Abstract
Purpose
To assess assisted reproductive technology (ART) outcomes in patients with one ovary compared to two ovaries.
Methods
We performed a retrospective cohort study of all patients with one ovary who underwent ≥ 1 ART cycle between 2012 and 2020 at a large university-affiliated fertility center. Patients were 3-to-1 matched with two ovary controls during the same period. Primary outcome was metaphase II oocytes (MIIs) retrieved per cycle. Secondary outcomes included ovarian reserve markers, laboratory outcomes, and live birth rates (LBRs).
Results
A total of 104 one ovary patients (158 cycles; median age 35.5 years) were matched to 312 two ovary patients (474 cycles; median age 35.0 years). In one ovary patients, anti-Mullerian hormone was lower (median 1.1 vs. 2.2, p < 0.01) and day 2 follicle-stimulating hormone was higher (median 7.4 vs. 6.2, p < 0.01). One ovary patients yielded median 7.5 MIIs and 10 oocytes per cycle, fewer than two ovary patients (11.0 and 14.5, respectively; p < 0.01). However, one ovary patients had ≥ 50% the MII and oocyte yield of two ovary patients (Z > 5.8, p < 0.01). Fertilization and blastocyst formation rates, euploidy rate, and rate of ≥ 1 embryo for transfer were equivalent between groups (p > 0.40). Among the one and two ovary groups, LBRs per transfer (45.8% vs. 46.6%, p = 1.00) and per patient who underwent transfer (68.3% vs. 73.9%, p = 0.55) were equivalent.
Conclusion
One ovary patients yielded fewer MIIs and oocytes than two ovary patients, but had ≥ 50% the yield of two ovary patients, suggesting a compensatory mechanism in oocyte yield in the solitary ovary. One and two ovary patients had equivalent LBRs.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10815-022-02534-9.
Keywords: Compensatory follicular recruitment, Single ovary, Unilateral oophorectomy, Oocyte yield, In vitro fertilization
Introduction
People with a single ovary are a small, but important subgroup of those undergoing fertility evaluation and treatment. These patients are often concerned that having a single ovary will result in poor assisted reproductive technology (ART) outcomes. The prevalence of one ovary in reproductive-aged women who desire future childbearing is unknown, but studies have reported that the prevalence of one ovary in women with other causes of infertility ranges between 5.5 and 17% [1, 2].
Since the advent of ART, studies have attempted to clarify the impact of a single ovary on ART outcomes; however, existing data is conflicting and inconclusive. Initial studies on ART outcomes in women with one ovary suggested that these women were at no disadvantage compared to women with two ovaries, with regard to both total number of oocytes retrieved [3–5] and pregnancy rate [3, 4]. The authors of these studies speculated that there is a compensatory increase in mature follicle development in the single, hyperstimulated ovary, and ventured that the response of a single, hyperstimulated ovary is comparable to that of two ovaries in a counterpart woman. However, subsequent studies concluded that women with one ovary have an inferior response to ovarian stimulation, with fewer oocytes retrieved [6–10]. Others have speculated compensatory measures in oocyte quality, rather than oocyte quantity, in patients with one ovary. This hypothesis is based on studies demonstrating higher fertilization rates [6] and pregnancy rates [11] in patients with one ovary. Studies from the last decade continue to report conflicting data, with no general consensus on oocyte retrieval outcomes, laboratory outcomes, or live birth outcomes [11–13].
Importantly, the majority of studies examining the impact of a single ovary on ART outcomes are outdated and do not reflect modern-day ART. For example, a thorough literature review failed to identify a single study that reviewed preimplantation genetic testing for aneuploidy (PGT-A) or oocyte cryopreservation outcomes in women with one ovary. Prior studies have limited applicability to current ART patients with a single ovary as the utilization of both of these techniques (PGT-A and oocyte cryopreservation) has exponentially increased in the past decade.
Finally, physicians and patients often rely on hormonal markers of ovarian reserve, such as follicle-stimulating hormone (FSH) and anti-Mullerian hormone (AMH), to predict success with fertility treatments. AMH is a well-established predictor of ovarian response to hyperstimulation, with a recognized linear relationship between AMH and oocyte yield in populations undergoing ART; further, AMH is known to decrease with age [14, 15]. However, the impact of having a single ovary on these hormonal markers is not well established. Existing data suggests that patients with one ovary may have a higher basal FSH level than patients with two ovaries [16–18], but the degree of this discrepancy is not well defined. Furthermore, we only identified two studies evaluating the impact of having one ovary on AMH. One small study reported lower AMH in women with one ovary compared to women with two ovaries [18], and another reported that, after unilateral oophorectomy, AMH increases with increasing age (opposite to the decreasing trend typically seen in women with two ovaries) [19].
