The association between age at menarche and infertility: a systematic review and meta-analysis of observational studies

Abstract

Objective

Both early and late age at menarche have been associated with various health issues and may influence the risk of infertility. This present study investigated the relationship between age at menarche and infertility risk.

Methods

This study follows PRISMA guidelines. Databases including PubMed, Scopus, Web of Science, Embase, and Cochrane were searched in December 2024. Odds ratios with 95% confidence intervals were estimated using a random-effects model. Heterogeneity was assessed with the I² index and chi-square, and publication bias was evaluated using Egger’s test and a funnel plot. Sensitivity analysis and meta-regression examined study impact and variable influence on heterogeneity.

Results

Out of 7,267 articles screened, 18 primary studies were included, yielding 21 pieces of evidence. The odds ratio (OR) for infertility in the late menarche group compared to the normal menarche group was 1.44 (95% CI: 0.98–2.10), while the OR for the early menarche group versus the normal menarche group was 0.98 (95% CI: 0.68–1.42). Additionally, the OR for infertility in the early menarche group compared to the late menarche group was 0.77 (95% CI: 0.55–1.06). For primary infertility, the OR for the late menarche group relative to the normal menarche group was 1.98 (95% CI: 1.02–3.85), whereas the OR for the early menarche group compared to the late menarche group was 0.59 (95% CI: 0.36–0.97).

Conclusion

Although the overall meta-analysis lacked statistical significance, subgroup analysis revealed a notable association between late menarche and primary infertility. Women with late menarche had higher odds of infertility, supporting a dose-responsive relationship. The observed 44% increase in infertility odds highlights late menarche as a potential risk factor, warranting further investigation into its implications for reproductive health.

Introduction

Infertility is defined as the inability to achieve pregnancy after 12 months of unprotected intercourse, which is associated with psychological distress, physical stress, and medical detriments in couples [1–4]. Globally, an estimated 9% of women experience infertility, with secondary infertility being more prevalent [4–6]. Infertility rates are higher in Eastern Europe, North Africa, and the Middle East [2, 5], reflecting geographical differences, which are attributed to various environmental, cultural, social, and economic factors, as well as access to healthcare systems [6, 7]. While various factors contribute to infertility—including age, body mass index (BMI), smoking, and stress—reproductive development, particularly the timing of menarche, has emerged as a potential determinant [1, 4, 8, 9].

Menarche, the onset of menstruation, marks the beginning of ovulation and fertility but does not guarantee them [10–12]. Menarche between 12 and 14 years of age is now considered clinically normal [13], though some researchers have defined other thresholds [14, 15]. The average age at menarche has declined by five years since the mid-19th century in most developed countries and continues to decrease globally [16–19]. Early and late menarche have been linked to multiple adverse health outcomes [7]. Early menarche is associated with increased risks of breast cancer [10, 20], cardiovascular related mortality and diseases [21], and mental health disorders [22, 23], while late menarche has been linked to osteoporosis and bone fractures [14, 24].

Despite increasing evidence on the impact of menarche timing on fertility and pregnancy outcomes, its association with infertility remains unclear [10, 25, 26]. While some studies suggest that both early and late menarche may be linked to infertility, findings have been inconsistent across populations and study designs [4, 10, 26, 27]. Given the conflicting evidence, we conducted this systematic review and meta-analysis to provide a comprehensive evaluation of the relationship between age at menarche and infertility. To our knowledge, this is the first systematic review and meta-analysis to specifically investigate this association.

Materials and methods

This study was designed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [28].

Inclusion and exclusion criteria

The inclusion and exclusion criteria were structured using PICO. The review adhered to pre-defined objectives and eligibility criteria to ensure consistency and focus. We included every peer-reviewed observational case-control, cohort, and cross-sectional study that reported age at menarche in fertile and infertile women, categorized into groups of early and late menarche or early, normal, and late menarche. Only studies in English and Persian from inception until December 6, 2024, were included. Case reports, case series, theses, and conference abstracts were excluded.

