Association between menstrual cycle length and ovarian reserve markers and stimulation response: a retrospective cohort study

Abstract

Aim

To evaluate the relationship between menstrual cycle length (MCL) categories and ovarian reserve biomarkers and controlled ovarian stimulation (COS) response levels.

Materials and methods

Medical records of women undergoing infertility evaluations between January 2022 and April 2025 were reviewed. A total of 550 women undergoing fresh IVF/ICSI cycles were screened for eligibility, and 440 normo-ovulatory women who met all inclusion criteria were included. Patients were grouped according to MCL (21–28 and 29–35 days). Demographic characteristics (age, body mass index), baseline hormonal parameters (anti-Müllerian hormone (AMH), cycle day-3 follicle-stimulating hormone, luteinizing hormone, and estradiol), and antral follicle count (AFC) were recorded. COS outcomes including the number of retrieved and metaphase II (MII) oocytes, follicular output index (FOI), and follicle-to-oocyte index (FORT) were documented. Fertilization was achieved via IVF or ICSI, and embryo transfer was performed on days 2–5.

Results

AMH levels were significantly lower in the MCL (21–28 days) group than in the MCL (29–35 days) group (p < 0.001). AFC levels were significantly lower in the MCL (21–28 days) group than in the MCL (29–35 days) group (p = 0.003). The number of oocytes developed after stimulation was significantly lower in the MCL (21–28 days) group than in the MCL (29–35 days) group (p < 0.001). The number of MII oocytes was significantly lower in the MCL (21–28 days) group than in the MCL (29–35 days) group (p < 0.001). The clinical pregnancy rate was significantly lower in the MCL (21–28 days) group than in the MCL (29–35 days) group (p = 0.017).

Conclusion

Within the normal range of MCLs, longer cycles are associated with higher AMH and AFC levels and greater oocyte yields after controlled ovarian stimulation. Although women with longer cycles showed higher crude clinical pregnancy rates, MCL was not an independent predictor of clinical pregnancy once ovarian reserve markers were accounted for. These findings suggest that MCL is a simple, history-based indicator of underlying ovarian reserve and ovarian response.

Introduction

Menstrual cycle length (MCL) reflects follicular phase duration and follicular growth dynamics, providing an accessible clinical clue to the integrity of the hypothalamic–pituitary–ovarian axis. Although MCL gradually decreases with advancing age, primarily due to shortening of the follicular phase, substantial interindividual variation in ovarian reserve exists among women of the same chronologic age [1, 2]. Accordingly, MCL has been proposed as a measure of quantitative and qualitative dimensions of ovarian reserve that may operate independently of chronologic age.

Systematic reviews and meta-analyses indicate that, among women with regular cycles, a short MCL (e.g., 21–27 days) is associated with lower anti-Müllerian hormone (AMH) levels and reduced antral follicle count (AFC), decreased fecundability in natural cycles, and, within assisted reproductive technologies (ART), fewer retrieved oocytes and lower clinical pregnancy rates [1]. Prospective cohort data further show that higher AMH is associated with longer MCL, whereas low AMH correlates with a shorter follicular phase [2]. These findings suggest that MCL may serve as an indicator of ovarian aging independent of chronologic age.

In ART populations, large cohort studies report that longer MCL is positively, and age-independently, associated with in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) outcomes (pregnancy/live birth), whereas short MCL is linked to higher gonadotropin requirements, fewer oocytes, and lower embryo quality [3]. Consequently, MCL has been proposed as a low-cost, accessible, history-based adjunct to biochemical (AMH, follicle-stimulating hormone (FSH), inhibin B) and sonographic (AFC, ovarian volume) assessments of ovarian reserve [3, 4]. In the literature, MCL is commonly analyzed in short (21–27 days), normal (28–31 days), and long (32–35 days) strata; this framework suggests that, even within the normal range, ovarian reserve biomarkers and reproductive outcomes may vary gradually in parallel with cycle length [1–3]. In clinical practice, however, considering the normal range in two groups (21–28 vs. 29–35 days) is more practical. This categorization enables direct evaluation of whether AMH/AFC profiles and controlled ovarian stimulation (COS) response metrics (numbers of growing/retrieved/MII oocytes after stimulation, clinical pregnancy outcomes) differ between the 21–28 and 29–35-days groups [1–4].

The aim of this study was to evaluate the relationship between MCL categories and ovarian reserve biomarkers and COS response levels.

Materials and methods

This is a retrospective cohort study conducted at the Ege University Faculty of Medicine, Center for Assisted Reproductive Technologies. Informed consent was obtained from all participants. The study was performed in line with the principles of the Declaration of Helsinki. The study was approved by the Ege University Clinical Research Ethics Committee (Date: 26.06.2025, File Number: 25-6.1T/7).

