged 18 to 40 years, with a normal uterine cavity and at least 1 patent fallopian tube and a male partner with at least 5 million total motile sperm. Tubal patency was confirmed with either hysterosalpingogram or saline sonogram. This trial was conducted at 12 clinical sites throughout the United States; institutional review board approval was obtained at each site. All participants gave written, informed consent. OS with intrauterine insemination was carried out in a similar manner at all sites, using an identical clinical protocol. Initial monitoring was performed on cycle day 3 (±2 days), at which time follicular number and size was measured. Medications were continued for 5 days for clomiphene and letrozole and monitoring was repeated within 3 days of completion of the course of oral medication. For women randomized to gonadotropins, reevaluation with estradiol and ultrasound measurements was performed after 4 days of treatment. Monitoring visits were individualized thereafter, with the option of gonadotropin dose adjustment for women by 37.5 to 75 IU/d beginning on cycle day 7 (a starting dose for cycle 1 of 150 IU of gonadotropins was used for all participants in that arm). Future cycles could be started at doses ranging from 75 to 225 IU/d, depending on the ovarian response in the previous cycle. Dose reduction or escalation of letrozole and clomiphene was also allowed in subsequent cycles such that letrozole doses ranged from 2.5 to 7.5 mg and clomiphene from 50 to 150 mg. Criteria for hCG administration was identical for all groups upon the first occurrence of: (1) the lead follicle reaching 20 mm average diameter, (2) 2 lead follicles >18 mm average diameter, and (3) the day after the lead follicle reached 18 mm average diameter, or on the day of detection of presumptive ovulation by ultrasound and estradiol monitoring. hCG was withheld if more than 4 follicles >18 mm mean diameter were present or E2 was >3000 pg/mL.
Outcome Assessment
Women in the study were randomized to 1 of 3 treatment groups: letrozole, clomiphene, or gonadotropin. This analysis assessed clinical pregnancy and live birth. A clinical pregnancy was defined as an intrauterine pregnancy with positive cardiac activity seen via transvaginal ultrasound. Live birth was defined as the delivery of at least 1 viable infant.
Follicles were measured at each site using transvaginal ultrasound and recorded on a standard case report form. The frequency of follicle measurements was determined by clinical considerations. All follicles were measured in 2 dimensions and the mean of the 2 dimensions was recorded. The total number of follicles in various size categories was captured (≥20 mm, ≥ 18-< 20 mm, ≥ 16-< 18 mm, ≥ 14-< 16 mm, ≥ 10-< 14 mm, < 10 mm) for each ovary and recorded on the case report form. Endometrial thickness was also measured in millimeters and recorded. All individuals performing ultrasound were credentialed to do so at each site and underwent specific training related to AMIGOS ultrasound protocols.
Statistical Analysis:
Demographic characteristics of women in the AMIGOS trial were compared by the number of follicles ≥ 16 mm on the day of trigger in the first cycle of study participation among the 795 women who completed at least 1 cycle of fertility treatment (hCG trigger administered) and had information on follicle size in the first completed cycle. For the regression models, information from all 900 women enrolled in AMIGOS was used, with multiple imputation used to account for missing data. Log binomial regression models with robust error variance were used to estimate associations between number of follicles ≥ 16 mm on the day of trigger (1 vs ≥ 2 follicles) and clinical pregnancy and live birth. Associations were estimated overall and stratified by treatment cycle (clomiphene, letrozole, gonadotropins). The models take multiple cycles into account and were adjusted for age, body mass index (BMI), number of months attempting conception, and history of previous live birth. Results were compared with models that additionally adjusted for endometrial thickness at the time of trigger, antithyroid antibodies and antichlamydial antibodies, because these factors have previously been shown to be associated with outcomes in this trial [11, 12]. Several metabolic factors have been associated with lower pregnancy rates and long-term health issues in women including homeostasis model assessment for insulin resistance, total cholesterol, and high-sensitivity C-reactive protein; thus, an additional model controlled for these factors [13–15]. A sensitivity analysis evaluating the number of follicles ≥18 mm on the day of trigger was also completed. The mean number of follicles after OS was compared by treatment group using repeated measures ANOVA and adjusting for multiple comparisons. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
Results
Table 1 indicates the baseline characteristics of the study sample. There were no differences in age, education, or household income among women with monofollicular development vs those with multifollicular development. The mean BMI was higher in those with monofollicular development (27.4 ± 6.9) than in those with multifollicular development (26.4 ± 6.1), which was statistically significant (P = 0.03). Most participants in both groups were of self-identified White race.
