Peak bone mass and bone mineral density correlates for 9 to 24 year-old Mexican women, using corrected BMD

Pico mineral óseo y factores asociados a la densidad mineral ósea en mujeres mexicanas de 9 a 24 años de edad usando densidad mineral ósea corregida

Gabriela Torres-Mejía, MD, PhD^I; Rubén Guzmán Pineda, MSc^II; Martha María Téllez-Rojo, MSc, DrSc^III; Eduardo Lazcano-Ponce, MD, DrSc^I

^ICentro de Investigación en Salud Poblacional, Instituto Nacional de Salud Pública, Cuernavaca, Morelos, México
^IIHospital General Regional UMF No. 1, Instituto Mexicano del Seguro Social (IMSS), Cuernavaca, Mor ]]> ^IIIDirección de Estadística, Centro de Investigación en Evaluación y Encuestas, Instituto Nacional de Salud Pública, Cuernavaca, Mor

ABSTRACT

OBJECTIVE: To determine the age of peak bone mass (PBM) in Mexican women and factors associated with both BMDa and corrected BMD (BMDcorr) at the femoral neck and the spine (L2-L4).
MATERIAL AND METHODS: Data on 461 women between 9 and 24 years old was used. An interview was performed and height and weight were measured. BMDa was measured by a densitometer and BMDcorr by the method proposed by Kröger et al. (1992).
RESULTS: PBM at the spine (L2-L4) was observed later than at the femoral neck. Both BMDa and BMDcorr at the lumbar spine correlate with age, socio-economic status, body fat percentage and height. BMDa at the femoral neck correlates with overweight and obesity, body fat percentage, height and moderate physical activity; the same variables were associated with BMDcorr except for height.
CONCLUSIONS: The method proposed by Kröger et al. was more precise at the femoral neck than at the spine.

Key words: corrected bone mineral density; peak bone mass; Mexican women

RESUMEN

Palabras clave: densidad mineral ósea corregida; pico mineral óseo; mujeres mexicanas

It has been reported that the prevalence of osteoporosis in Mexican postmenopausal women over the age of 50 is 20%, and as in other countries, this prevalence is higher for women than for men.^1-3 It has been suggested that the prevention of this illness might begin in early childhood due to the fact that, according to some authors,^4-5 the maximum mineralization of bone tissue –peak bone mass (PBM)– occurs between the ages of 18 and 30. In addition, it has been suggested that approximately 45% of bone mass is established during adolescence.^6,7 These observations have led several authors to assess correlates of bone mass in adolescents. Some characteristics that have been studied are age, gender, height, body mass index (BMI), calcium consumption, and physical activity, among others.^8-12

For most of the studies, the measurement of bone mass has been done in terms of areal bone mineral density (BMDa), however this measurement is confounded by the size of the bone. BMDa is defined as the bone mineral content (BMC) in grams per bone mass measured in square centimeters (g/cm²). This measurement has been widely criticized by several authors because it only permits the approximate estimation of bone resistance. Since this measurement does not take into consideration the anteroposterior diameter of the bone, and is therefore influenced by size, it shows a higher density effect for larger bones than for smaller ones.¹³

Volumetric bone mineral density (BMDv) has been proposed as a more specific measurement of bone resistance due to the fact that it takes size into consideration. BMDv is defined as BMC in grams per bone volume in cubic centimeters (g/cm³). This measurement may be precisely obtained by means of peripheral quantitative computed tomography, and in a less precise way, by certain densitometers that use dual energy x-ray absorption (DEXA "Dual energy x-ray absorptiometry").

The objective of this study was to assess the average age at which PBM was reached and the correlates associated with BMDa and BMDcorr at the femoral neck and lumbar spine (L2-L4) using the method proposed by Kröger, et al.¹⁴

Methods

Study population

The information that was used for this work has been previously described.¹⁷ For the present study, 9 to 24-year-old women who participated in the PBM study and whose densitometries were recovered were included.