Given the paucity of data, it is unclear whether patients with one ovary have a lower chance of success with ART than patients with two ovaries. This uncertainty makes it difficult to counsel these patients and help them achieve their reproductive goals. Therefore, our objective was to assess ART outcomes in patients with one ovary compared to patients with two ovaries (matched by age, number of cycles, ART type, and time frame).
Materials and methods
Design
We conducted a retrospective cohort study of all patients with one ovary who underwent ART between January 1, 2012, and October 19, 2020, at the New York University Langone Fertility Center. The study was performed with New York University Institutional Review Board approval (s13-00,389).
Subjects
All patients who had single ovary were identified by chart review of initial pelvic ultrasounds and/or documented history. All one ovary patients reviewed who underwent at least one ART cycle at our center during the study period were included. Next, patients with one ovary were matched three-to-one to patients with two ovaries. A three-to-one matching ratio was utilized to minimize confounding factors and to increase statistical power (given the small sample size of one ovary patients). Matching parameters included patient age at ART cycle start, number of ART cycles completed, ART cycle type (oocyte cryopreservation vs. in vitro fertilization (IVF) vs. embryo banking), and time frame of ART cycle (2012–2015 vs. 2016–2020). The two time frames were chosen given changes in PGT-A methodology at our institution (array comparative genomic hybridization prior was used from 2012 to 2015, and next-generating sequencing was used from 2016 to 2020). If no exact match was available, first priority was given to match by age at ART cycle start and second priority was given to ART cycle type. For example, a one ovary patient who underwent one cycle of oocyte cryopreservation at age 26 in 2012–2015 was paired to 2 two ovary patients who underwent one cycle of oocyte cryopreservation at age 26 in 2012–2015, and 1 two ovary patient who underwent once cycle of oocyte cryopreservation at age 26 in 2016–2020.
Data collection and outcomes
All demographics and outcome variables were collected from the electronic medical record in November 2020. Variables collected included age at initial ART cycle start, day 2 estradiol (E2), day 2 FSH, AMH, total gonadotropin dose, type of gonadotropins used, number of MIIs retrieved per ART cycle, and number of total oocytes retrieved per ART cycle. For IVF and embryo banking cycles, we also obtained the 2 pronuclear (2PN) fertilization rate, blastocyst formation rate, the number of embryos suitable for transfer, and the number of euploid embryos if PGT-A was used. For patients who underwent embryo transfer, we also obtained the number of clinical pregnancies and live births. A clinical pregnancy was defined as a pregnancy with an intrauterine gestational sac on transvaginal ultrasound.
For the one ovary cohort, we also collected the reason for absence of one ovary and the laterality of ovary remaining in situ. Reasons for absence of one ovary were categorized as (1) history of unilateral oophorectomy due to ovarian torsion, (2) benign cyst including endometriosis, (3) ovarian neoplasm, or (4) congenital absence of one ovary. Subgroup analyses were performed for all outcomes comparing reason for absence one ovary and laterality of ovary in situ.
Our primary outcome was number of MIIs retrieved per ART cycle. Secondary outcomes included day 2 E2, day 2 FSH, AMH, total gonadotropin dose, mean daily gonadotropin dose, number of total oocytes retrieved per ART cycle, percentage of MIIs among total oocytes retrieved per ART cycle, 2PN fertilization rate, blastocyst formation rate, number of embryos suitable for transfer, number of euploid embryos, euploidy rate per cycle, clinical pregnancy rate per patient and per transfer, and live birth rate per patient and per transfer.
Ovarian stimulation, laboratory techniques, and embryo transfer
Ovarian stimulation protocols were chosen by the treating physician based on age, ovarian reserve, and semen parameters. All patients received injectable gonadotropins: FSH and/or human menopausal gonadotropin (HMG). One of three ovarian stimulation protocol types were employed: gonadotropin-releasing hormone (GnRH) antagonist, GnRH antagonist with clomiphene citrate, or microdose leuprolide acetate. Oocyte retrieval was performed via ultrasound-guided transvaginal aspiration 35 hours after trigger administration.
