The association between birth weight/rapid weight gain and early menarche: a systematic review and meta-analysis based on cohort studies.

This meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Statement, and was pre-registered with PROSPERO international prospective of systematic reviews (Registration number: CRD42022327587). We systematically searched for studies published before September 16, 2023 in Pubmed, Embase, Web of Science, and Cochrane Library. The search focused on studies reporting data related to birth weight/rapid weight gain, and early menarche. We used a combination of Medical Subject Headings (MeSH) terms, keywords synonyms, and near synonyms for ‘birth weight’, ‘small for gestational age’, ‘fetal macrosomia’, ‘intrauterine growth retardation’, ‘rapid weight gain’, ‘menarche’, ‘puberty’, ‘youth’ and ‘adolescence’. Manual searches and reference lists of other relevant studies were used to find more potentially eligible studies. Our primary exposure variables of interest were birth weight and rapid weight gain in girls, and the primary outcome variable was early menarche. Birth weight was categorized as low (< 2500 g), reference (2500 to < 4000 g), or high(4000 g) [ 26 ]. Early menarche was defined as the onset of menarche before age 12 [ 2 ]. Rapid weight gain was defined as increasing across two major Centers for Disease Control and Prevention growth percentiles [ 21 , 27 ]. In order to obtain more relevant researches, the title and abstract have been appropriately expanded. First, studies in English reporting birth weight/rapid weight gain and early menarche would be included. The full text of the included studies was then reviewed by the researchers. Included studies must have the following characteristics: (1) were cohort studies; (2) birth weight/rapid weight gain and early menarche were well defined; (3) reported association between birth weight/rapid weight gain and early menarche. The excluded studies had the following characteristics: (1) were intervention studies, reviews, letters to the editor, abstracts of international conferences; (2) cannot provide exact risk values and confidence intervals. In the case of multiple studies from the same population in the same research area, only the most comprehensive or recently published studies were included. Extracted data included: first author, year of publication, study duration, follow-up duration and age of participation, sample size. Data for each included study was independently extracted and evaluated by two authors, and any disagreement was mediated by the Principal Investigator. Two authors (X.Z and CY.C) independently assessed the quality of included studys using the Newcastle–Ottawa Scale (NOS) [ 28 ]. NOS consists of eight entries. Ratings range from 1 to 9 stars, and each study was evaluated on five qualities across three modules: selection, comparability, and outcome or exposure. The final score ≥ 7 was considered high quality. The relationship between birth weight and early menarche was expressed as a risk ratio (RR) in random effects model. Heterogeneity was assessed using the chi-square test and I 2 statistic (I 2  > 50% indicates a high degree of heterogeneity). Egger's linear regression test was used to check whether publication bias existed in the included study ( P  > 0.05 represents no publication bias). Subgroup analyses were used to look for potential sources of heterogeneity: geographical region (e.g. Europe, US, Asia), sample size (for studies of birth weight, the sample size was divided into ≤ 500, > 500; for studies of early weight gain, the sample size was divided into ≤ 1000, > 1000), sample source (e.g. hospital, population), follow-up duration (e.g. ≥ 10 years, < 10 years), rapid weight gain in different periods(e.g. 0-2y, 2-6y),quality assessment(e.g. ≥ 7, < 7) and adjusted confounders (e.g. none, adjusted). Sensitivity analysis excluded one study at a time to assess the robustness of the results. All statistical analyses were performed by RevMan version 5.3 (The Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark) and R version 3.4.1 (The R Foundation for Statistical Computing).