Infertility was defined as the failure to achieve a pregnancy after 12 months or more of regular unprotected sexual intercourse. The age at menarche varies across populations due to numerous influencing factors and may also differ within the same population over time. To address this variability, we included only studies that explicitly classified their samples into categories such as early, normal, and late menarche. This approach ensured that each study compared its population internally and reported its findings on the relationship between early, normal, and late menarche and infertility.

Search strategy

Databases including PubMed, Scopus, Embase, Web of Science, Cochrane, and the search engine Google Scholar were searched using specified MeSH and non-MeSH terms, along with the operators OR and AND. The Persian database Scientific Information Database (SID) was searched using Persian keywords for “Age at menarche”, “Infertility”, and “Fertility”. We also carefully reviewed the references of included studies to enhance the sensitivity of our search process. No restrictions were applied to the search. For studies with incomplete information, the authors were contacted to obtain full details. EndNote X9 (Thomson Reuters, America) was used to manage the studies. An example of the PubMed search strategy is as follows:

((“menarch“[All Fields] OR “menarchal“[All Fields] OR “menarche“[MeSH Terms] OR “menarche“[All Fields] OR “menarcheal“[All Fields] OR “menarches“[All Fields]) AND (“infertiles“[All Fields] OR “infertilities“[All Fields] OR “infertility“[MeSH Terms] OR “infertility“[All Fields] OR “infertile“[All Fields] OR “infertility s“[All Fields])) OR ((“menarch“[All Fields] OR “menarchal“[All Fields] OR “menarche“[MeSH Terms] OR “menarche“[All Fields] OR “menarcheal“[All Fields] OR “menarches“[All Fields]) AND (“fertiles“[All Fields] OR “fertility“[MeSH Terms] OR “fertility“[All Fields] OR “fertile“[All Fields] OR “fertilities“[All Fields]))

Study selection

We initially removed duplicates using EndNote software, then screened the studies based on title and abstract to eliminate irrelevant records. The full text of the selected studies was carefully evaluated to review the predefined inclusion and exclusion criteria. This process was conducted independently by two authors, and in case of disagreement, the study was discussed with a third author who made the final decision.

Quality assessment

The risk of bias was assessed using the Newcastle-Ottawa Scale (NOS) checklist [29]. Assessment was done independently by two authors. This checklist includes three sections, namely Selection, Comparability and Exposure, and globally score studies from 0 to 9 points. Studies that obtained a score < 5 points were excluded. Selection criteria maximum score is 4, Comparability is 2 and Exposure is 3.

Data extraction

Data extraction was independently performed by two authors using a pre-defined form. In case of disagreement, the study was discussed with a third author who made the final decision. The extracted data from each study included the following: the first author’s name, study title, year of publication, country of study, study type, type of infertility (primary or secondary or both), number of participants in the fertile and infertile groups, mean age of the fertile and infertile groups with standard deviation, definition of early, normal, and late menarche, and the number of fertile and infertile samples in the groups with early, normal, and late menarche.

Statistical analysis

Data analysis was performed using Stata Ver. 11 software. To estimate the odds ratio (OR) using a two-by-two table, the number of individuals with early, normal, and late menarche was extracted separately for the infertile and fertile women groups from each primary study. Using a random effects model and inverse variance, the ORs were estimated for early menarche compared to normal menarche, late menarche compared to normal menarche, and early menarche compared to late menarche, with a 95% confidence interval (CI). The significance criterion for the OR between the two groups (with and without infertility) was the exclusion of the number one from the upper and lower confidence interval of the OR. Heterogeneity between the results of primary studies was assessed using the I-square index, chi-square, and publication bias with the Egger test and funnel plot. Sensitivity analysis was also conducted to examine the impact of each primary study on the overall estimate. Meta-regression was performed to investigate the impact of the criteria for early menarche, late menarche, and type of infertility (primary/secondary) on heterogeneity among the primary studies.

Results

Based on the search strategy across the mentioned databases, 7276 articles were retrieved. Using EndNote software, 3148 duplicate articles were removed. Subsequently, 3881 articles were excluded after screening the titles and abstracts. Full-text evaluation was performed for 247 articles, and a total of 229 articles were excluded due to being review articles (32 articles), not reporting the necessary data for analysis (158 articles), and lacking a control group (39 articles) (Fig. 1).