In IVF and embryo banking cycles, conventional insemination was used for all cycles, which is standard in our lab, except where intracytoplasmic sperm injection (ICSI) was indicated based on semen parameters or male history. Embryos were then cultured until fresh transfer, trophectoderm biopsy for PGT-A, or cryopreservation based on patient preference and physician orders.
In oocyte cryopreservation cycles, oocytes were vitrified, thawed, and fertilized with ICSI using techniques previously described by our group [20, 21]. Embryos were then cultured until transfer, trophectoderm biopsy for PGT-A, or cryopreservation based on patient preference and physician orders.
Embryo transfer protocols were chosen by the treating physician. A programmed or hormone-replaced embryo transfer protocol was used for patients who underwent fresh embryo transfer. A programmed, natural cycle, or modified natural cycle (with letrozole, clomiphene citrate, or gonadotropins) embryo transfer protocol was used for patients who underwent frozen embryo transfer.
Analysis
Continuous variables were assessed for normality using the Kolmogorov-Smirnoff test, with all determined to be nonparametric. Therefore, Mann–Whitney U and Kruskal–Wallis tests were used to compare continuous variables. Fisher’s exact or chi-squared tests were used for categorical variables. Wilcoxon signed-rank tests were used to compare continuous groups to an expected median. Multiple linear regressions were performed adjusting for significant covariates. An alpha error of 0.05 was considered statistically significant. Results are reported as median or percentages, unless otherwise indicated.
Results
A total of 10,053 patients undergoing 15,847 ART cycles were initially evaluated. Among these patients, 104 patients were identified with one ovary (1.03%); all patients with one ovary underwent at least one ART cycle and therefore met inclusion criteria. The 104 one ovary patients, undergoing a total of 158 ART cycles, were then matched to 312 patients with two ovaries, who underwent 474 ART cycles. Median age at ART cycle start was 35.5 years (range 17–44) for the one ovary cohort and 35.0 years (range 16–44) for the two ovary cohort, with median age at first ART cycle 34.0 years in both groups (one ovary age range 16–43, two ovary age range 17–43) (p = 0.87). Patients in both cohorts underwent a median of one ART cycle (range 1–7). In the one ovary cohort, there were 89 IVF, 17 embryo banking, and 52 oocyte cryopreservation cycles evaluated, compared to 285 IVF, 41 embryo banking, and 148 oocyte cryopreservation cycles in the two ovary cohort. In the one ovary cohort, 39 cycles were completed in 2012–2015 and 119 cycles were completed in 2016–2020; in the two ovary cohort, 137 were completed in 2012–2015 and 337 cycles were completed in 2016–2020. Of the 104 patients in the one ovary cohort, 19 did not have three exact two ovary matches.
Supplemental Table 1 presents the demographics of the one ovary cohort. The most common reason for absence of one ovary was history of a unilateral oophorectomy for an ovarian neoplasm (46%), followed by history of a unilateral oophorectomy for a benign cyst (30%) and history of a unilateral oophorectomy for torsion (18%). Only 6% of patients had a congenital absence of one ovary. Sixty-two percent of patients with one ovary had the left ovary in situ, while 38% of patients with one ovary had the right ovary in situ.
Patients with one ovary yielded a median of 7.5 MIIs and a median of 10 total oocytes per cycle, which was significantly lower than the MII and total oocyte yield in patients with two ovaries (Fig. 1). However, patients with one ovary yielded more than half the number of MII and total oocytes than patients with two ovaries (Z > 5.8, p < 0.01).
Fig. 1.
Median number of MII and total oocytes retrieved per ART cycle by number of ovaries. The bars display the interquartile ranges. MII oocytes, p < 0.01; total oocytes, p < 0.01. Analysis includes 158 one ovary cycles and 474 two ovary cycles
Table 1 presents ART and PGT-A outcomes in the one ovary and two ovary cohorts. The percentage of MIIs retrieved per cycle did not differ among groups. For IVF and embryo banking cycles, 2PN fertilization rate and blastocyst formation rate also did not differ among groups. In the one ovary cohort, 63% (n = 67) of the IVF and embryo banking cycles involved PGT-A; in the two ovary cohort, 67% (n = 217) of the IVF and embryo banking cycles involved PGT-A. For PGT-A cycles, the median number of euploid embryos was 1.0 in both groups; however, there was a trend towards more euploid embryos created in patients with two ovaries (mean 1.3 vs. 2.0, p = 0.03). Still, the euploidy rate per PGT-A cycle did not differ among groups.