The medical records of women who underwent infertility evaluation and fresh IVF/ICSI treatment between January 2022 and April 2025 were retrospectively reviewed.

Eligibility criteria

A total of 550 women undergoing fresh IVF/ICSI cycles between January 2022 and April 2025 were screened for eligibility. Inclusion criteria were: women aged 24–42 years with normo-ovulatory, eumenorrheic cycles undergoing fresh IVF/ICSI treatment and having complete baseline ovarian reserve data. Exclusion criteria included polycystic ovary syndrome (n = 34), endocrine disorders affecting the hypothalamic–pituitary–ovarian axis such as diabetes mellitus, thyroid dysfunction, or hyperprolactinemia (n = 24), prior chemotherapy or radiotherapy (n = 5), severe male factor infertility (n = 16), ovarian mass or history of ovarian surgery (n = 10), and age < 24 or > 42 years (n = 21). After applying these criteria, 440 normo-ovulatory women were included in the final analysis (Fig. 1).

Fig. 1.

Study flow diagram of patient inclusion and outcomes

The patients’ demographic characteristics, including age and body mass index (BMI), were recorded. Baseline hormonal assessments were performed at the beginning of the treatment cycle, including serum AMH levels and day-3 measurements of FSH, luteinizing hormone (LH), and estradiol (E2). On day 3 of the menstrual cycle, antral follicle counts (AFC) were determined using transvaginal ultrasonography. During COS, the number of developing follicles was monitored, and on the day of oocyte retrieval (oocyte pick-up - OPU), both the total number of oocytes retrieved and the number of mature (MII) oocytes were recorded. In addition, the follicular output index (FOI) and follicle-to-oocyte index (FORT) were calculated for each patient.

MCL was defined as the interval from the first day of bleeding to the day before the next menses. At treatment initiation, patient-reported MCL was recorded in the electronic medical records. Patients were categorized into two predefined groups based on MCL: 21–28 days and 29–35 days. COS was performed according to ovarian reserve parameters and center standards using either a long gonadotropin releasing hormone (GnRH) agonist or short GnRH antagonist protocol. The starting gonadotropin dose ranged between 150 and 300 IU per day and was individualized according to age, AMH level, AFC, and clinician assessment. Dose adjustments were performed during stimulation based on follicular growth dynamics and serum estradiol levels, in line with routine clinical practice at our center. Ovulation trigger was administered using urinary hCG or recombinant hCG when the leading follicle measured 18–22 mm with appropriate serum E2 levels. OPU was scheduled 35 h after trigger. Fertilization was achieved by IVF or ICSI. Embryo transfer was planned on day 2, 3, or 5 according to embryonic development. Luteal phase support was provided per center protocol. Recorded covariates included age, body mass index (BMI), AMH (ng/mL), AFC; both ovaries (2–10 mm, measured on cycle days 2–4), and basal FSH, LH, and E2 (cycle days 2–4). COS response and reproductive outcomes comprised the number of developing follicles/oocytes after stimulation, the number of oocytes retrieved at OPU, the number of metaphase II (MII) oocytes, and clinical pregnancy, defined as an intrauterine gestational sac with fetal cardiac activity on ultrasound.

Statistical analysis

Statistical analysis was performed using the SPSS version 26.0 software package (IBM Inc., Chicago, IL, USA). The normality of the data distribution was evaluated using the Kolmogorov–Smirnov test. Descriptive statistics are presented as mean ± standard deviation, median (interquartile range), or n (%). Student’s t-test or the Mann–Whitney U test was used for continuous variables, and the Chi-square or Fisher’s exact test was used for categorical variables in between-group comparisons. Multivariable linear (for continuous outcomes) and logistic regression (for binary outcomes) models adjusted for age, BMI, AMH, AFC, and stimulation protocol were used to evaluate associations between MCL categories and COS responses/clinical pregnancies. Age-adjusted partial Pearson (rₚ) and partial Spearman (ρₚ) correlation coefficients with corresponding p-values were reported. To control the family-wise error rate at 5%, a Bonferroni correction was applied across the two co-primary endpoints, using a two-sided α = 0.025 for each. Results were evaluated at 95% confidence intervals (CI). As this study was designed as a retrospective cohort analysis including all eligible patients within the predefined study period, no a priori sample size calculation was applicable.