A total of 11,121 related documents were retrieved. After rigorous screening, 21 studies were included in the meta-analysis. [ 16 – 18 , 29 – 46 ]. The study screening flow chart was shown in Fig.  1 . Fig. 1 Study PRISMA diagram Study PRISMA diagram Table 1 showed studies examining the association between birth weight and early menarche. Table 2 showed studies examining the association between rapid weight gain and early menarche. The studies with a total of 159,526 participants were published between 2002 and 2022. Of them, nine studies reported rapid weight gain, sixteen studies reported birth weight, and four studies reported both; eight studies were conducted in US, ten in Europe, one in Asia and two in Australian. The follow-up time ranging from 4 to 23 years. Most studies ( N  = 18) adjusted for possible confounders when assessing the association of birth weight/rapid weight gain with early menarche. Each study included in this meta-analyses was reviewed and critically assessed using the NOS of cohort studies. Table 1 Studies examining the association between birth weight and early menarche (16 studies) First author, Year Geographic region Follow-up (years) Sample size Sample source Adjusted confounders Quality score Koziel 2002 [ 29 ] Poland 14 1060 Population Socio-economic status and body mass index 8 Silva 2002 [ 30 ] UK 15 2058 Population Height and body mass index 8 Romundstad 2003 [ 31 ] Norway 13 3343 Population Parental height, body mass index and maternal early menarche 7 Sloboda 2007 [ 16 ] Australian 14 776 Population Height and body mass index 8 Nadina 2009 [ 32 ] Germany 19 87 Population Body mass index or body fat percentage 1, 2, or 3 y before pubertal growth spurt 9 Terry 2009 [ 18 ] US 7 262 Population Percentile change in weight and height, birth length, family socio-economic status at age 7 years, maternal early menarche 6 Wehkalampi 2011 [ 33 ] Finland 23 259 Hospital Gestational age at birth 8 Wang 2012 [ 17 ] US 5 305 Hospital Smoking during pregnancy 6 D'Aloisio 2013 [ 34 ] US 15 33,501 Population Race/ethnicity, childhood family income 7 Sorensen 2013 [ 35 ] Denmark 13 3446 Population None 6 Behie 2015 [ 36 ] Australian 14 1493 Population None 6 Kelly 2016 [ 37 ] UK 12 5839 Population Income and adiposity 9 Flom 2017 [ 38 ] US 4 923 Population Birth length, and maternal early menarche 8 Aurino 2017 [ 39 ] India and Vietnam 12 2120 Population Maternal height in cm, maternal age at girl’s birth, maternal education in years, urban location at age 1 year, socioeconomic status at 8 year 6 Appelhans 2021 [ 40 ] US 10 426 Population Breastfeeding, maternal menarcheal age, and child race/ethnicity 8 Julia 2021 [ 41 ] Finland 16 298 Population None 6 Table 2 Studies examining the association between rapid weight gain and early menarche (9 studies) First author, Year Geographic region Follow-up (years) Sample size Sample source Adjusted confounders Quality score Silva 2002 [ 30 ] UK 15 2058 Population Height and body mass index 8 Lee 2006 [ 42 ] US 10 354 Population Race, age of maternal menarche, and maternal education 9 Terry 2009 [ 18 ] US 7 262 Population Percentile change in weight and height, birth length, family socio-economic status at age 7 years, maternal early menarche 6 Ong 2009 [ 43 ] UK 13 2402 Population Age, height, mother’s education 7 Morris 2010 [ 44 ] UK 7 81,606 Population Socio-economic status and birth year, gestation length, birthweight 7 Flom 2017 [ 38 ] US 4 923 Population Birth length, and maternal early menarche 8 Appelhans 2021 [ 40 ] US 10 426 Population Breastfeeding, maternal menarcheal age, and child race/ethnicity 8 Aris 2022 [ 45 ] US 10 3772 Population Child’s birth year, race, and Hispanic ethnicity and maternal age at delivery; educational level during pregnancy; annual household income during pregnancy; prenatal cigarette smoking; whether mother was nulliparous; whether mother had gestational diabetes, hypertension, or preeclampsia; mode of delivery; prepregnancy BMI; gestational weight gain; and gestational age at delivery 9 Aghaee 2022 [ 46 ] US 15 15,196 Population Maternal age, education, prior livebirths, maternal gestational weight gain, and girl’s birthweight, gestational age, breastfeeding duration, and race 7 Studies examining the association between birth weight and early menarche (16 studies) Studies examining the association between rapid weight gain and early menarche (9 studies) Figure  2 A showed the pooled risk estimates between high birth weight and risk of early onset of menarche. Overall, there was no statistically difference between low birth weight and early menarche (Fig.  2 B). High birth weight was associated with a low risk of early menarche compared to normal birth weight ( RR  = 0.91 , 95% CI: 0.84–0.98). Significant heterogeneity was found between studies ( P  < 0.001 ; I 2  = 79 % ). No potential publication bias identified using Egger's linear regression test ( t  = −0.8 2, P  = 0.4256). Sensitivity analyses showed that exclusion of any studies did not significantly alter the combined effect of high birth weight/low birth weight on early menarche (Fig.  3 A and B). Fig. 2 Forest plot of the association between birth weight and early menarche ( A ) High birthweight vs Normal birthweight; ( B ) Low birthweight vs Normal birthweight Fig. 3 Forest plot of sensitivity analysis ( A ) High birthweight; ( B ) Low birthweight Forest plot of the association between birth weight and early menarche ( A ) High birthweight vs Normal birthweight; ( B ) Low birthweight vs Normal birthweight Forest plot of sensitivity analysis ( A ) High birthweight; ( B ) Low birthweight Subgroup analyses of the association between low birth weight and early menarche was shown in Table S1. Subgroup analysis showed that the risk of early menarche was higher in low birth weight