Table 1.
Comparison of ART outcomes and PGT-A outcomes by number of ovaries
| One ovary | 95% CI | Two ovaries | 95% CI | p value | |
|---|---|---|---|---|---|
| Laboratory outcomes per cycle | |||||
| % MII of total oocytes retrieved | 79.5 | 77.6–81.4 | 77.6 | 76.6–78.5 | 0.08 |
| 2PN fertilization rate (%) | 75.2 | 72.4–78.0 | 73.8 | 72.4–75.2 | 0.40 |
| Blastocyst formation rate (%) | 57.7 | 53.1–62.2 | 58.2 | 56.1–60.3 | 0.83 |
| PGT-A outcomes per cycle | |||||
| Median number of euploid embryos (IQR) | 1.0 (0–2) | 1.0 (0–3) | 0.03 | ||
| Euploidy rate (%) | 34.2 | 28.4–39.9 | 36.1 | 33.4–38.7 | 0.62 |
Values represent overall percentage, unless otherwise noted. CI, confidence interval; IQR, interquartile range
Figure 2 presents embryo transfer outcomes in the one and two ovary cohorts. Approximately 90% of patients in each cohort who underwent IVF or embryo banking had at least one euploid or untested embryo available for transfer (p = 0.81). In the one ovary cohort, 71% (n = 41) of patients with an embryo suitable for transfer underwent at least one embryo transfer; in the two ovary cohort, 81% (n = 142) of patients with an embryo suitable for transfer underwent at least one embryo transfer. In total, 83 transfers were evaluated in the one ovary cohort compared to 279 transfers in the two ovary cohort. In both groups, patients underwent a median of two embryo transfers (p = 0.91) and a median of one embryo was transferred per cycle, with a trend towards more embryos transferred per cycle in the one ovary group (mean of 1.36 embryos transferred in the one ovary group compared to mean of 1.21 embryos transferred in the two ovary group; p = 0.02). There were no significant differences in clinical pregnancy rates between the one and two ovary groups. Live birth rate per transfer was equivalent in the one and two ovary cohorts (45.8% vs. 46.6%, p = 1.00). When controlling for differences in AMH and day 2 FSH, a multiple linear regression model continued to show no difference in live birth rate among one vs. two ovary groups (B = 0.18, p = 0.84). For patients who underwent ≥ 1 transfer, live birth rate per patient was also equivalent in the one and two ovary cohorts (68.3% vs. 73.9%, p = 0.55).
Fig. 2.
Embryo transfer outcomes by number of ovaries. The following values are displayed: percentage of cases with each outcome, p values. Patients were included in ≥ 1 embryo for transfer group if they had ≥ 1 euploid or untested embryo. CPR, clinical pregnancy rate; LBR, live birth rate
Table 2 presents markers of ovarian reserve prior to initial ART cycle. Of note, AMH, FSH, and E2 were not available for every patient in our study (see the number of patients with an available value for each parameter in Table 2). In patients with one ovary, AMH was lower (1.13 in one ovary group vs. 2.17 in two ovary group; p < 0.01) and day 2 FSH was higher (7.37 in one ovary group vs. 6.20 in two ovary group; p = 0.01). Day 2 E2 did not differ among the two groups (p = 0.29).
Table 2.
Ovarian reserve markers prior to first ART cycle
| One ovary | Two ovary | p value | |
|---|---|---|---|
| AMH (ng/mL) | 1.13 (0.65–2.14) (n = 89, 85.6%) | 2.17 (1.23–4.16) (n = 273, 87.5%) | < 0.01 |
| Basal FSH (IU/mL) | 7.37 (5.00–10.78) (n = 70, 67.3%) | 6.20 (4.70–8.20) (n = 205, 65.7%) | 0.01 |
| Basal E2 (pg/mL) | 34.0 (24.01–51.50) (n = 70, 67.3%) | 37.0 (27.00–56.00) (n = 189, 60.6%) | 0.29 |
Values represent median (interquartile range); n = number of patients with available value, % of cohort
ng, nanograms; IU, international units; pg, picograms; mL, milliliter
In both the one ovary and the two ovary groups, there were negative correlations between patient age and AMH (one ovary cohort rs = − 0.288, p < 0.05; two ovary cohort rs = − 0.272, p < 0.05) as well as between patient age and number of MIIs yielded per cycle (one ovary cohort rs = − 0.219, p < 0.05; two ovary cohort rs = − 0.312, p < 0.05).