To summarize, the information was provided by a cross-sectional study that was performed on 461 Mexican women who were randomly selected from different age strata in 1999. The participants were students from elementary, junior high, high school and college levels. The study included clinically healthy women between 9 to 24 years of age who were born in the state of Morelos, who signed a consent form to participate in the study, and in the case of minors, those whose parents signed a consent form. The study was approved by the National Institute of Public Health IRB.

Women who reported previous fractures and chronic degenerative illnesses were excluded from the study. The population for the study was essentially urban (based on governmental statistics). Socio-economic status (SES) was generated using principal components analysis.

All women were interviewed. The questionnaire included socio-demographic characteristics and gynecological history (e.g. age at menarche, age at menopause), frequency of food consumption¹⁸ (e.g. calories/24 hrs, vitamin D, calcium, dairy products and tortillas), and time spent doing physical activity, watching TV and sleeping.¹⁹ Height and weight were measured using a stadimeter and a floor scale; women were dressed in light clothes and did not wear shoes.

A certified technician measured BMDa at the femoral neck and the lumbar spine (L2 to L4) using a DEXA Hologic type series A densitometer. Finally, for this particular study, BMDcorr at the femoral neck and spine (L2-L4) was estimated using the method proposed by Kröger et al.¹⁴

In order to calculate BMDcorr, volumes at the femoral neck and vertebral bodies were estimated using the method proposed by Kröger, et al.¹⁴ From the images obtained by the densitometer, height (h*) and width (b*) of each vertebra (L2, L3, L4) and femoral neck were measured using a digitalized vernier (Mitutoyo). To calculate real height (h) and width (b), a scale factor was obtained using the area (A) obtained by the densitometer, as follows:

A= (α b*) (α h*)

A= α² (b* h*)

α= √ A/(b* h*)

Where: A= real coronal area at the vertebra or femoral neck

α= constant

b*= width of the vertebra measured with a vernier

h*= height of the vertebra measured with a vernier

Then, to obtain the real width (b) and height (h), b* and h* were multiplied by the constant a:

h= αh*

Subsequently, the cylinder volume formula was applied to obtain the volume as follows:

Femoral Neck

Volume= π x (A² / 4h)

Where A= real coronal area at the femoral neck

h= real height at the neck

The BMDcorr at the femoral neck was measured as follows:

BMDcorr= BMC/volume

Vertebral bodies

Where b= width for each of the vertebra

A= coronal area of each vertebra

The BMDcorr from each vertebra was measured as follows:

BMDcorr= BMC/volume

Once the volume for each vertebra was calculated, we proceeded to calculate the BMDcorr from the lumbar column (L2 , L3 and L4) as follows:

Lean body mass and body fat mass (%) were obtained from the densitometer.

Calculation of the body mass index

Body mass index is defined as the weight in kilograms (kg) divided by the square of the height in meters (m²), which is expressed in kg/m². For women between 9 to 17 years of age, weight status was defined, according to CDC and WHO recommendations, by using age- and gender-specific BMI percentiles from the revised 2000 Center for Health Statistics growth charts for the United States.²⁰ For women 18 and over, a BMI of 25 for overweight and 30 for obesity were used. Women under 18 years were classified into three categories according to their BMI, based on percentiles: adequate weight (<85), overweight (>85, <95) and obesity (>95).

Initially, the study population was described. To determine the age at which PBM was reached, spine and femoral neck BMDa and BMDcorr were estimated by using two-year strata. Multiple lineal regressions were used to estimate the association between all characteristics and BMDa and BMDcorr. In addition, generalized additive models were used in order to graphically observe the association between continuous predictors and the outcomes.²¹ The statistical analysis was done with the use of STATA statistical packages version 10.

The intraclass correlation coefficient was used to assess intra- and inter-observer reliability.

Results

Reliability

The reliability study was performed for 138 measurements obtained at the femoral neck and at each vertebra. The intraclass correlation for the femoral neck height was 0.90 (CI 95% 0.86-0.95) and was 0.80 (CI 95% 0.72-0.89) for the base. The intraclass correlation for the height of each vertebra (L2-L4) varied from 0.73 (CI 95% 0.62-0.84) for L4 to 0.95 (CI 95% 0.93-0.97) for L3. The intraclass correlation corresponding to the base varied from 0.70 (CI 95% 0.58-0.82) for L2 to 0.94 (CI 95% 0.92-0.97) for L4.