Table 1.
Demographics and characteristics of women in the AMIGOS trial by number of follicles ≥16 mm in the first completed cycle of fertility treatment (hCG trigger administered)
| Total N = 900 |
N = 876a | 1 follicle N = 325 |
≥ 2 follicles N = 470 |
P | |
|---|---|---|---|---|---|
| Age, y: mean ± SD | 32.2 ± 4.2 | 32.2 ± 4.2 | 32.2 ± 4.3 | 32.3 ± 4.2 | 0.83 |
| BMI, kg/m2: mean ± SD | 26.9 ± 6.5 | 26.9 ± 6.5 | 27.4 ± 6.9 | 26.4 ± 6.1 | 0.03 |
| Months attempting conception: mean ± SD | 34.7 ± 25.7 | 34.6 ± 25.6 | 34.8 ± 24.5 | 34.7 ± 26.7 | 0.95 |
| Anti-TPO antibodies, IU/mL: mean ± SD | 54.8 ± 158.9 | 55.7 ± 160.3 | 58.7 ± 164.7 | 51.9 ± 154.7 | 0.57 |
| Endometrial thickness at baseline, mm: mean ± SD | 6.7 ± 3.2 | 6.7 ± 3.1 | 6.8 ± 3.3 | 6.7 ± 3.1 | 0.71 |
| Endometrial thickness at time of trigger, mm: mean ± SD | 9 ± 2.7 | 9 ± 2.7 | 9.0 ± 2.8 | 9.1 ± 2.6 | 0.80 |
| HOMA-IR: mean ± SD | 2.1 ± 3.6 | 2.1 ± 3.3 | 2.0 ± 2.7 | 2.1 ± 3.7 | 0.50 |
| Total cholesterol: mean ± SD | 169.4 ± 31.3 | 169.7 ± 31.3 | 171.1 ± 31.7 | 168.7 ± 31.1 | 0.29 |
| hs-CRP: mean ± SD | 3.3 ± 4.9 | 3.3 ± 5.1 | 3.5 ± 5.1 | 3.2 ± 5.1 | 0.32 |
| Race, n (%) | |||||
| White | 639 (71) | 631 (72) | 231 (71.1) | 347 (73.8) | 0.47 |
| Black | 82 (9.1) | 73 (8.3) | 23 (7.1) | 42 (8.9) | |
| Asian | 58 (6.4) | 54 (6.2) | 20 (6.2) | 31 (6.6) | |
| American Indian | 7 (0.8) | 7 (0.8) | 2 (0.6) | 3 (0.6) | |
| Mixed race | 89 (9.9) | 86 (9.8) | 39 (12) | 35 (7.4) | |
| Hispanic or Latino | 5 (0.6) | 5 (0.6) | 2 (0.6) | 3 (0.6) | |
| Other | 20 (2.2) | 20 (2.3) | 8 (2.5) | 9 (1.9) | |
| Education, n (%) | |||||
| 8th grade or less | 3 (0.3) | 3 (0.3) | 1 (0.3) | 2 (0.4) | 0.14 |
| Some high school | 15 (1.7) | 15 (1.7) | 6 (1.8) | 7 (1.5) | |
| High school graduate | 55 (6.1) | 54 (6.2) | 16 (4.9) | 32 (6.8) | |
| Some college | 204 (22.7) | 191 (21.8) | 82 (25.2) | 92 (19.6) | |
| College graduate | 385 (42.8) | 380 (43.4) | 146 (44.9) | 196 (41.7) | |
| Graduate degree | 238 (26.4) | 233 (26.6) | 74 (22.8) | 141 (30.0) | |
| Household income, n (%) | |||||
| <$25 000 | 38 (4.2) | 36 (4.1) | 13 (4.0) | 18 (3.8) | 0.07 |
| $25 000-$49 999 | 114 (12.7) | 110 (12.6) | 50 (15.4) | 52 (11.1) | |
| $50 000-$74 999 | 200 (22.2) | 196 (22.4) | 64 (19.7) | 114 (24.3) | |
| $75 000-$100 000 | 199 (22.1) | 196 (22.4) | 86 (26.5) | 94 (20.0) | |
| >$100 000 | 191 (21.2) | 185 (21.1) | 61 (18.8) | 107 (22.8) | |
| Wish to not answer | 158 (17.6) | 153 (17.5) | 51 (15.7) | 85 (18.1) | |
| Antichlamydial antibodies, n (%) | 123 (16.8) | 121 (16.7) | 37 (13.3) | 65 (17.0) | 0.23 |
| Prior live birth, n (%) | 183 (20.3) | 181 (20.7) | 70 (21.5) | 96 (20.4) | 0.72 |