Table 3 displays the gonadotropin doses, number of stimulation days, and stimulation protocols used in the one and two ovary groups. The one ovary cohort required significantly more total gonadotropins per cycle (p < 0.01), and the one ovary cohort required a median of one more day of stimulation compared to the two ovary cohort (p = 0.01). The median of each cycles’ mean daily gonadotropin use was significantly higher in the one ovary cohort (409 IU/day, IQR: 348–450) compared to the two ovary cohort (373 IU/day, IQR: 356–413) (p < 0.01). There were no significant differences in stimulation protocols (GnRH agonist vs. GnRH agonist with clomiphene citrate vs. microdose leuprolide acetate) used in the two cohorts. There were no significant differences in type of injectable gonadotropins used in the one and two ovary groups: 78% and 82%, respectively, received both FSH and HMG (p = 0.35), 16% and 12% received FSH only (p = 0.13), and 2% and 4% received HMG only (p = 0.81). When a multiple linear regression was performed to control for total gonadotropins used, it was found that one vs. two ovaries still significantly predicted the number of MIIs retrieved (B = 1.710, p = 0.005).
Table 3.
Stimulation cycles: gonadotropin dose, days of stimulation, and stimulation protocols used
| One ovary n = 158 cycles |
Two ovary n = 474 cycles |
p value | |
|---|---|---|---|
| Median total IUs of gonadotropins (IQR) | 4500 (3600–5400) | 3825 (2650–4500) | < 0.01 |
| Median total IUs of FSH (IQR) | 3000 (2400–3600) | 2475 (1625–3000) | < 0.01 |
| Median total IUs of HMG (IQR) | 1500 (1200–1800) | 1350 (975–1650) | < 0.01 |
| Median days of stimulation (IQR) | 11 (10–12) | 10 (10–12) | 0.01 |
| No. cycles (%) with each stimulation protocol | |||
| GnRH antagonist | 114 (72.2%) | 357 (75.3%) | 0.46 |
| GnRH antagonist + clomiphene citrate | 30 (19.0%) | 91 (19.2%) | 1.00 |
| Microdose leuprolide acetate | 14 (8.9%) | 26 (5.5%) | 0.13 |
IU, international units; FSH, follicle-stimulating hormone; HMG, human menopausal gonadotropin; GnRH, gonadotropin-releasing hormone
When a subgroup analysis was performed for laterality of the single ovary, the only ART outcome that differed was percentage of MIIs retrieved (78.1% for left ovary vs. 82.9% for right ovary; p = 0.02). There were no differences in AMH, day 2 FSH, or day 2 E2 based on laterality of the single ovary. There was also no significant difference in total (p = 0.71) or daily (p = 0.71) gonadotropin dose based on laterality. When a subgroup analysis was performed for reason for absence of one ovary (torsion vs. benign cyst vs. neoplasm vs. congenital absence), there were no differences among any outcomes.
Discussion
Previous literature on ART outcomes in patients with one ovary is conflicting and inconclusive. Our results provide important insight into this debated topic. We demonstrate that ART patients with one ovary have equivalent laboratory outcomes and live birth rates compared to ART patients with two ovaries, despite yielding fewer MII and total oocytes. This new information is helpful for counseling ART patients with one ovary.
Across the one and two ovary cohorts, we found no difference in laboratory outcomes. The percentage of MIIs, 2PN fertilization rate, blastocyst formation rate, median number of euploid embryos, and euploidy rate were equivalent between the one and two ovary groups. Our findings suggest that having one ovary does not negatively impact oocyte quality (maturity or fertilization) or embryo quality (euploidy rate). To our knowledge, our study is the first to evaluate euploidy rate, which is arguably the most important marker of embryo quality, in patients with one ovary compared to two ovaries.
We also found no difference in the percentage of patients with at least one embryo available for transfer, clinical pregnancy rate per transfer or per patient who underwent ≥ 1 transfer, and live birth rate per transfer or per patient who underwent ≥ 1 transfer among the one and two ovary cohorts. These findings are consistent with previous studies that have demonstrated no difference in clinical pregnancy rates in patients with one ovary compared to two [7–10].