Fig. 1.

Process for searching and selecting primary studies

The characteristics of the studies included in this systematic review and meta-analysis, along with the results of the risk of bias assessment, are presented in Table 1. The publication years of the articles ranged from 1993 to 2022. The studies were conducted in China [30], Egypt [31], Ethiopia [32], Iran [33, 34], India [35–40], Pakistan [41], Sri Lanka [42], Turkey [43], and the United States [44–47]. The risk of bias scores for the included articles varied from 5 to 7. Notably, three studies [36, 42, 46] reported data for more than two groups, and we analyzed these groups as independent datasets, which explains the discrepancy between the number of studies and the total pieces of evidence.

Table 1.

Characteristics of primary studies included to the present meta-analysis

ID	First Author (Country, publication year)	Type study	Type infertility	Definition of menarche age		Number of participants with early menarche		Number of participants with late menarche		Number of participants with normal menarche		Quality score*
				Early	Late	Infertile group	Fertile group	Infertile group	Fertile group	Infertile group	Fertile group
1	Rich-Edwards (USA, 1993) [45]	Case-control	Primary infertility	< 11	≥ 17	190	3401	47	277	2290	43,040	6
2	Signorello (USA, 1997) [46]	Case-control	Primary/secondary infertility	≤ 11	≥ 13	9	16	27	48	14	25	5
3	Signorello (USA, 1997) [46]	Case-control	Primary/secondary infertility	≤ 11	≥ 13	5	16	27	48	15	25
4	Samarakoon (Sri Lanka, 2002) [42]	Cross-sectional	Primary infertility	< 14	≥ 14	48	765	33	820	-	-	6
5	Samarakoon (Sri Lanka, 2002) [42]	Cross-sectional	Primary/secondary infertility	< 14	≥ 14	174	765	143	820	-	-
6	Mokhtar (Egypt, 2006) [31]	Case-control	Primary/secondary infertility	≤ 15	> 15	191	214	24	1	-	-	5
7	Delpishe (Iran, 2014) [33]	Cross-sectional	Primary/secondary infertility	< 8	> 14	5	6	42	265	44	550	6
8	Gokler (Turkey, 2014) [48]	Cross-sectional	Primary/secondary infertility	≤ 12	≥ 15	14	132	9	92	50	273	5
9	Saoji (India, 2014) [37]	Case-control	Primary infertility	< 14	> 14	42	25	78	95	-	-	5
10	Yang (China, 2016) [30]	Cross-sectional	Primary infertility	≤ 13	≥ 17	141	1605	96	563	563	4057	7
11	Katole (India, 2019) [35]	Cross-sectional	Primary infertility	< 14	> 14	16	361	35	158	-	-	7
12	Khan (Pakistan, 2019) [41]	Cross-sectional	Primary/secondary infertility	≤ 11	≥ 15	71	33	50	21	302	334	7
13	Bayu (Ethiopia, 2020) [32]	Case-control	Primary/secondary infertility	< 14	≥ 14	30	113	63	75	-	-	5
14	Kamboj (India, 2022) [36]	Cross-sectional	Primary infertility	< 12	> 16	8	17	2	0	65	35	5
15	Kamboj (India, 2022) [36]	Cross-sectional	Primary infertility	< 12	> 16	8	1	5	0	79	114
16	SoriaContreras (USA, 2022) [44]	Cross-sectional	Primary/secondary infertility	< 12	≥ 15	37	144	19	97	153	615	7
17	Banerjee (India, 2023) [38]	Cross-sectional	Primary infertility	≤ 14	≥ 15	61	1014	74	846	-	-	6
18	Jenabi (Iran, 2023) [49]	Case-control	Primary infertility	< 12	> 14	35	56	28	19	137	125	5
19	Kataria (India, 2023) [39]	Cross-sectional	Primary/secondary infertility	≤ 13	> 13	17	112	33	282	-	-	6
20	Sharma (India, 2024) [40]	Cross-sectional	Primary/secondary infertility	< 14	≥ 14	13	200	36	418	-	-	7
21	Wang (USA, 2024) [47]	Cross-sectional	Primary/secondary infertility	< 14	≥ 14	159	1070	32	316	-	-	7