Results

The mean age of the 29–35 days MCL group was found to be younger than that of the 21–28 MCL group (p = 0.030). AMH levels were significantly lower in the MCL (21–28 days) group than in the MCL (29–35 days) group (p < 0.001). FSH levels were significantly higher in the 21–28 days MCL group than in the 29–35 days MCL group (p = 0.008). E2 levels were significantly higher in the 21–28 days MCL group than in the 29–35 days MCL group (p = 0.045) (Table 1).

Table 1.

Comparison of demographic and laboratory data between groups

Variables	MCL (21–28 days)	MCL (29–35 days)	p
	(Mean ± SD)
Age (years)	34.09 ± 4.47	33.13 ± 4.37	0.030
BMI (kg/m²)	25.27 ± 5.05	25.93 ± 4.49	0.055
AMH (ng/mL)	1.07 ± 0.92	1.89 ± 1.42	< 0.001
FSH (IU/L)	9.51 ± 5.34	8.05 ± 3.66	0.008
LH (IU/L)	6.41 ± 3.23	6.18 ± 3.18	0.414
E2 (pg/mL)	46.50 ± 19.10	43.23 ± 15.84	0.045

AFC levels were significantly lower in the 21–28 days MCL group compared with the 29–35 days MCL group (p = 0.003). Similarly, after stimulation, fewer oocytes developed in the 21–28 days MCL group than in the 29–35 days MCL group (p < 0.001). The number of oocytes retrieved was also significantly reduced in the 21–28 days MCL group vs. the 29–35 days MCL group (p < 0.001). Additionally, participants in the 21–28 days MCL group had considerably fewer MII oocytes than those in the 29–35 days group (p < 0.001). Lastly, the clinical pregnancy rate was significantly lower in the 21–28 days MCL group compared with the 29–35 days MCL group (p = 0.017; Table 2).

Table 2.

Comparison of stimulation results between groups

Variables	MCL (21–28 days)	MCL (29–35 days)	p
	(Mean ± SD)
AFC (n)	6.93 ± 4.36	8.67 ± 5.39	0.003
Oocytes developed after stimulation (n)	5.88 ± 2.76	7.30 ± 3.36	< 0.001
Oocytes retrieved (n)	5.65 ± 3.84	7.57 ± 4.72	< 0.001
MII oocytes (n)	4.53 ± 3.23	5.85 ± 3.63	< 0.001
Follicular Output Index (FOI, %)	101.68 ± 90.20	119.30 ± 116.31	0.061
Follicle-to-Oocyte Index (FORT, %)	112.41 ± 95.01	128.00 ± 132.89	0.918
Clinical Pregnancy Rates (n-%)	71/310 (22.9%)	44/130 (33.8%)	0.017

Even after adjusting for the effect of age, the 29–35-days MCL group maintained an independent advantage in favor of ovarian reserve (higher AMH and AFC, lower FSH) and oocyte outcomes (greater numbers of developing, retrieved, and MII oocytes). This pattern, consistent with the unadjusted analyses, suggests that longer MCL (29–35 days) is positively associated with ovarian reserve and oocyte yield independently of age, yet does not demonstrate a clear or consistent difference in proportional response metrics (FOI/FORT) (Table 3).

Table 3.

Linear regression analysis (age-adjusted)

Variables	β (29–35 vs. 21–28)	95% CI	p
FSH (IU/L)	−1.20	−2.03 to − 0.36	0.005
LH (IU/L)	−0.16	−0.82 to 0.50	0.632
E2 (pg/mL)	−3.37	−6.82 to 0.09	0.056
AFC	+ 1.52	+ 0.51 to + 2.53	0.003
AMH (ng/mL)	+ 0.74	+ 0.49 to + 0.98	< 0.001
Developed oocytes (n)	+ 1.23	+ 0.63 to + 1.84	< 0.001
Retrieved oocytes (n)	+ 1.69	+ 0.82 to + 2.56	< 0.001
MII oocytes (n)	+ 1.14	+ 0.46 to + 1.83	0.001
FOI (%)	+ 16.45	−6.67 to + 39.58	0.163
FORT (%)	+ 15.35	−9.70 to + 40.41	0.230

Even after adjusting for the effect of age, the likelihood of achieving a clinical pregnancy remained higher in the group with longer MCL (29–35 days) (p = 0.024). In the multivariate logistic regression analyses, no statistically significant association was found between MCL and clinical pregnancy (OR = 1.31; 95% CI: [0.81–2.13]; p = 0.273). Conversely, each one-unit increase in AMH level was associated with approximately a 35% higher likelihood of achieving pregnancy, indicating a significant positive relationship between AMH levels and clinical pregnancy probability (OR = 1.35; 95% CI: [1.07–1.71]; p = 0.013) (Table 4).