Study population characteristics

Out of 461 women, 21 were excluded from the analysis due to the fact that they had no menarche, and 41 were excluded because information from the questionnaire on food frequency was not complete. Finally, data on 399 women were analyzed. The mean age was 19.6 years (range 10.2 to 25.9 years). The majority of women belonged to middle SES (65.7%). The mean age at menarche was 12 years (range 9 to 15 years). It was observed that 79% of women had normal weight at the time the study was performed, 17.3% were overweight and 3.3% were obese. The mean intake of calcium was 507 mg/day, the mean time performing moderate physical activity was 0.6 hrs/day and vigorous activity was 0.5 hrs/day. Women reported watching TV a mean of 3 hours/day and sleeping 8.6 hours/day. Results for only 175 women were obtained for calcium and vitamin D.

BMDa, BMDcorr and BMC at the spine and femoral neck had a normal distribution. The mean BMDa at the spine was 0.954 g/cm², while for BMDcorr it was 0.283 g/cm³. The mean values at the femoral neck were 0.826 g/cm² and 0.348 gr/cm³ for BMDa and BMDcorr, respectively.

Univariate analysis

Spine

Regarding the correlates of BMDa at the spine, the associated variables observed were: age (R²= 0.27), years post menarche (R²= 0.24), weight (R²=.33), BMI (R²= 0.22), height (R²= 0.12), body fat percentage (R²⁼ 0.13) and lean body mass percentage (R²= 0.13). The results were similar for the BMDcorr except for height, where R² was 0.002. In both analyses, age at menarche was positively associated (p= 0.005; p= 0.004, respectively). Calcium consumption was inversely associated with both BMDa and BMDcorr, although the association was borderline for BMDa (p= 0.06) and statistically significant for BMDcorr (p= 0.04). Likewise, there was a negative association between calorie consumption and both densities (p= 0.007; p= 0.006, respectively). Regarding physical activity, there was no statistically significant association.

Femoral neck

Age was positively associated with BMDa and BMDcorr at the femoral neck. In both analyses, the association was statistically significant (p<0.001), however the strength of the association was negligible (R²= 0.07). Age at menarche and diet were not statistically significantly associated with either density. Regarding height, it was statistically significantly associated with BMDa (p<0.001), however with BMDcorr there was no association (p= 0.734). Regarding body composition, there was a positive and statistically significant association between total body fat percentage and both densities (p<0.001; p<0.001, respectively). In contrast, there was a negative association between lean mass percentage and both densities (p<0.001; p<0.001, respectively). Finally, total physical activity was positively associated with BMDa and BMDcorr (p= 0.04; p= 0.03) and moderate physical activity was only associated with BMDcorr (p= 0.081).

Multivariate analysis

Results for the multiple lineal regression analyses are shown in Table I

Characteristics associated with BMDa and BMDcorr at the spine (L2-L4)

There was a positive association between BMDa and age (β=12.7 mg/cm²; CI 95% 9.9, 15.4 mg/cm²), body fat percentage (β= 6.9 mg/cm²; CI 95% 4.7, 15.4 mg/cm²) and height (β= 5.7% mg/cm²; CI 95% 4.1, 7.2 mg/cm²). Weight status (normal/overweight and obesity) was not associated with BMDa and, therefore, it was not included in the model. The model was adjusted for calories and SES; all variables explained 42% of the variability (Table I) (Figure 1). A similar figure was observed for BMDcorr (Table I) (Figure 2).

Characteristics associated with BMDa and BMDcorr at the femoral neck

There was a positive association between BMDa and age (β= 5.0 mg/cm²; CI 95% 2.1, 7.9 mg/cm²), weight status (β= 52 mg/cm²; CI 95% 19.9, 84.2 mg/cm²), height (β= 3.9 mg/cm²; CI 95% 2.2, 5.6 mg/cm²), and body fat percentage (β= 5.8 mg/cm²; CI 95% 2.8, 8.7 mg/cm²). Moderate physical activity was not statistically significantly associated with BMDa (β= 35.2 mg/cm²; CI 95% -2.3, 72.7 mg/cm²). The model was adjusted for calories and all variables explained 26% of the variability (Table1) (Figure 3). A similar figure was observed for BMDcorr, and height was not included in the model (Table1) (Figure 4).