individuals in Asian, and lower in US populations. Similarly, low birth weight samples from hospitals ( RR  = 3.80, 95% C I: 1.50–9.63) have a higher risk of early menarche than samples from the population. Figure  4 showed the pooled risk estimates between rapid weight gain and risk of early menarche. Rapid weight gain was associated with an increased risk of early menarche ( RR  = 1.41, 95% CI :1.01–1.96). No potential publication bias detected by Egger's linear regression test ( t  = −0.98, P  = 0.3540). Sensitivity analyses showed that exclusion of any studies did not significantly alter the combined effect of rapid weight gain on early menarche (Fig.  5 ). Fig. 4 Forest plot of the association between rapid weight gain and early menarche Fig. 5 Forest plot of sensitivity analysis of rapid weight gain and early menarche Forest plot of the association between rapid weight gain and early menarche Forest plot of sensitivity analysis of rapid weight gain and early menarche All included studies of rapid weight gain were adjusted for confounding factors, so subgroup analyses were performed for geographic region, sample size, follow-up duration, rapid weight gain in different periods, and quality assessment, but not for adjusted confounders (Table S2). In subgroup analyses, rapid weight gain was associated with a higher risk of early menarche when adjusted for birth weight ( RR : 2.21;95% CI :2.15–2.26) compared to analyses without birth weight adjustment ( RR :1.16;95% CI :1.10–1.23) (Fig.  6 ). Fig. 6 Forest plot of the association between rapid weight gain and early menarche after adjusting for birth weight Forest plot of the association between rapid weight gain and early menarche after adjusting for birth weight

This systematic review and meta-analysis aimed to assess the relationship between birth weight/rapid weight gain and the risk of early menarche. Our study showed that high birth weight was associated with a low risk of early menarche compared to low birth weight ( RR  = 0.91). In contrast, rapid weight gain was associated with an increased risk of early menarche ( RR  = 1.41), especially in girls with rapid weight gain between 0 and 2 years of age, who face a 1.42-fold higher risk of early menarche. Subgroup analyses showed that studies from different regions and samples from different sources (hospitals and populations) were both responsible for the high heterogeneity. To our knowledge, this is the first comprehensive meta-analysis to describe the effect of birth weight/rapid weight gain on early menarche. These results can provide theoretical support for preventing excessive weight gain in infancy, and provide new ideas for formulating targeted prevention and intervention measures. Previous systematic reviews had not quantitatively analyzed the relationship between birth weight/rapid weight gain and early menarche, leaving the relationship unclear. Our study provided new evidence that high birth weight was associated with a reduced risk of early menarche, which was consistent with findings from studies in Asian populations [ 39 ] and a cohort study conducted by Kelly et al. in the United Kingdom [ 37 ]. In addition, our results showed that rapid weight gain was associated with an increased risk of early menarche, aligning with cohort studies from Italy and Danish [ 47 , 48 ]. Meanwhile, a cohort study conducted in five low-income and middle-income countries also found that rapid weight gain was associated with an increased risk of early menarche [ 49 ]. Biologically, rapid weight gain was associated with elevated concentrations of insulin-like growth factor 1, insulin resistance, adipokines (e.g., leptin), and elevated concentrations of adrenal androgens. These factors may influence gonadotropin-releasing hormone (GnRH) pulsatility, thus affecting the timing of puberty. In girls, adipokines may play a role in the initiation of menstruation [ 45 , 50 , 51 ]. Studies also suggested that fat cells secrete leptin, lipocalin, insulin-like growth factors and active steroids, which were closely related to the onset of puberty and may act as key signals for the initiation of the hypothalamic-pituitary–gonadal axis [ 52 – 54 ]. However, this biological mechanism has not been fully elucidated. Dietary issues such as time of breastfeeding or eating certain types of food can also be associated with rapid weight gain and/or obesity, also considered a trigger for precocious puberty. Multiple studies have demonstrated that high intake of yogurt, longer duration of breastfeeding, and food insecurity decreased the possibility of earlier menarche, while high intake of animal protein and low intake of plant protein increased that risk [ 55 – 57 ]. This study has several strengths. Firstly, it is the first comprehensive systematic review and meta-analysis to evaluate the effect of birth weight/rapid weight gain on early menarche, filling a significant research gap. Additionally, our study explored the relationship between rapid weight gain after age 2 and early menarche, providing a more nuanced understanding of this relationship. Secondly, we conducted a thorough literature search across three major databases, supplemented by manual searches, ensuring a comprehensive review that accurately reflects the relationship between birth weight, rapid weight gain, and early menarche. Thirdly, our analysis included only cohort studies, with a total sample size of 159,526 participants, which can reflect more real situations and enhance the ability to verify cause and effect, thereby providing more reliable and accurate risk estimates. Despite these strengths, there are some limitations to our study. Firstly, there are four types of abnormal birth weight, high birth weight, low birth weight, very low birth weight and ultra-low birth weight. Due to the