*Quality of studies were assessed using Newcastle-Ottawa Scale Checklist

Comparison of infertility in women with late menarche vs. normal menarche age

Eleven studies compared infertility rates between women with late menarche and those with normal menarche. These studies included 1,782 women in the late menarche group and 52,905 women in the normal menarche group. Among the studies, 63.6% (7/11) reported a higher odds ratio (OR) for infertility in the late menarche group, with statistically significant differences in 27.3% (3/11). In the remaining four studies, where the OR for infertility in the late menarche group was equal to or below one, no statistically significant differences were observed (Fig. 2).

Fig. 2.

Odds ratio of infertility in women with late menarche compared to normal menarche with 95% confidence interval based on primary studies and overall estimate

The pooled OR for infertility in the late menarche group compared to the normal menarche group was 1.44 (95% CI: 0.98, 2.10), with substantial heterogeneity (I² = 79.6%). Neither the Egger test nor the funnel plot indicated publication bias (Fig. 3). Sensitivity analysis showed that no single study significantly influenced the overall estimate (Table 2). Meta-regression analysis, considering late and early menarche criteria and type of infertility (primary/secondary), did not identify significant sources of heterogeneity (Table 3). Subgroup analysis revealed that the OR for primary infertility in the late menarche group compared to the normal menarche group was 1.98 (95% CI: 1.02, 3.85). For studies where the type of infertility was unspecified, the OR was 1.17 (95% CI: 0.71, 1.94).

Fig. 3.

Funnel plot to assess publication bias in the estimated odds ratio of infertility in women with late menarche compared to normal menarche

Table 2.

Pooled estimate of the odds ratio of infertility by menarche age

menarche age	Number of evidence	Sample size (expose group)	Sample size (non-expose)	Pooled estimate of the odds ratio of infertility		Publication Bias total (Egger test)		Sensitivity analysis result (Yes, No) *	Heterogeneity
				OR (95% CI)	Significant impact of menarche age on the infertility (Yes, No)	β	P-value		I−square; %	p−value for Q
Early vs. late menarche age	21	11,340	6164	0.77(0.55, 1.06)	No	0.60	0.590	No	85.6	< 0.001
Early vs. normal menarche age	11	5950	52,905	0.98 (0.68, 1.42)	No	0.63	0.637	No	85.7	< 0.001
late vs. normal menarche age	11	1782	52,905	1.44(0.98, 2.10)	No	-0.27	0.842	No	79.6	< 0.001

*Is there a significant difference in the impact of each of the primary studies on the overall estimate?

Table 3.

Meta-regression in order to investigate the related factors with heterogeneity on odds ratio (OR)

Variables	OR for Early vs. late menarche		OR for Early vs. normal menarche		OR for late vs. normal menarche
	β	P-value	β	P-value	β	P-value
Type of infertility	0.49	0.310	-0.07	0.979	-3.31	0.206
Definition of early menarche	-0.08	0.574	-0.77	0.525	-0.98	0.432
Definition of late menarche	-0.23	0.253	0.03	0.974	0.94	0.338

Comparison of infertility in women with early menarche vs. normal menarche age

Eleven studies also compared infertility rates between women with early menarche and those with normal menarche. These studies included 5,950 women in the early menarche group and 52,905 women in the normal menarche group. Of these studies, 45.5% (5/11) found a higher OR for infertility in the early menarche group, with statistically significant differences in 27.3% (3/11). In six studies where the OR was equal to or below one, significant differences were noted in three cases: Jenabi et al. (OR: 0.57, 95% CI: 0.35, 0.93), Kamboj et al. (OR: 0.25, 95% CI: 0.10, 0.65), and Yang et al. (OR: 0.63, 95% CI: 0.52, 0.77) (Fig. 4).

Fig. 4.