Discussion

In the present study, it was observed that PBM for the spine (L2-L4) was reached between 24 and 25 years, while for the femoral neck the peak was reached at a lower age (between 18 and 19 years). Similar results have been observed before.⁵

It has been reported that the association between age and BMD begins in puberty^22,23 due to an increase of growth hormones and sex steroid blood levels which have a positive effect on bone mineralization.²⁴ In the present study, PBM was reached at the same age whether using BMDa or BMDcorr. This suggests that BMD is not associated with the size of the bones as other studies have suggested. ^{4,5,14,16,22,23}

The correction proposed by Kroger et al. was more precise for the femoral neck than for the spine, probably due to the femoral neck being more similar to a cylinder than vertebrae. In the present study, the correlates of BMDa and BMDcorr at the spine were the same: age, body fat percentage, height, and calorie consumption. The effect of height decreased for BMDcorr, but did not disappear. In contrast, at the neck, height was only associated with BMDa and not with BMDcorr.

In this study, age at menarche was not included in the model due to the high correlation of this variable with age (r= 0.956, <0.00l). However, other studies have found a positive association.^5,14,16,25

According to the hypotheses formulated by Glauber et al,²⁶ if the relationship between obesity and BMD is due to a mechanical effect of load, the effect of weight should be higher for bones which support a larger load (e.g. femoral neck). Our results are consistent with this hypothesis because weight status (normal/overweight and obesity) was associated with BMDa and BMDcorr at the femoral neck but not with bone density at the spine. Our results support those who have suggested that high BMI is associated with a decrease in femoral or lumbar spine fractures due to osteoporosis.^13,27 Results from other studies are similar to what we found.^11,15

Our results are consistent with those who have found a positive association between adiposity and BMDa.^28,29 If the effect of adipose tissue on BMD is due to metabolic and hormonal activity, the effect of body fat percentage should be a determinant of BMD for any bone. In our study, body fat percentage was associated with both BMDa and BMDcorr at the femoral neck and the spine. Lazcano et al. showed an inverse association between the percentage of fat and BMD at the spine; when analyzing these women, however, they used saturated models for their analyses.

Our results showed an inverse association between SES and BMDa and BMDcorr at the spine (data not shown). This is probably due to the fact that women with higher SES were thinner than those with lower SES. Some studies have shown that women with higher SES consume fewer calories and are thinner than women from a high socioeconomic level.³⁰ Furthermore, more fractures have been observed among slim women and those with a high socio-economic status.^2,3 SES was not associated with BMD at the femoral neck.

In the present study, neither calcium consumption nor vitamin D consumption showed a statistically significant association with either BMDa or BMDcorr at the spine or the femoral neck. Our results are consistent with other studies,^4,5,14,16 although it is well known that an adequate consumption of calcium is necessary for bone growth and probably for reaching PBM at the appropriate time, thus reducing the risk of osteoporosis.³⁷

This is the first study to report BMDcorr in Mexican women, though it was faced with the inherent limitations of cross-sectional studies.

Conclusions

In order to lower costs and lower the dose of radiation in developing countries with little resources, BMDcorr measured using the method proposed by Kroger et al (1992) is an acceptable option to correct for the size of the bone.

References

1. NIH Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 1993; 94:646-650.        [ Links ]

2. Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996; 312:1254-1259.        [ Links ]

3. Woodhouse A, Black DM. BMD at various sites for de prediction of hip fracture: a meta-analysis. J Bone Miner Res 2000; 15:S145.        [ Links ]

4. Boot AM, de Ridder MA, Pols HA, Krenning EP, de Muinck Keizer-Schrama SM. Bone mineral density in children and adolescents: relation to puberty, calcium intake, and physical activity. J Clin Endocrinol Metab 1997; 82:57-62.        [ Links ]