Missing months attempting conception n = 9, anti-TPO antibodies n = 68, endometrial thickness at baseline n = 5, endometrial thickness at time of trigger n = 75, antichlamydial antibodies n = 169, HOMA-IR n = 26, total cholesterol n = 56, hs-CRP n = 28.
Abbreviations: AMIGOS, Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation; BMI, body mass index; hCG, human chorionic gonadotropin; HOMA-IR, homeostatic model assessment for insulin resistance; hs-CRP, high-sensitivity C-reactive protein; TPO, thyroid peroxidase.
876 women completed at least 1 cycle of fertility treatment (hCG trigger administered); 795 had information about follicle size in the first completed cycle.
Women will often have a mature oocyte in follicles measuring at least 16 mm. In the overall cohort, women with a single mature follicle ≥ 16 mm identified at the time of the HCG trigger had a decreased likelihood of clinical pregnancy (relative risk [RR], 0.70; 95% CI, 0.54-0.90) and live birth (RR, 0.67; 95% CI, 0.51-0.89) compared with women with ≥ 2 follicles larger than 16 mm (Table 2). When a cutoff of 18 mm was used, similar results were seen for both clinical pregnancy (RR, 0.77; 95% CI, 0.60-0.98) and live birth (RR, 0.75; 95% CI, 0.58-0.97) (Table 2).
Table 2.
Clinical pregnancy and live birth in women with 1 follicle vs ≥ 2 follicles with unexplained infertility and undergoing ovulation stimulation in the AMIGOS trial
| 1 follicle vs ≥ 2 follicles (≥16 mm) |
1 follicle vs ≥ 2 follicles (≥18 mm) |
||
|---|---|---|---|
| Model | RR (95% CI) | RR (95% CI) | |
| N, cycles | 1 follicle: n (%) | 1031 (38.7) | 1553 (58.3) |
| N, women | >50% cycles with 1 follicle: n (%) | 245 (27.2) | 483 (53.7) |
| Clinical pregnancy | % cycles resulting in clinical pregnancy 1 vs ≥ 2 follicles | 8% vs 12% | 9% vs 12% |
| Unadjusted | 0.68 (0.52-0.87)* | 0.74 (0.58-0.95)* | |
| Adjusted model 1a | 0.69 (0.53-0.89)* | 0.76 (0.60-0.97)* | |
| Adjusted model 2b | 0.70 (0.54-0.90)* | 0.77 (0.60-0.98)* | |
| Adjusted model 3c | 0.70 (0.54-0.90)* | 0.77 (0.61-0.98)* | |
| Live birth | % cycles resulting in live birth 1 vs ≥ 2 follicles | 7% vs 10% | 8% vs 11% |
| Unadjusted | 0.65 (0.49-0.87)* | 0.73 (0.56-0.94)* | |
| Adjusted model 1a | 0.67 (0.50-0.88)* | 0.74 (0.57-0.96)* | |
| Adjusted model 2b | 0.67 (0.51-0.89)* | 0.75 (0.58-0.97)* | |
| Adjusted model 3c | 0.67 (0.51-0.88)* | 0.75 (0.58-0.94)* |
All 900 women enrolled in the study contributed to the analysis. Multiple imputation used to account for missing data. Reference group is ≥2 follicles group. The * is used to denote statistically significant findings.