Prior studies have suggested that patients with one ovary may have a compensatory mechanism that allows for similar live birth outcomes compared to two ovary patients; however, the precise mechanism of compensation remains unknown. Studies have suggested multiple mechanisms, including ovarian hypertrophy [22, 23], increased follicular recruitment and development [3, 4, 13], formation of oocytes with higher fertilization potential [6], and enhanced oocyte quality leading to increased chance of pregnancy [11]. Given that one ovary patients have half as much ovarian tissue as two ovary patients, one would expect that patients with one ovary would yield half the number of oocytes in response to ART. However, contrary to this expectation, we found that one ovary patients yielded more than half the number of MII and total oocytes than two ovary patients. This finding supports the hypothesis that patients with one ovary have a compensatory mechanism that allows the single ovary to yield a greater number of oocytes. This greater oocyte yield may be due to increased follicular recruitment and development, which was also posited by a recent study by Khan et al. [13]. Therefore, our study not only demonstrates equivalent live birth rates in patients with one ovary, but also suggests a compensatory mechanism that can explain this important finding.
We found a slightly higher basal FSH in the one ovary cohort, consistent with previous reports in the literature [17, 18], though this difference may not be clinically significant. We also confirmed the finding that AMH is lower in patients with one ovary compared to two, which is consistent with previous studies [18]. Our finding that patients with a lower AMH yielded fewer oocytes is also consistent with prior studies that show a correlation between AMH and oocyte yield in the general population [14, 15]. Due to a recent increase in fertility awareness and interest in fertility preservation (autologous oocyte cryopreservation and embryo banking), many patients are now inquiring about AMH and its utility as a marker of both age-related fertility decline and response to ART. Patients are increasingly performing at-home AMH testing and using their results to determine whether and when to pursue fertility preservation. Since patients with a single ovary tend to be concerned about future fertility and often inquire about fertility preservation, understanding the meaning of AMH levels in this subgroup is essential. Our study suggests that patients with one ovary should expect a lower AMH than their two ovary counterparts, but have > 50% the oocyte yield of their two ovary counterparts and still have an excellent chance of embryo transfer and live birth from ART. Therefore, patients with one ovary can be counseled that fertility preservation and ART are both viable options to help them achieve their reproductive goals.
Our study also supports previous observations that patients with one ovary require greater gonadotropin doses for ovarian stimulation compared to patients with one ovary [9, 10, 13]. This is not surprising given that patients with one ovary have lower AMH levels, and physicians tend to choose to use higher gonadotropin doses for patients with lower AMH levels.
One strength of our study is its comparatively large sample size. Our study is one of the largest studies in the literature that reviews ART patients with a single ovary. Additionally, our study reflects modern-day ART technology and therefore provides more up-to-date data compared prior studies (mostly published in the mid-1980s to early 2000s) [3–8, 10, 11, 17]. To our knowledge, this is the first study evaluating patients with a single ovary undergoing oocyte cryopreservation and the first study assessing euploidy as a measure of embryo quality in patients with one ovary compared to two.
Our study was able to detect a difference in our primary outcome (number of MIIs retrieved per ART cycle); however, our study was not powered to detect a type II error in live birth rates. We performed a post hoc power analysis for live birth rate per transfer and live birth rate per patient who underwent ≥ 1 transfer, and the powers for these outcomes were 3.3% and 16.7%, respectively. We also performed a priori power analyses for live birth rate per patient who underwent transfer; if we input α = 0.05, β = 0.2, and aim to detect a 10% difference in live birth rate per patient who underwent transfer (assuming that the live birth rate per patient who underwent transfer in two ovary patients = 73.9%), then 219 one ovary patients and 657 matched two ovary controls would be required attain a power of 80%. In contrast, if we aimed to detect a 15% difference in live birth rate per patient who underwent transfer, then 100 one ovary patients and 300 matched two ovary controls would be required to attain a power of 80%. Given that our sample included 104 one ovary patients and 312 matched two ovary controls, we were powered to detect a 15% difference in live birth rate per patient who underwent transfer, but not a smaller difference. This restricted power is certainly a limitation of our study; however, we did not have more one ovary patients available for inclusion given that patients with one ovary only comprise a small proportion of ART patients.