5. Bachrach LK, Hastie T, Wang M-C, Narisimhan B, Marcus R. Bone mineral acquisition in healthy Asian, Hispanic, Black, and Caucasian youth: A longitudinal study. J Clin Endocrinol Metab 1999; 84:4702-4712.        [ Links ]

6. Matkovic V, Jelic T, Wardlaw GM, Ilich JZ, Goel PK, Wright JK, Andon MB, Smith KT, Heaney RP. Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. Inference from a cross-sectional model. J Clin Invest 1994; 93:799-808.        [ Links ]

7. Ott SM. Attainment of peak bone mass. J Clin Endocrinol Metab 1990; 71:1082A-1082C.        [ Links ]

8. Clark P, de la Pena F, Gomez Garcia F, Orozco JA, Tugwell P. Risk factors for osteoporotic hip fractures in Mexicans. Arch Med Res 1998; 29:253-257.        [ Links ]

9. Murillo-Uribe A, Carranza-Lira S, Martinez-Trejo NA, Santos Gonzalez JE. Epidemiologic variables in postmenopausal women. Ginecol Obstet Mex. 1999;67:478-483.        [ Links ]

10. Deleze M, Cons-Molina F, Villa AR, Morales-Torres J, Gonzalez-Gonzalez JG, Calva JJ, Murillo A, Briceno A, Orozco J, Morales-Franco G, Pena-Rios H, Guerrero-Yeo G, Aguirre E, Elizondo J. Geographic differences in bone mineral density of Mexican women. Osteoporos Int 2000; 11:562-569.        [ Links ]

11. Moyer-Mileur L, Xie B, Ball S, Bainbridge C, Stadler D, Jee WS. Predictors of bone mass by peripheral quantitative computed tomography in early adolescent girls. J Clin Densitom 2001; 4:313-323.        [ Links ]

12. Lazcano-Ponce E, Tamayo J, Cruz-Valdez A, Diaz R, Hernandez B, Del Cueto R, Hernandez-Avila M. Peak bone mineral area density and determinants among female aged 9 to 24 years in México. Osteoporos Int 2003; 14:539-547.        [ Links ]

13. Seeman E. Clinical review 137: Sexual dimorphism in skeletal size, density, and strength. J Clin Endocrinol Metab 2001; 86:4576-4584.        [ Links ]

14. Kroger H, Kotaniemi A, Vainio P, Alhava E. Bone densitometry of the spine and femur in children by dual-energy x-ray absorptiometry. Bone Miner 1992; 17:75-85.        [ Links ]

15. Kroger H, Vainio P, Nieminen J, Kotaniemi A. Comparison of different models for interpreting bone mineral density measurements using DXA and MRI technology. Bone 1995;17:157-159.        [ Links ]

16. Lu PW, Cowell CT, Lloyd-Jones SA, Briody JN, Howman-Giles R. Volumetric bone mineral density in normal subjects, aged 5-27 years. J Clin Endocrinol Metab 1996; 81:1586-1590.        [ Links ]

17. Lazcano-Ponce EC, Hernandez B, Cruz-Valdez A, Allen B, Diaz R, Hernandez C, Anaya R, Hernandez-Avila M. Chronic disease risk factors among healthy adolescents attending public schools in the state of Morelos, Mexico. Arch Med Res 2003; 34:222-236.        [ Links ]

18. Hernández-Avila M, Romieu I, Parra S, Hernández-Avila J, Madrigal H, Willett W. Validity and reproducibility of a food frequency questionnaire to assess dietary intake of women living in Mexico City. Salud Publica Mex 1998; 40:133-140.        [ Links ]

19. Hernandez B, Gortmaker SL, Colditz GA, Peterson KE, Laird NM, Parra Cabrera S. Association of obesity with physical activity, television programs and other forms of video viewing among children in Mexico City. Int J Obesity. 1999; 23:845-854.         [ Links ]

20. Kuczmarski RJ, Ogden CL, Guo SS, Grummer-Strawn LM, Flegal KM, Mei Z, Wei R, Curtin LR, Roche AF, Johnson CL. 2000 CDC Growth Charts for the United States: methods and development. Vital Health Stat 11 2002 May; 246:1-190.