Abbreviations: AMIGOS, Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation; BMI, body mass index; CRP, C-reactive protein; HOMA-IR, homeostatic model assessment for insulin resistance; RR, relative risk; TPO, .
Adjusted model 1 includes age, BMI, number of months attempting conception, and history of prior live birth.
Adjusted model 2 includes age, BMI, number of months attempting conception, history of prior live birth, anti-TPO antibodies, antichlamydial antibodies, and endometrial thickness at time of trigger.
Adjusted model 3 includes age, BMI, number of months attempting conception, history of prior live birth, anti-TPO antibodies, antichlamydial antibodies, endometrial thickness at time of trigger, HOMA-IR, total cholesterol, and CRP.
After stratifying the data by treatment modality (gonadotropins, letrozole, or clomiphene), we found that women who were treated with gonadotropins having ≥ 2 follicles were associated with higher clinical pregnancy and live birth (Table 3). Results for the clomiphene group also suggested potential positive associations with ≥ 2 follicles, though the estimates were less precise because of smaller numbers (odds ratio for live birth, 0.66 [0.40-1.11] in unadjusted and 0.71 [0.42-1.19] in the fully adjusted model). Women treated with letrozole demonstrated no association with follicle number 1 vs 2 or more and live births (0.97 [0.58-1.63] in the unadjusted and 0.91 [0.54-1.54] in the fully adjusted model). Similar results were observed when a cutoff of 18 mm was used [Table 3]. Treatment modality was related to the mean number of follicles at the time of HCG trigger, when using the 16 mm cutoff: clomiphene (2.03 ± SE 0.04), letrozole (1.72 ± SE 0.03), and gonadotropins (2.21 ± SE 0.05) (Table 4). Of the treatment cycles, the percent that resulted in a live birth or multiple live birth (calculated as number of live births/number of cycles) is shown by follicle size (Table 5). Overall, there is a lower percent of multiple live births in cycles with a single follicle.
Table 3.
Clinical pregnancy and live birth in women with 1 follicle vs ≥ 2 follicles with unexplained infertility and undergoing ovulation stimulation in the AMIGOS trial, stratified by treatment cycle
| 1 follicle vs ≥ 2 follicles (≥16 mm) |
1 follicle vs ≥ 2 follicles (≥18 mm) |
||||||
|---|---|---|---|---|---|---|---|
| Clomiphene | Letrozole | Gonadotropins | Clomiphene | Letrozole | Gonadotropins | ||
| Model | RR (95% CI) | RR (95% CI) | RR (95% CI) | RR (95% CI) | RR (95% CI) | RR (95% CI) | |
| N, cycles | 1 follicle: n (%) | 309 (33.9) | 411 (44.2) | 311 (37.8) | 497 (54.4) | 582 (62.7) | 474 (57.6) |
| N, women | >50% cycles with 1 follicle: n (%) | 66 (22.0) | 102 (34.1) | 77 (25.6) | 144 (48.0) | 180 (60.2) | 159 (52.8) |
| Clinical pregnancy | % cycles resulting in clinical pregnancy 1 vs ≥ 2 follicles | 8% vs 11% | 8% vs 7% | 8% vs 17% | 8% vs 12% | 8% vs 8% | 11% vs 17% |