Table 4.

Multivariate logistic regression analyses for clinical pregnancy

Variables	OR	95% CI	p
MCL group (29–35 vs. 21–28 days)	1.31	0.81–2.13	0.273
Age (years)	1.01	0.95–1.06	0.823
BMI (kg/m²)	1.03	0.98–1.07	0.219
AMH (ng/mL)	1.35	1.07–1.71	0.013
AFC (n)	1.01	0.95–1.06	0.812

Age-adjusted partial correlations likewise showed positive associations of longer MCL with AMH/AFC and oocyte counts, and a negative association with FSH, with no association for FOI/FORT. Collectively, these findings indicate that, independent of age, longer menstrual cycles are associated with a more favorable ovarian-reserve profile and higher oocyte yield, and any potential effect on clinical pregnancy is likely mediated by reserve markers such as AMH and AFC (Table 5).

Table 5.

Age-adjusted partial correlations between menstrual cycle length and study variables

Variables	Partial r (Pearson)	p (Pearson)	Partial ρ (Spearman)	p (Spearman)
BMI (kg/m²)	0.061	0.198	0.092	0.054
AMH (ng/mL)	0.311	< 0.001	0.245	< 0.001
FSH (IU/L)	-0.114	0.017	-0.105	0.027
LH (IU/L)	-0.023	0.634	-0.029	0.550
Estradiol (pg/mL)	-0.084	0.079	-0.096	0.044
Antral follicle count (AFC)	0.149	0.002	0.125	0.009
Developed oocytes (n)	0.196	< 0.001	0.193	< 0.001
Retrieved oocytes (n)	0.189	< 0.001	0.178	< 0.001
MII oocytes (n)	0.158	< 0.001	0.167	< 0.001
Follicular Output Index (FOI, %)	0.076	0.112	0.087	0.067
Follicle-to-Oocyte Index (FORT, %)	0.065	0.175	0.003	0.954

Discussion

This study demonstrates that MCL is a clinically meaningful determinant of ovarian reserve and ovarian response. Women with longer cycles (29–35 days) had significantly higher AMH levels and AFCs, and greater oocyte yields compared with women with shorter cycles (21–28 days). The persistence of these associations after adjustment for age suggests that MCL is a marker reflecting intrinsic ovarian biology [1, 2]. In other words, MCL is a simple and non-invasive clinical parameter that provides clues about the size of the follicle pool and the developmental potential of follicles.

It has long been known that short cycles are associated with diminished ovarian reserve. In the meta-analysis by Younis et al., short cycles were reported to be associated with lower AMH levels, lower AFCs, and higher FSH levels [1]. In the prospective cohort study by Harris et al., AMH levels were also shown to vary positively with MCL, meaning that AMH levels increased as cycle length increased [2]. These findings indicate that the duration of the follicular phase is related to the size of the follicle pool. Studies focusing on hormonal dynamics, such as the BioCycle study, have also shown that short cycles are associated with higher E2 levels during the follicular phase, earlier rises in FSH, and alterations in the timing of ovulatory processes [3]. These findings point to accelerated follicular development and changes in hypothalamic–pituitary–ovarian axis response patterns, which are consistent with the lower reserve and reduced oocyte yield observed in our study.

Large-scale studies conducted in ART populations have also revealed similar findings. Brodin et al. reported that women with longer MCL obtained more oocytes, required lower gonadotropin doses, and achieved higher clinical pregnancy rates [4]. Gizzo et al. suggested that MCL provides complementary contributions to reserve parameters such as AMH and AFC and may therefore enhance prognostic accuracy [5]. Our findings are in complete agreement with this literature, indicating that women with longer cycles have a larger follicle pool and respond better to COS.

The absence of significant differences in FOI and FORT in our study demonstrates the distinction between absolute response and proportional response. FOI is defined as an index that reflects granulosa cell responsiveness to gonadotropins and ovarian sensitivity rather than the size of the basal follicle pool [6, 7]. FORT has been reported to quantitatively reflect antral follicle responsiveness to FSH during controlled ovarian hyperstimulation and to be influenced by factors such as the stimulation protocol used, gonadotropin type, and dose [8–11]. Therefore, although women with longer cycles have more follicles and more oocytes retrieved, proportional indices such as FOI and FORT do not change; the pool is larger, but the proportion utilized remains constant.