21. Hastie, T.J. and Tibshiani R.J.(1990) Generalized Additive Models. London. Chapman & Hall. Pgs. 281,284.        [ Links ]

22. Gilsanz V, Roe TF, Mora S, et al. Changes in vertebral bone mineral density in blacks girls and white girls during childhood and puberty. N Engl J Med 1991; 325:1597-1600.        [ Links ]

23. Gilsanz V, Skaggs DL, Kovanlikaya A, Sayre J, Loro ML, Kaufman F, Korenman SG. Differential effect of race on the axial and appendicular skeletons of children. J Clin Endocrinol Metab 1998; 83:1420-1427.        [ Links ]

24. Frank GR. The role of strogen in pubertal skeletal physiology: epiphyseal maturation and mineralization of the skeleton. Acta Paediatr. 1995; 84:627-630.        [ Links ]

25. Loro ML, Sayre J, Roe TF, Goran MI, Kaufman FR, Gilsanz V. Early identification of children predisposed to low peak bone mass and osteoporosis later in life. J Clin Endocrinol Metab 2000; 85:3908-3918.        [ Links ]

26. Glauber HS, Vollmer WM, Nevitt MC, Ensrud KE, Orwoll ES. Body weight versus body fat distribution, adiposity, and frame size as predictors of bone density. J Clin Endocrinol Metab 1995;80:1118-1123.        [ Links ]

27. Aloia JF, Cohn SH, Vaswani A, Yeh JK, Yuen K, Ellis K. Risk factor for postmenopausal osteoporosis. Am J Med 1985; 78:95-100.        [ Links ]

28. Reid IR, Ames R, Evans MC, et al. Determinants of total body, and regional bone mineral density in normal postmenopausal women-a key role for fat mass. J Clin Endocrinol Metab 1992; 75:45-51.        [ Links ]

29. Leunissen RW, Stijnen T, Boot AM, Hokken-Koelega AC.Influence of birth size and body composition on bone mineral density in early adulthood. The PROGRAM-study. Clin Endocrinol (Oxf). 2008 Feb 18 (Epub ahead of print).        [ Links ]

30. Neville CE, Murray LJ, Boreham CAG, Gallagher AM, Twisk J, Robson PJ, Savage JM, Kemper HCG, Ralston SH, Smith GD. Relationship between physical activity and bode mineral status in young adults: The Northern Ireland Young Hearts Project. Bone 2002; 30:792-798.        [ Links ]

31. Reid IR, Legge M, Stapleton JP, Evans MC, Grey AB. Regular exercise dissociates fat mass and bone density in premenopausal women. J Clin Endocrinol Metab 1995; 80:1764-1768.        [ Links ]

32. Janz KF, Burns TL, Torner JC, Levy SM, Paulos R, Willing MC, Warren JJ. Physical activity and bone measures in young children: the Iowa bone development study. Pediatrics 2001; 107:1387-1393.        [ Links ]

33. McLean JA, Barr SI, Prior JC. Dietary restraint, exercise, and bone density in young women: are they related?. Med Sci Sports Exerc 2001; 33:1292-1296.        [ Links ]

34. Mackelvie KJ, McKay HA, Khan KM, Crocker PR. Lifestyle risk factors for osteoporosis in Asian and Caucasian girls. Med Sci Sports Exerc 2001; 33:1818-1824.        [ Links ]

35. Barbeau P, Johnson MH, Howe CA, Allison J, Davis CL, Gutin B, Lemmon CR. Ten months of exercise improves general and visceral adiposity, bone, and fitness in black girls. Obesity 2007; 15:2077-2085.        [ Links ]

36. Ortega FB, Ruiz JR, Castillo MJ, Sjöström M. Physical fitness in childhood and adolescence: a powerful marker of health. Int J Obes 2008; 32:1-11.        [ Links ]

37. Cadogan J, Eastell R, Jones N, Barker ME. Milk intake and bone mineral acquisition in adolescent girls: randomised, controlled intervention trial. BMJ 1997; 315:1255-1260.        [ Links ]

Received on: October 10, 2008
Accepted on: December 11, 2008