| Unadjusted | 0.76 (0.48-1.18) | 1.06 (0.66-1.70) | 0.46 (0.29-0.72)* | 0.68 (0.45-1.03) | 1.00 (0.62-1.63) | 0.69 (0.47-1.00) | |
| Adjusted modela | 0.81 (0.51-1.26) | 1.04 (0.65-1.67) | 0.46 (0.29-0.72)* | 0.73 (0.48-1.10) | 1.00 (0.61-1.62) | 0.69 (0.48-1.01) | |
| Adjusted modelb | 0.80 (0.51-1.27) | 1.04 (0.65-1.67) | 0.47 (0.30-0.74)* | 0.73 (0.47-1.11) | 1.01 (0.62-1.65) | 0.70 (0.48-1.02) | |
| Adjusted modelc | 0.84 (0.54-1.30) | 1.00 (0.62-1.61) | 0.48 (0.31-0.73)* | 0.76 (0.50-1.14) | 0.98 (0.60-1.59) | 0.70 (0.49- 1.00)* | |
| Live birth | % cycles resulting in live birth 1 vs ≥ 2 follicles | 6% vs 9% | 6% vs 7% | 8% vs 15% | 7% vs 20% | 6% vs 7% | 10% vs 15% |
| Unadjusted | 0.66 (0.40-1.11) | 0.97 (0.58-1.63) | 0.52 (0.32-0.82)* | 0.67 (0.42-1.06) | 0.89 (0.53,1.49) | 0.71 (0.48-1.06) | |
| Adjusted modela | 0.71 (0.42-1.19) | 0.95 (0.57-1.60) | 0.52 (0.32-0.83)* | 0.72 (0.45-1.14) | 0.88 (0.53,1.47) | 0.71 (0.48-1.06) | |
| Adjusted modelb | 0.71 (0.42-1.19) | 0.91 (0.54-1.54) | 0.53 (0.34-0.85)* | 0.72 (0.45-1.14) | 0.87 (0.52-1.45) | 0.72 (0.49-1.07) | |
| Adjusted modelc | 0.72 (0.42, 1.22) | 0.84 (0.50, 1.42) | 0.54 (0.35, 0.84)* | 0.73 (0.46, 1.15) | 0.82 (0.49, 1.36) | 0.72 (0.50, 1.06) | |
Reference group is ≥2 follicles group. The * is used to denote statistically significant findings.
Abbreviations: AMIGOS, Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation; BMI, body mass index; CRP, C-reactive protein; HOMA-IR, homeostatic model assessment for insulin resistance; RR, relative risk; TPO, .
Adjusted model 1 includes age, BMI, number of months attempting conception, and history of prior live birth.
Adjusted model 2 includes age, BMI, number of months attempting conception, history of prior live birth, anti-TPO antibodies, antichlamydial antibodies, and endometrial thickness at time of trigger.
Adjusted model 3 includes age, BMI, number of months attempting conception, history of prior live birth, anti-TPO antibodies, antichlamydial antibodies, endometrial thickness at time of trigger, HOMA-IR, total cholesterol, and CRP.
Table 4.
Mean number of follicles after ovarian stimulation for unexplained infertility among women in the Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation trial by treatment groups (gonadotropin, clomiphene, or letrozole)
| Number of follicles (≥16 mm) Mean ± SE |
Number of follicles (≥18 mm) Mean ± SE |
|
|---|---|---|
| Clomiphene | 2.03 ± 0.04 | 1.56 ± 0.02 |
| Letrozole | 1.72 ± 0.03 | 1.42 ± 0.02 |
| Gonadotropins | 2.21 ± 0.05 | 1.54 ± 0.03 |
Table 5.
Percentage of treatment cycles resulting in live birth or multiple live birth (n live births/n cycles) by follicle size
| No. follicles ≥ 16 mm | 1 | 2 | 3 | 4 | 5+ |
|---|---|---|---|---|---|
| N (cycles) | 917 | 948 | 351 | 113 | 78 |
| Live birth, n (%) | 62 (6.8) | 85 (9) | 38 (11) | 13 (12) | 15 (19) |
| Multiple live birth, n (%) | 5 (0.5) | 16 (1.7) | 10 (2.8) | 2 (1.8) | 7 (8.9) |
Note: Women may contribute multiple cycles.
