Table of Contents /Previous Topic/Next Topic / Index

Original and Papers

A Cohort Study of Correlation between Body Mass Index and Age at Menarche in Healthy Bengali Girls

I Banerjee*, S Chakraborty**, N Ghia Bhattacharyya***, S Bandyopadhyay****, H N Saiyed*****, D Mukherjee******

.

Key words : Menarche, Bengali, body mass index, cohort study.

There appears to be a trend for decreasing age at menarche worldwide. Other factors aside, it is often assumed that this trend is due to an increase in body fat mass secondary to excess calorie consumption. The evidence for this is not compelling. There are reports1-3, often with methodological inadequacies, of decreasing age at menarche with increasing body weight. In contrast, there appears to be a lack of association of weight gain and menarche observed in a major longitudinal study in North American children4. There has been an increased incidence of obesity in the Indian subcontinent5 and the question whether an earlier menarche is attributable to weight gain, is therefore an important one. It is not known if comparatively greater body mass index (BMI) standard deviation scores (SDS) are associated with lower age at menarche in Bengali adolescents. In this study, BMI was determined in a group of Bengali adolescent girls from a middle income family background and investigated the possible association with age at menarche. The aims of the study were the following: (1) To determine the age at menarche in healthy Bengali girls in a longitudinal follow-up cohort. (2) To investigate if BMI SDS at menarche or at or one year preceding menarche was in correlation with age at menarche. (3) To ascertain if change in BMI in the year preceding menarche was in correlation with age at menarche.
.

Material And Method

Healthy girls (age range 8-16 years) of adequate nutrition and from middle income families were consented and recruited from an inner area school in North Kolkata6. This cohort was assessed longitudinally between 1998 and 2001 with serial measurements undertaken on a designated day every year. Age was recorded in decimal years. Standing height was measured using a stadiometer by a single trained observer and recorded to the nearest 0.1cm. Weight was recorded by a SECA beam balance to the nearest 0.1kg. These values were converted to BMI = weight in kg/ (height in metres)2. National Center for Health Statistics (NCHS) BMI charts (year 2000) were used to calculate BMI SDS with the aid of growth analyser version 3.0 software (Dutch Growth Foundation, Ferring Pharmaceuticals). Breast staging was adopted as the method to assess puberty using the sex maturity rating of Tanner7. Age at menarche was ascertained by prospective questioning in longitudinal assessments and by recall in those girls who were postmenarcheal.
The Kolmogorov-Smirnoff method was used to test for normality of data. When normally distributed, analysis of variance (ANOVA) was undertaken to test for differences between groups. Univariate ANOVA was used to test for the interaction of factors and covariates on the dependent variable. Bivariate correlations were tested using Pearson or Spearman rank correlation. The data was stored in SPSS 11.5 (SPSS Inc, Chicago, Illinois) and all statistical analyses, parametric or non-parametric, were performed using its programmes.

. Observations

Longitudinal assessments were performed in 349 girls and data was obtained on the age at menarche in 273 girls. Menarcheal age was not normally distributed (p<0.001) and was skewed to the right [skewness 0.43, standard error (SE) 0.14]. The median age at menarche of girls in our group was 12.0 (range 10.0 – 16.0) years. The frequencies and percentages of girls achieving menarche at different ages are depicted in Table 1.
Data on the stage of breast development was available in 170 girls in the cohort. Fifty-seven (33%), 88 (52%) and 25 (15%) girls were at breast stages 3, 4 and 5 respectively. Like menarcheal age, BMI at menarche was skewed to the right [skewness 0.65, SE 0.20] with median BMI of 16.4 (range 12.5 – 23.3) kg/m2. However on conversion to SDS, BMI SDS at menarche was normally distributed [mean (SD) –0.9 (0.9), p=0.2] in the whole group as well in the individual breast stages 3 (p=0.2), 4 (p=0.2) and 5 (p=0.2). The BMI SDS values for breast stage of girls are shown in Table 2. There was no significant difference of BMI SDS between the 3 breast stages in ANOVA (p=0.1).