Another limitation of our study is that patients with one ovary were not matched to patients with two ovaries based on ART indication; this limitation was due to the fact the ART indication for many one ovary patients was history of a prior oophorectomy, which is inherently unmatchable. We chose a three-to-one matching algorithm in order to minimize selection bias that could have occurred regarding ART indication.
Our study was also limited by the fact that we examined a relatively small sample of patients from a single high-volume urban academic institution; therefore, our conclusions may not be generalizable to all patients with one ovary. Further studies with larger samples of patients from diverse geographic locations and ART center types are required to validate our conclusions.
We were unable to find any studies exploring live birth rates after spontaneous conception in women with one ovary. Therefore, it remains unknown whether women with a single ovary have similar live birth rates to women with two ovaries after spontaneous conception, or just after ART. Further studies are needed to explore this topic.
In conclusion, our results provide an important resource to help counsel patients with one ovary about their predicted ART and fertility preservation outcomes. Patients with one ovary can be counseled that while their markers of ovarian reserve and oocyte yield may be less than that of a counterpart woman with two ovaries, their oocyte yield is not expected to decrease by 50%, likely due to a compensatory mechanism in oocyte yield of the remaining ovary. Furthermore, these women can be assured that there is no expected decrease in oocyte quality or laboratory outcomes, such as fertilization rate, blastocyst formation rate, or euploidy rate. Finally, these women can be counseled that they are as just likely to reach embryo transfer as patients with two ovaries and that their chance of live birth appears equivalent to counterpart women with two ovaries. Therefore, while there may be a decrease in AMH and oocyte yield in this population, this decrease does not seem to decrease the likelihood that patients achieve the ultimate goal of live birth.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thankfully acknowledge the patients included in this study.
Declarations
Conflict of interest
The authors declare no competing interests.
Footnotes
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Lass A. The fertility potential of women with a single ovary. Hum Reprod Update. 1999;5:546–550. doi: 10.1093/humupd/5.5.546. [DOI] [PubMed] [Google Scholar]
- 2.Boutteville C, Muasher SJ, Acosta AA, Jones HW, Rosenwaks Z. Results of in vitro fertilization attempts in patients with one or two ovaries. Fertil Steril. 1987;47:821–827. doi: 10.1016/S0015-0282(16)59172-9. [DOI] [PubMed] [Google Scholar]
- 3.Alper MM, Seibel MM, Oskowitz SP, Smith BD, Ransil BJ, Taymor ML. Comparison of follicular response in patients with one or two ovaries in a program of in vitro fertilization. Fertil Steril. 1985;44:652–655. doi: 10.1016/S0015-0282(16)48982-X. [DOI] [PubMed] [Google Scholar]
- 4.Dodds WG, Chin N, Awadalla SG, Miller F, Friedman C, Kim M. In vitro fertilization and embryo transfer in patients with one ovary. Fertil Steril. 1987;48:249–253. doi: 10.1016/S0015-0282(16)59351-0. [DOI] [PubMed] [Google Scholar]
- 5.Diamond MP, Wentz AC, Herbert CM, Pittaway DE, Maxson WS, Daniell JF. One ovary or two: differences in ovulation induction, estradiol levels, and follicular development in a program for in vitro fertilization. Fertil Steril. 1984;41:524–529. doi: 10.1016/S0015-0282(16)47772-1. [DOI] [PubMed] [Google Scholar]
- 6.Lam SY, McKenna M, McBain JC, Baker HW, Johnston WI. Outcome of patients with one ovary in an in vitro fertilization program. J In Vitro Fert Embryo Transf. 1987;4:319–323. doi: 10.1007/BF01555378. [DOI] [PubMed] [Google Scholar]
- 7.Dodson MG, Young RL, Poindexter AN, Gibbons WE, Rossavik IK, Findley WE. Comparison of in vitro fertilization results in women with one or two ovaries. J Reprod Med. 1987;32:359–362. [PubMed] [Google Scholar]