Discussion
In this secondary analysis of a large, randomized controlled trial, women with monofollicular development had a significantly lower probability of pregnancy and live birth when compared with those with multifollicular development. This held true when evaluating the data using both 16-mm and 18-mm follicle size cutoffs. When stratified by treatment modality, this association appeared largely driven by gonadotropin treatment. No relationship between clinical pregnancy or live birth was observed with 1 compared with 2 or more follicles in women with unexplained infertility treated with letrozole.
It is widely believed that more mature follicles achieved in women with unexplained infertility increases the pregnancy rate [16, 17]. Prior studies of this question have provided mixed results, with some confirming [18] and others refuting this concept [7, 8, 19, 20]. A 2008 meta-analysis that evaluated 14 studies reporting on more than 11 000 cycles concluded that follicle number influenced pregnancy rate. Compared with monofollicular growth, pregnancy rates increased by 5%, 8%, and 8% when 2, 3, or 4 follicles were seen; the absolute pregnancy rate was 8.4% for monofollicular growth vs 15% for multifollicular growth [8]. These authors also noted that in cycles with 3 or 4 follicles, the multiple pregnancy rate was substantially increased but the overall pregnancy rate was not significantly impacted [8]. The studies within the meta-analysis had significant heterogeneity and varied primary outcome measures with only 7 reporting on live birth; 13 of the studies were retrospective cohorts and only 1 was a prospective study. Conversely, the current study reports on live birth and the data were collected prospectively. Patients in those studies received gonadotropins and/or clomiphene citrate for ovulation induction but none used letrozole.
A recent large retrospective cohort study also was conducted to address this question [9]. The authors determined that singleton pregnancy rates were not significantly improved when more than 2 mature follicles were present, but the rate of multiple gestations rose steadily with more follicles of 14 mm or more. This report analyzed more than 50 000 OS cycles, used rigorous statistical methodology, and defined a mature follicle as all follicles measuring 14 mm or more. However, it was a retrospective cohort study, and the primary outcome measure was clinical pregnancy rate—not live births—and multiple gestations. Although this study was from 1 large clinical practice, there may have been more variation around the clinical protocols than we had in the Reproductive Medicine Network study. The study population consisted of patients with various causes of infertility, with the majority diagnosed with unexplained infertility (39.5%). Additionally, few cycles used letrozole alone (6.1%) and a subanalysis based on treatment type was not performed.
The majority of studies that have attempted to address questions of optimal follicle size and number were retrospective cohorts [7, 8, 18, 19, 21] and there was heterogeneity in the inclusion criteria, infertility diagnosis, treatments used, and follicle size cutoffs used [8]. A key feature of our analysis is that results were stratified by treatment group. Although the number of mature follicles was similar across the 3 treatment groups, gonadotropins produced the highest mean number of follicles (at both the 16- and 18-mm cutoffs) and letrozole produced the lowest.
Van Rumste et al noted an absolute rate of multiple pregnancies of 0.3% in cases of monofollicular development and 2.8% rate when multiple follicles develop [8]. In 2017, a retrospective study evaluated the risk of multiple pregnancy in women undergoing OS with insemination secondary to UI. These authors found that history of prior pregnancy and 3 or more follicles ≥ 15 mm was associated with a higher risk of multiple pregnancy but did not comment on the overall pregnancy rate [21]. Another study examined 861 gonadotropin-IUI cycles; of the 95 pregnancies that resulted in a live birth, 11.5% (n = 11) were characterized as high-order multiples (ie, triplets or greater). In that series, neither the size nor the number of follicles predicted multiple pregnancies [18].
The current study has multiple strengths. It uses information from a randomized controlled trial that ensured that all sites used identical protocols and standardized scanning procedures and had a low dropout rate. Further, all couples had been diagnosed with UI. The prospective nature of the parent study avoids many of the pitfalls inherent with retrospective studies. Moreover, prior studies reported clinical pregnancy rates and not live birth rates; both are reported in the current study. The parent study is a randomized controlled trial that was meticulously executed, using standardized treatment protocols and scanning techniques with excellent follow-up and excellent capture of exposures and outcomes. Because the data were collected prospectively, accurate assessments of clinical pregnancy rate were readily available. The large number of patients examined in 3 separate treatment arms also allowed subgroup analyses, which might not have been possible in other circumstances. Rigorous statistical methods, including multiple imputation, were undertaken to account for the impact of missing data. The results of the study are consistent across many different analyses and the CIs are not wide. There are some limitations to this study as well. Because all couples in this study had been diagnosed with UI, the results are not generalizable to patients with other causes of infertility. Although all female partners in the study had confirmed ovulation, subclinical ovulatory disturbances were not excluded as a cause for infertility. We relied on local clinical protocols at each site for decision making about when to perform ultrasound, and local equipment was used. However, a standard procedure and systematic scanning were maintained.