As expected, breast staging was positively correlated to the age at menarche (r=0.20, p=0.008). BMI at menarche had weak linear positive correlation with age at menarche (r=0.27, p=0.001), indicating an increment in BMI with age. BMI at one year preceding menarche was not correlated with age at menarche (p=0.2). On conversion of BMI to BMI SDS, thereby attenuating the age dependency of BMI, there was no correlation of BMI SDS at menarche (Fig 1, p=0.5) or BMI SDS at one year preceding menarche (Fig 2, p=0.8) with age at menarche. Menarche can occur at different breast stages, usually between stages 3 and 5. It is therefore assessed in this cohort that breast staging had an independent effect on age at menarche. Univariate ANOVA was performed with age at menarche as the dependent variable, breast stage as a factor, and BMI at menarche as a covariate. The interaction of the variables breast stage and BMI at menarche did not significantly influence menarcheal age (p=0.5). In other words, for a given breast stage, relative BMI increments did not determine a reduction in the age at menarche.
It is well known for BMI to increase through puberty6. Therefore it was examined for the change in BMI in the year preceding menarche and noted whether it was associated with age at menarche. There was a mean (SD) BMI increment of 1.6 (1.3) kg/m2 but the magnitude of this increment was not in correlation with age at menarche [p=0.4].

Discussion

In this study, age at menarche in Bengali girls [mean 12.3 (SD 0.8) years] was similar to that of affluent Indian girls (mean age 12.6 years)8 but somewhat lower than that of urban Punjabi girls [median 13.2 (SD 1.2)]9. The decrease in menarcheal age noted is likely to be due to the variability in the methods of recording, but may also reflect a secular downward trend over a decade. This however needs confirmation in a larger population – the present study was not powered to test for this.

The age at menarche in girls in a lower socio-economic group in western India was considerably delayed (mean 15.4 years)3. This extreme delay in menarche is likely to reflect malnutrition and its effects on delaying puberty. Children in this group are therefore not comparable to girls in the cohort. In another study of Bengali girls within a low socio-economic group, menarcheal age was found to be lower in heavier individuals10 but this observation did not take into account either the stage of puberty or BMI SDS. BMI is age dependent in adolescence6,11 and menarche is dependent on the stage of puberty12. In the present cohort, we therefore controlled for the variability of BMI with age (by SD scores), and adjusted for the stage of puberty (in univariate analysis). The results did not suggest that BMI at menarche was significantly associated with age at menarche. This is in keeping with the results of the Fels longitudinal study, which examined six 10-year birth cohorts and concluded that population shifts in BMI and the timing of menarche are largely independent4.
The results in the present study are in contradiction with that of a large study, albeit in a different population, which found an inverse correlation with BMI1. However the latter and other observations10,13,14 were retrospective in nature and therefore unlikely to represent truly the BMI at the time of menarche. As BMI increases through adolescence6,11, it would be an incorrect assumption to extrapolate BMI data in adult life back to the time of menarche. The present study, although relatively small in number, is strengthened by its prospective longitudinal design, measurements of BMI at the time of menarche and the use of SD scores.
Age at menarche can be influenced by a number of factors such as maternal age at menarche15, formula feeding in infancy16 and birth weight17, which were not taken into account. We did not enquire about premature thelarche as a cause of early menarche18, although it can be safely assumed that the numbers of such girls were likely to be reassuringly low.
The group of children had relatively low BMI SDS, with a maximum of 1.2 SD. It is possible that the inclusion of heavier girls would have altered the present findings. However it would be unrealistic to expect BMI values in the Bengali population to be similar to that of western standards6. We have used BMI as an index of body fat mass. Although a reasonable tool to estimate adiposity in the community19, it is only a ratio and therefore an abstract index. One must therefore exert a degree of caution in its interpretation. Nevertheless the use of BMI SDS as a surrogate marker of body fat is a more refined method than simple weight measurement, which is liable to be influenced by stature.
The age at menarche in healthy Bengali girls of middle income families in this study was similar to that of other Indian observations. BMI at menarche or in the year preceding menarche did not influence the menarcheal age in such girls.