Odds ratio of infertility in women with early menarche compared to normal menarche with 95% confidence interval based on primary studies and overall estimate

The pooled OR for infertility in the early menarche group compared to the normal menarche group was 0.98 (95% CI: 0.68, 1.42), with substantial heterogeneity (I² = 85.7%). Publication bias was not evident based on the Egger test and funnel plot (Fig. 5). Sensitivity analysis showed no significant influence of individual studies on the overall estimate (Table 2). Meta-regression analysis did not identify significant sources of heterogeneity (Table 3). Notably, there was significant variability in the definition of early menarche across studies. In one study (Delpishe et al.), early menarche was defined as occurring before eight years of age, with an OR of 10.42 (95% CI: 3.06, 35.49) for infertility compared to normal menarche.

Fig. 5.

Funnel plot to assess publication bias in the estimated odds ratio of infertility in women with early menarche compared to normal menarche

Comparison of infertility in women with early menarche vs. late menarche

Twenty-one studies compared infertility rates between women with early and late menarche. These studies included 11,340 women in the early menarche group and 6,164 women in the late menarche group. Among these studies, 38.1% (8/21) reported a higher OR for infertility in the early menarche group, with statistically significant differences in 14.3% (3/21). In contrast, 57.1% (12/21) reported a lower OR for infertility in the early menarche group, with statistically significant differences in 28.6% (6/21). One study reported an OR of one (Fig. 6).

Fig. 6.

Odds ratio of infertility in women with early menarche compared to late menarche with 95% confidence interval based on primary studies and overall estimate

The pooled OR for infertility in the early menarche group compared to the late menarche group was 0.77 (95% CI: 0.55, 1.06), with substantial heterogeneity (I² = 85.6%). Publication bias was not evident based on the Egger test and funnel plot (Fig. 7). Sensitivity analysis confirmed that no single study significantly influenced the overall estimate (Table 2). Meta-regression analysis did not identify significant sources of heterogeneity (Table 3). Subgroup analysis revealed that the OR for primary infertility in the early menarche group compared to the late menarche group was 0.59 (95% CI: 0.36, 0.97). For studies where the type of infertility was unspecified, the OR was 0.94 (95% CI: 0.65, 1.38).

Fig. 7.

Funnel plot to assess publication bias in the estimated odds ratio of infertility in women with early menarche compared to late menarche

Discussion

In this systematic review and meta-analysis, the relationship between age at menarche and infertility was investigated. 21 pieces of evidence were analyzed, comparing infertility rates at three levels: late menarche vs. normal, early menarche vs. normal, early menarche vs. late menarche. While the overall meta-analysis results were not statistically significant for any of these comparisons, subgroup analysis revealed a significant increase in the likelihood of primary infertility associated with late menarche. This finding suggests that late menarche plays a significant role in primary infertility, with women who experience late menarche having 98% and 41% higher odds of primary infertility compared to those with normal or early menarche, respectively. Although statistical significance was not reached, the observed effect size—indicating a 44% increase in infertility odds—may still represent a potential clinical concern, particularly for reproductive health assessments. These findings suggest that late menarche likely has a greater impact on infertility than early or normal menarche in a dose-responsive manner, supporting the hypothesis that delayed menarche is a risk factor for infertility.

Cao et al. (2024) conducted a systematic review and meta-analysis examining the relationship between menstrual characteristics such as age at menarche and fertility outcomes [50]. Their findings indicated that early menarche had minimal impact on clinical pregnancy rates, while late menarche was significantly associated with reduced fertility rates. In comparison, our meta-analysis did not find a statistically significant overall association between menarche timing and infertility; however, subgroup analysis revealed a substantial increase in the likelihood of primary infertility among women with late menarche. These findings align with Cao et al.‘s conclusions, reinforcing the idea that late menarche may be a risk factor for infertility. However, while their study focused on broader fertility outcomes, our analysis specifically highlights the impact of late menarche on primary infertility, suggesting a dose-responsive relationship that warrants further investigation.