In summary, we have found that, in a randomized controlled trial population of women with well-characterized unexplained infertility and undergoing OS and intrauterine insemination, both pregnancy and live birth rates tend to be lower in women with a single dominant follicle among women receiving gonadotropins. Taken in conjunction with the existing data, we conclude that multifollicular development increases live birth rates but also increases the risk of high-order multiples.
Acknowledgments
We thank Dr. Esther Eisenberg for her expert input in the writing of this manuscript and the Reproductive Medicine Network for its input and guidance.
Abbreviations
- AMIGOS
Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation
- BMI
body mass index
- hCG
human chorionic gonadotropin
- IUI
intrauterine insemination
- LBR
live birth rate
- OS
ovulation stimulation
- RR
relative risk
- UI
unexplained infertility
Contributor Information
Torie C Plowden, Email: torie.c.plowden.mil@mail.mil, Department of Gynecologic Surgery and Obstetrics, Womack Army Medical Center, Fort Bragg, NC 28310, USA.
Sunni L Mumford, Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH, Bethesda, MD 20892, USA.
Robert A Wild, Department of Obstetrics and Gynecology, OUHSC, Oklahoma City, OK 73117, USA.
Marcelle I Cedars, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, San Francisco, CA 94158, CA, USA.
Anne Z Steiner, Duke University Medical Center, Durham, NC 27710, USA.
Jason M Franasiak, RMA New Jersey, Thomas Jefferson University, Basking Ridge, NJ 07920, USA.
Michael P Diamond, Augusta University Medical Center, Augusta, GA 30912, USA.
Nanette Santoro, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Aurora, CO 80045, USA.
Funding
Supported by grants from the National Institutes of Health, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R25 HD-075737, to N.S., U10 HD39005, to M.P.D.). S.L.M. was 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.
Disclosures
The views expressed herein are those of the author(s) and do not necessarily reflect the official policy of the Department of the Army, Department of Defense, or the US Government. The authors have no other disclosures.
The Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation (AMIGOS) was registered in ClinicalTrials.gov as trial number NCT01044862.
Data Availability
Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
References
- 1. Practice Committee of the American Society for Reproductive Medicine . Definitions of infertility and recurrent pregnancy loss: a committee opinion. Fertil Steril. 2020;113(3):533–535. [DOI] [PubMed] [Google Scholar]
- 2. Thoma ME, McLain AC, Louis JF, et al. Prevalence of infertility in the United States as estimated by the current duration approach and a traditional constructed approach. Fertil Steril. 2013;99(5):1324–1331.e1321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Practice Committee of the American Society for Reproductive Medicine . Effectiveness and treatment for unexplained infertility. Fertil Steril. 2006;86(5 Suppl 1):S111–S114. [DOI] [PubMed] [Google Scholar]
- 4. Practice Committee of the American Society for Reproductive Medicine . Evidence-based treatments for couples with unexplained infertility: a guideline. Fertil Steril. 2020;113(2):305–322. [DOI] [PubMed] [Google Scholar]
- 5. Diamond MP, Legro RS, Coutifaris C, et al. Letrozole, gonadotropin, or clomiphene for unexplained infertility. N Engl J Med. 2015;373(13):1230–1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Balayla J, Granger L, St-Michel P, et al. Rescue in vitro fertilization using a GnRH antagonist in hyper-responders from gonadotropin intrauterine insemination (IUI) cycles. J Assist Reprod Genet. 2013;30(6):773–778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. van Rumste MM, den Hartog JE, Dumoulin JC, Evers JL, Land JA. Is controlled ovarian stimulation in intrauterine insemination an acceptable therapy in couples with unexplained non-conception in the perspective of multiple pregnancies? Hum Reprod. 2006;21(3):701–704. [DOI] [PubMed] [Google Scholar]
