Nutrition and Health Sciences, Department of


Document Type


Date of this Version



Published in Bone 88 (2016), pp 101–112. doi:10.1016/j.bone.2016.01.030


Copyright © 2016 Elsevier Inc. Used by permission.


Menstrual status, both past and current, has been established as an important determinant of bone mineral density (BMD) in young exercising women. However, little is known regarding the association between the cumulative effect of menstrual status and indices of bone health beyond BMD, such as bone geometry and estimated bone strength.

Purpose: This study explores the association between cumulative menstrual status and indices of bone health assessed using dualenergy x-ray absorptiometry (DXA), including femoral neck geometry and strength and areal BMD (aBMD), in exercising women.

Methods: 101 exercising women (22.0 ± 0.4 years, BMI 21.0 ± 0.2 kg/m2, 520±40 min/week of self-reported exercise) participated in this cross-sectional study. Women were divided into three groups as follows based on their self-reported current and past menstrual status: 1) current and past regular menstrual cycles (C + P-R) (n=23), 2) current and past irregular menstrual cycles (C+P-IR) (n=56), 3) and current or past irregular cycles (C/P-RIR) (n=22). Current menstrual status was confirmed using daily urinary metabolites of reproductive hormones. DXA was used to assess estimates of femoral neck geometry and strength from hip strength analysis (HSA), aBMD, and body composition. Cross-sectional moment of inertia (CSMI), cross-sectional area (CSA), strength index (SI), diameter, and section modulus (Z) were calculated at the femoral neck. Low CSMI, CSA, SI, diameter, and Z were operationally defined as values below the median. Areal BMD (g/cm2) and Z-scores were determined at the lumbar spine, femoral neck, and total hip. Low BMD was defined as a Z-score < −1.0. Chi-square tests and multivariable logistic regression were performed to compare the prevalence and determine the odds, respectively, of low bone geometry, strength, and aBMD among groups.

Results: Cumulative menstrual status was identified as a significant predictor of low femoral neck CSMI (p = 0.005), CSA (p ≤ 0.024), and diameter (p = 0.042) after controlling for confounding variables. C + P-IR or C/PRIR were four to eight times more likely to exhibit low femoral neck CSMI or CSA when compared with C + PR. Lumbar spine aBMD and Z-score were lower in C + P-IR when compared with C + P-R (p ≤ 0.003). A significant association between menstrual group and low aBMD was observed at the lumbar spine (p = 0.006) but not at the femoral neck or total hip (p > 0.05). However, after controlling for confounding variables, cumulative menstrual status was not a significant predictor of low aBMD.

Conclusion: In exercising women, the cumulative effect of current and past menstrual irregularity appears to be an important predictor of lower estimates of femoral neck geometry, as observed by smaller CSMI and CSA, which may serve as an another means, beyond BMD, by which menstrual irregularity compromises bone strength. As such, evaluation of both current and past menstrual status is recommended to determine potential risk for relatively small bone geometry at the femoral neck.