lack of relevant data, only the relationship between high birth weight and low birth weight and early menarche was analyzed in this paper. In addition, we were unable to analyse the association of prematurity and small for gestational age with early menarche due to insufficient data. Previous studies showed that all components of pubertal growth occurred earlier in VLBW subjects than in normal birth weight, including menarche, acceleration of pubertal growth, pubertal peak height velocity and adult height attainment. This link indicated that prenatal conditions causing poor growth may influence pubertal maturation later in adolescence, which is consistent with our findings [ 33 , 58 – 60 ].Secondly, although subgroup analyses of multivariate were performed in this study to explore sources of heterogeneity, heterogeneity did exist in this study.Thirdly, due to the limitation of the data included in the study, this study did not make a detailed group analysis of the age of rapid weight gain, and only roughly divided it into 0–2 years old and 2–6 years old, so it was not possible to derive a more specific "sensitive period" for rapid weight gain. Fourthly, the results of our subgroup analyses found that rapid weight gain was associated with a higher risk ratio of early menarche when birth weight was not controlled for ( RR  = 2.21) compared to when it was ( RR  = 1.16). This suggests confounding effects. Subsequent studies need to adjust for birth weight to avoid underestimating the risk ratio of early menarche. Finally, our study included only published studies, which may have resulted in the omission of some important unpublished studies that met the inclusion criteria. Our meta-analysis showed that high birth weight was associated with a reduced risk of early menarche, and rapid weight gain was associated with an increased risk of early menarche. These findings suggested that individuals with abnormal birth weight and rapid weight gain in infancy may need to be closely monitored for precocious puberty to maximize the prevention of later metabolic disease. In the long term, the results of this study may have important implications for further research into early menstrual onset and the development of targeted preventive measures.

Menarche is a key developmental marker of a girl’s healthy transition from childhood into young adulthood, and an important clinical indicator of girls’ physical, nutritional, and reproductive health [ 1 ]. Most studies defined early menarche as the onset of menarche before age 12, though some set the threshold at 10 years old, with others considering it under 13 years old [ 2 ]. Early menarche has become increasingly prevalent worldwide [ 3 , 4 ]. For example, the mean early menarche among South African black females decreased from 14.9 years old in 1956 to 12.4 years old in 2006, reflecting an average decline of 0.50 years old per decade, in contrast to an approximately 0.08 years old per decade decline observed in contemporary Dutch girls [ 5 , 6 ]. In China, the incidence of early menarche among girls was reported at 11.47% [ 3 ]. Early menarche may increase girls' risk of several negative psychosocial and physical effects, including low self-esteem, high levels of anxiety, irritability, and eating disorders [ 7 , 8 ]. Additionally, early menarche was also associated with obesity, various cardiometabolic risk factors [ 9 , 10 ], and chronic conditions such as breast cancer, endometriosis, asthma and type 2 diabetes in adulthood [ 11 , 12 ]. Previous studies have identified multiple factors influencing early menarche, and birth weight/rapid weight gain played an vital role [ 13 , 14 ]. However, the relationship between birth weight/rapid weight gain and early menarche remains contentious. The results of previous studies have shown that low birth weight may be associated with early menarche [ 15 , 16 ]. Studies also showed that high birth weight was associated with early menarche in girls [ 17 ]. Conversely, other research showed that high birth weight was not statistically different from early menarche [ 18 ]. So the relationship between birth weight and early menarche is unclear. Some studies showed that rapid weight gain in infancy was associated with early menarche [ 18 – 21 ], but others suggested there may be no link [ 22 – 24 ]. To date, only one systematic review has addressed the connection between birth weight/rapid weight gain and early menarche. In this study, the relationship between birth weight and early menarche was contradictory in different studies and had not conclusively analyzed quantitatively. In addition, this study revealed that rapid weight gain in infants at 6–9 months was associated with early menarche, but the relationship between rapid weight gain after 2 years of age and early menarche was not clear [ 25 ]. Given the conflicting evidence surrounding the association between birth weight/rapid weight gain and early menarche, we aim to conduct a comprehensive systematic review and meta-analysis to clarify these relationships. Our goal is to provide a foundation for developing targeted preventive strategies for early menarche.

Supplementary Material 1. Table S1: Subgroup analyses for the association between low birth weight and early menarche. Supplementary Material 2. Table S2: Subgroup analyses for the association between rapid weight gain and early menarche. Supplementary Material 3. Table S3: Bibliographic search strategies. Supplementary Material 4. Supplementary Material 1. Table S1: Subgroup analyses for the association between low birth weight and early menarche. Supplementary Material 2. Table S2: Subgroup analyses for the association between rapid weight gain and early menarche. Supplementary Material 3. Table S3: Bibliographic search strategies. Supplementary Material 4.