- 8.Lass A, Paul M, Margara R, Winston RM. Women with one ovary have decreased response to GnRHa/HMG ovulation protocol in IVF but the same pregnancy rate as women with two ovaries. Hum Reprod. 1997;12:298–300. doi: 10.1093/humrep/12.2.298. [DOI] [PubMed] [Google Scholar]
- 9.Hendricks MS, Chin H, Loh SF. Treatment outcome of women with a single ovary undergoing in vitro fertilization cycles. Singapore Med J. 2010;51:698–701. [PubMed] [Google Scholar]
- 10.Al-Hasani S, Asimakopoulos B, Nikolettos N, Diedrich K. Comparison of the response to ovarian stimulation between women with one ovary and those with two ovaries, in a program of ICSI/ET. Acta Obstet Gynecol Scand. 2003;82:845–849. doi: 10.1034/j.1600-0412.2003.00201.x. [DOI] [PubMed] [Google Scholar]
- 11.Levitas E, Furman B, Shoham-Vardi I, Lunenfeld E, Potashnik G. Treatment outcome in women with a single ovary versus patients with two ovaries undergoing in vitro fertilization and embryo transfer (IVF/ET) Eur J Obstet Gynecol Reprod Biol. 2000;88:197–200. doi: 10.1016/S0301-2115(99)00155-4. [DOI] [PubMed] [Google Scholar]
- 12.Lind T, Holte J, Olofsson JI, et al. Reduced live-birth rates after IVF/ICSI in women with previous unilateral oophorectomy: results of a multicentre cohort study. Hum Reprod. 2018;33:238–247. doi: 10.1093/humrep/dex358. [DOI] [PubMed] [Google Scholar]
- 13.Khan Z, Gada RP, Tabbaa ZM, et al. Unilateral oophorectomy results in compensatory follicular recruitment in the remaining ovary at time of ovarian stimulation for in vitro fertilization. Fertil Steril. 2014;101:722–727. doi: 10.1016/j.fertnstert.2013.11.019. [DOI] [PubMed] [Google Scholar]
- 14.Dewailly D, Andersen CY, Balen A, et al. The physiology and clinical utility of anti-Mullerian hormone in women. Hum Reprod Update. 2014;20:370–385. doi: 10.1093/humupd/dmt062. [DOI] [PubMed] [Google Scholar]
- 15.Sun XU, Lan YZ, Liu S, et al. Relationship between anti-Mullerian hormone and in vitro fertilization embryo transfer in clinical pregnancy. Front Endocrinol. 2020;11:595448. doi: 10.3389/fendo.2020.595448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Younis JS, Naoum I, Salem N, Perlitz Y, Izhaki I. The impact of unilateral oophorectomy on ovarian reserve in assisted reproduction: a systematic review and meta-analysis. BJOG. 2018;125:26–35. doi: 10.1111/1471-0528.14913. [DOI] [PubMed] [Google Scholar]
- 17.Khalifa E, Toner JP, Muasher SJ, Acosta AA. Significance of basal follicle-stimulating hormone levels in women with one ovary in a program of in vitro fertilization. Fertil Steril. 1992;57:835–839. doi: 10.1016/S0015-0282(16)54967-X. [DOI] [PubMed] [Google Scholar]
- 18.Rustamov O, Krishnan M, Roberts SA, Fitzgerald CT. Effect of salpingectomy, ovarian cystectomy, and unilateral salpingo-oophorectomy on ovarian reserve. Gynecol Surg. 2016;13:173–178. doi: 10.1007/s10397-016-0940-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wilkosz P, Greggains GD, Tanbo TG, Fedorcsak P. Female reproductive decline is determined by remaining ovarian reserve and age. PLoS ONE. 2014;9:1–7. doi: 10.1371/journal.pone.0108343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Grifo JA, Noyes N. Delivery rate using cryopreserved oocytes is comparable to conventional in vitro fertilization for female cancer patients. Fertil Steril. 2010;93:391–396. doi: 10.1016/j.fertnstert.2009.02.067. [DOI] [PubMed] [Google Scholar]
- 21.Noyes N, Knopman J, Labella P, McCaffrey C, Clark-Williams M, Grifo J. Oocyte cryopreservation outcomes including pre-cryopreservation and post-thaw meiotic spindle evaluation following slow cooling and vitrification of human oocytes. Fertil Steril. 2010;94:2078–2082. doi: 10.1016/j.fertnstert.2010.01.019. [DOI] [PubMed] [Google Scholar]
- 22.Biggers JD, Finn CA, McLaren A. Long-term reproductive performance of female mice, effect of removing one ovary. J Reprod Fertil. 1962;3:303–312. doi: 10.1530/jrf.0.0030303. [DOI] [PubMed] [Google Scholar]
- 23.Hartman CG. Observations on the functional compensatory hypertrophy of the opossum ovary. Amer J Anat. 1925;35:1. doi: 10.1002/aja.1000350102. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.