Changes in the follicular phase of the menstrual cycle are among the earliest indicators of ovarian aging. Santoro et al. identified shortening of the follicular phase as one of the first clinical signs of reproductive aging [12]. Hale et al. demonstrated that hormonal dynamics in the late reproductive years were closely associated with cycle length [13]. AMH is known to be the most sensitive biomarker reflecting the decline in the follicle pool, and this biologic process progresses in parallel with shortening of the follicular phase [14]. Broekmans’ comprehensive evaluation of the pathophysiology of diminished ovarian reserve showed that follicle loss, cycle shortening, and hormonal changes were parts of the same mechanism [15]. Depmann’s mathematical models demonstrated that decreased follicle pool size directly affected cycle length, and O’Connor et al. defined follicular phase variability as an early sign of ovarian aging [16, 17].

This biological framework further strengthens the clinical relevance of our findings. The POSEIDON classification groups patients into prognostic categories based on ovarian reserve and expected oocyte yield [18]. Women with short cycles and low reserve are more likely to fall into poor-prognosis groups (groups 3–4), whereas women with longer cycles generally correspond to groups associated with better prognosis. Incorporating MCL into such individualized classification systems may provide additional value in predicting ovarian response.

In conclusion, MCL emerges as a robust clinical parameter reflecting both ovarian reserve and ovarian response. Women with longer cycles consistently demonstrated higher AMH levels and AFCs, and greater oocyte yields, and these associations remained significant even after adjusting for age, indicating that MCL might mirror intrinsic ovarian biology rather than chronologic aging alone. Although MCL did not independently predict clinical pregnancy after accounting for AMH and AFC, its strong correlation with key reserve markers highlights its potential value in patient counseling, initial prognostic assessment, and individualized stimulation planning. As a simple, history-based metric, MCL may provide complementary insight, particularly in settings where biochemical testing is unavailable or delayed. Longitudinal studies are needed to determine whether temporal changes in MCL parallel the trajectory of ovarian reserve decline and if incorporating MCL into predictive models can enhance reproductive prognostication.

Strengths and limitations

This study has several limitations. First, its retrospective design may introduce selection bias despite predefined eligibility criteria. Second, menstrual cycle length was based on patient self-report, which may be subject to recall bias. Third, the study was conducted at a single tertiary ART center, which may limit the generalizability of the findings to broader populations, particularly to women not undergoing IVF/ICSI treatment.

Although multivariable analyses were performed adjusting for age, BMI, AMH, AFC, and stimulation protocol, residual confounding cannot be completely excluded. Additionally, the unequal distribution between MCL groups (310 vs. 130 patients) may have influenced statistical comparisons, although the sample size was sufficient for detecting clinically meaningful differences.

Strengths of the study include the relatively large sample size, homogeneous inclusion criteria, detailed ovarian reserve assessment, and comprehensive evaluation of both absolute and proportional ovarian response indices.

Abbreviations

AFC

Antral Follicle Count

AMH

Anti-Müllerian Hormone

ART

Assisted Reproductive Technologies

BMI

Body Mass Index

CI

Confidence Interval

COS

Controlled Ovarian Stimulation

E2

Estradiol

FOI

Follicular Output Index

FORT

Follicle-to-Oocyte Index

FSH

Follicle-Stimulating Hormone

GnRH

Gonadotropin-Releasing Hormone

hCG

Human Chorionic Gonadotropin

hMG

Human Menopausal Gonadotropin

ICSI

Intracytoplasmic Sperm Injection

IVF

In Vitro Fertilization

LH

Luteinizing Hormone

MCL

Menstrual Cycle Length

MII

Metaphase II

OPU

Oocyte Pick-Up

OR

Odds Ratio

PCOS

Polycystic Ovary Syndrome

POSEIDON

Patient-Oriented Strategies Encompassing Individualized Oocyte Number

rFSH

Recombinant Follicle-Stimulating Hormone

Authors’ contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by BE, ENTG, UA, FA, GA and FTK. The first draft of the manuscript was written by ENTG, ET and all authors commented on previous versions of the manuscript. All authors read and approved of the final manuscript.

Funding

No funding was received for conducting this study. The authors have no financial or proprietary interest in any material discussed in this article.

Data availability

The datasets generated during the current study were available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study protocol was approved by the Ege University Clinical Research Ethics Committee (Date: 26.06.2025, File Number: 25-6.1T/7). Although the study was retrospective in design, written informed consent for participation in clinical research and use of anonymized medical data was obtained from all patients at the time of treatment initiation, in addition to the standard procedural consent for IVF/ICSI. The study was conducted in accordance with the Declaration of Helsinki.

All patient data were anonymized prior to analysis and handled in accordance with the Turkish Personal Data Protection Law (Law No. 6698) and institutional data protection regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.