Acknowledgment

We wish to acknowledge with great respect, Swami Sarvolokananda Maharaj, Secretary, Ramakrishna Mission Seva Prathishthan, Vivekananda Institute of Medical Sciences, Kolkata for providing a generous grant and logistic support to carry on the study. Our sincere thanks to Dr AK Ray Chowdhury, Officer-in-Charge, Dr AK Mukherjee, Dr P Ganguli and all the technical staff of ROHC(E) for their co-operation and help for the study. The paediatricians and residents (during the study) in the department of paediatrics, Ramakrishna Mission Seva Pratisthan, who volunteered to co-operate with this study, deserve special mention.

References

1 Mandel D, Zimlichman E, Mimouni FB, Grotto I, Kreiss Y — Age at menarche and body mass index: a population study. J Pediatr Endocrinol Metab 2004; 17: 1507-10.
2 Lin-Su K, Vogiatzi MG, New MI — Body mass index and age at menarche in an adolescent clinic population. Clin Pediatr (Phila) 2002; 41: 501-7.
3 Rao S, Joshi S, Kanade A — Height velocity, body fat and menarcheal age of Indian girls. Indian Pediatr 1998; 35: 619-28..

4 Demerath EW, Towne B, Chumlea WC, Sun SS, Czerwinski SA, Remsberg KE, et al — Recent decline in age at menarche: the Fels longitudinal study. Am J Hum Biol 2004; 16: 453-7.
5 Ramachandran A, Snehalatha C, Vinitha R, Thayyil M, Kumar CK, Sheeba L, et al — Prevalence of overweight in urban Indian adolescent school children. Diabetes Res Clin Pract 2002; 57: 185-90.
6 Banerjee I, Ghia N, Sayed HN, Mukherjee D — Body mass index in Bengali adolescents. Indian Pediatr 2005; 42: 262-7.
7 Tanner JM — Growth at Adolescence. 2nd ed. Oxford: Blackwell Scientific, 1962.
8 Agarwal DK, Agarwal KN, Upadhyay SK, Mittal R, Prakash R, Rai S — Physical and sexual growth pattern of affluent Indian children from 5 to 18 years of age. Indian Pediatr 1992; 29: 1203-82.
9 Sharma K — Age at menarche in northwest Indian females and a review of Indian data. Ann Hum Biol 1990; 17: 159-62.
10 Bharati S, Bharati P — Relationship between menarcheal age and nutritional anthropometry in urban girls of the Howrah District, West Bengal, India. Anthropol Anz 1998; 56: 57-61.
11 Agarwal KN, Saxena A, Bansal AK, Agarwal DK — Physical growth assessment in adolescence. Indian Pediatr 2001; 38: 1217-35.
12 Marshall WA, Tanner JM — Variations in pattern of pubertal changes in girls. Arch Dis Child 1969; 44: 291-303.
13 Vitalle MS, Tomioka CY, Juliano Y, Amancio OM — Body mass index, pubertal development and their relationship with menarche. Rev Assoc Med Bras 2003; 49: 429-33.
14 Fujii K, Demura S — Relationship between change in BMI with age and delayed menarche in female athletes. J Physiol Anthropol Appl Human Sci 2003; 22: 97-104.
15 Palmert MR, Boepple PA — Variation in the timing of puberty: clinical spectrum and genetic investigation. J Clin Endocrinol Metab 2001; 86: 2364-8.
16 Novotny R, Daida YG, Grove JS, Acharya S, Vogt TM — Formula feeding in infancy is associated with adolescent body fat and earlier menarche. Cell Mol Biol (Noisy-le-grand) 2003; 49: 1289-93.
17 Ibanez L, Valls C, Miro E, Marcos MV, de Zegher F — Early menarche and subclinical ovarian hyperandrogenism in girls with reduced adult height after low birth weight. J Pediatr Endocrinol Metab 2002; 15: 431-3.
18 Biro FM, Lucky AW, Simbartl LA, Barton BA, Daniels SR, Striegel-Moore R, et al — Pubertal maturation in girls and the relationship to anthropometric changes: pathways through puberty. J Pediatr 2003; 142: 643-6.
19 Pietrobelli A, Faith MS, Allison DB, Gallagher D, Chiumello G, Heymsfield SB — Body mass index as a measure of adiposity among children and adolescents: a validation. J Pediatr 1998; 132: 204-10.

.

Home
Copyright : JIMA