The relationship between age at menarche and infertility is not fully understood, but several mechanisms explaining how age at menarche affects infertility have been proposed. The onset of menstruation may be related to the size of the ovarian follicular pool and/or the rate of follicular atresia, which in turn predicts the decline in ovarian functional reserve in the future [51]. Anti-Müllerian hormone (AMH) is suggested as a marker of ovarian reserve, produced by pre-antral and small antral follicles, reflecting both the number of these growing small follicles and the number of primordial follicles [52–54]. Studies have shown that late menarche may be associated with lower levels of AMH [50, 55, 56]. In other words, late menarche might lead to a smaller ovarian reserve, which is a known factor in infertility [1, 2, 26]. Conversely, other studies have found that among women seeking infertility treatment, early onset of menstruation is associated with decreased ovarian reserve [25, 51, 57, 58]. Furthermore, early menarche is associated with premature and early menopause [59]. Some researchers speculate that this observed relationship might be explained by unknown prenatal exposures [10].

A lower age at first menstruation increases the incidence of diseases such as pelvic inflammatory disease, which can lead to infertility and spontaneous abortion in later years [43, 60]. In women with early menarche, 50% of their cycles are ovulatory in the first year, and nearly all cycles are ovulatory by the fifth year after menarche. In contrast, it takes about 8 to 12 years for all cycles to become ovulatory in women who experience later menarche [14, 61].

The observed relationship between age at menarche and infertility may be influenced by common underlying factors. Both lower and higher BMI compared to the normal range affect the age at menarche and infertility [7–9, 14, 61–65]. Exposure to environmental toxins, such as endocrine-disrupting chemicals (EDCs), is associated with earlier menarche and infertility markers like a lower antral follicle count (AFC) [7, 18, 66, 67]. There are also hypotheses about the influence of factors like genetics, malnutrition, intense physical exercise, psychological factors (e.g., anorexia nervosa), and chronic diseases (e.g., Autoimmune diseases) on the age at menarche and infertility [19, 68]. Nevertheless, further studies on the relationship between age at menarche and infertility and the investigation of their mechanisms are recommended.

Limitations

Our study has several limitations. One limitation is that the age at menarche was primarily self-reported based on recall. However, previous studies have shown that individuals can recall the age at menarche with reasonable accuracy [63, 69], and self-reported age at menarche for middle-aged women has a moderate correlation with recorded data from adolescence [70]. A key limitation of this study is the reliance on self-reported infertility, which may be subject to recall bias or incomplete disclosure of reproductive history. Additionally, we did not exclude studies that involved populations using oral contraceptives, assisted reproductive technologies, or those with a known history of infertility. While such factors could influence infertility outcomes, excluding these studies would have significantly reduced the available data and compromised the study’s generalizability. Instead, we applied a standardized definition of infertility—failure to conceive after one year of unprotected intercourse—to ensure consistency across included studies. Publication biases and language inclusion criteria might have limited the evaluated studies, but these factors were statistically accounted for. The primary limitation of any meta-analysis is the quality of the primary studies, as some had limited data and could not be included in the analysis.

Conclusion

Although the overall meta-analysis did not yield statistically significant results, subgroup analysis identified a notable association between late menarche and primary infertility. Women with late menarche had 98% and 41% higher odds of primary infertility compared to those with normal and early menarche, respectively. While statistical significance was not reached, the observed 44% increase in infertility odds suggests a potential clinical concern. These findings highlight late menarche as a possible risk factor for infertility, supporting a dose-responsive relationship. Further research is needed to clarify this association and its implications for reproductive health assessments and early intervention strategies.

Abbreviations

BMI

Body mass index

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

AMH

Anti-Müllerian hormone

OR

Odds ratio

CI

Confidence interval

NOS

Newcastle-Ottawa Scale

Author contributions

MM: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Visualization, Supervision, Project administration, Writing - Original Draft, Writing – review & editing. AHB: Methodology, Validation, Investigation, Resources, Data curation, Supervision, Project administration, Writing – review & editing. SMH: Methodology, Investigation, Resources, Data curation, Writing – review & editing. MGT: Conceptualization, Validation, Supervision, Writing – review & editing. SKS: Methodology, Investigation, Resources, Data curation, Writing – review & editing.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.