- 8. van Rumste MM, Custers IM, van der Veen F, van Wely M, Evers JL, Mol BW. The influence of the number of follicles on pregnancy rates in intrauterine insemination with ovarian stimulation: a meta-analysis. Hum Reprod Update. 2008;14(6):563–570. [DOI] [PubMed] [Google Scholar]
- 9. Evans MB, Richter KS, Connell MT, et al. Mature follicle count and multiple gestation risk based on patient age in ovulation induction-intrauterine insemination (OI-IUI) cycles. Fertil Steril. 2018;110(4):e30. [Google Scholar]
- 10. Diamond MP, Mitwally M, Casper R, et al. Estimating rates of multiple gestation pregnancies: sample size calculation from the assessment of multiple intrauterine gestations from ovarian stimulation (AMIGOS) trial. Contemp Clin Trials. 2011;32(6):902–908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Seungdamrong A, Steiner AZ, Gracia CR, et al. Preconceptional antithyroid peroxidase antibodies, but not thyroid-stimulating hormone, are associated with decreased live birth rates in infertile women. Fertil Steril. 2017;108:843–850. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Steiner AZ, Diamond MP, Legro RS, et al. Chlamydia trachomatis immunoglobulin G3 seropositivity is a predictor of reproductive outcomes in infertile women with patent fallopian tubes. Fertil Steril. 2015;104(6):1522–1526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Song H, Yu Z, Li P, Wang Y, Shi Y. HOMA-IR for predicting clinical pregnancy rate during IVF. Gynecol Endocrinol. 2022;38(1):33–38. [DOI] [PubMed] [Google Scholar]
- 14. Tinius RA, Yoho K, Blankenship MM, Maples JM. Postpartum metabolism: how does it change from pregnancy and what are the potential implications? Int J Womens Health. 2021;13:591–599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Sjaarda LA, Radin RG, Silver RM, et al. Preconception low-dose aspirin restores diminished pregnancy and live birth rates in women with low-grade inflammation: a secondary analysis of a randomized trial. J Clin Endocrinol Metab. 2017;102(5):1495–1504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Guzick DS, Sullivan MW, Adamson GD, et al. Efficacy of treatment for unexplained infertility. Fertil Steril. 1998;70(2):207–213. [DOI] [PubMed] [Google Scholar]
- 17. Cohlen BJ, te Velde ER, van Kooij RJ, Looman CW, Habbema JD. Controlled ovarian hyperstimulation and intrauterine insemination for treating male subfertility: a controlled study. Hum Reprod. 1998;13(6):1553–1558. [DOI] [PubMed] [Google Scholar]
- 18. Fong SA, Palta V, Oh C, Cho MM, Loughlin JS, McGovern PG. Multiple pregnancy after gonadotropin-intrauterine insemination: an unavoidable event? ISRN Obstet Gynecol. 2011;2011:465483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Moosavifar N, Aliakbarzadeh M, Shakeri M. Association of the number of follicles of various sizes and treatment protocol with multiple pregnancies following ovulation induction and intrauterine insemination. J Pak Med Assoc. 2008;58(1):18–22. [PubMed] [Google Scholar]
- 20. Dickey RP, Taylor SN, Lu PY, Sartor BM, Rye PH, Pyrzak R. Relationship of follicle numbers and estradiol levels to multiple implantation in 3,608 intrauterine insemination cycles. Fertil Steril. 2001;75(1):69–78. [DOI] [PubMed] [Google Scholar]
- 21. Ghesquiere SL, Castelain EG, Spiessens C, Meuleman CL, D’Hooghe TM. Relationship between follicle number and (multiple) live birth rate after controlled ovarian hyperstimulation and intrauterine insemination. Am J Obstet Gynecol. 2007;197(6):589.e1–589.e5. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.