Comparison of the Right and Left Femur Bone Mineral Densities in Postmenopausal Women

It has been documented that bone mineral density (BMD) in women reaches and maintains a maximum bone density level, and then rapidly decreases with menopause. As estrogen is involved in both bone formation and the prevention of bone resorption, postmenopausal osteoporosis occurs with the cessation of ovarian function as age progresses. BMD measurements by dual-energy radiography absorptiometry (DXA) are widely used to diagnose osteoporosis and predict individual fracture risk [1]. Bone loss is also influenced by age, weight, body mass index (BMI), comorbid diseases or medications [2]. In postmenopausal women, menopausal hormone therapy is known to prevent osteoporosis. When hormone therapy is stopped, further evaluation of BMD levels is recommended to determine the risk of osteoporosis [3, 4]. Low vitamin D levels are also associated with BMD [5].

The 2019 the International Society for Clinical Densitometry (ISCD) guidelines state that BMD should be measured in the posteroanterior spine and hip in all patients. Forearm BMD measurement is optional in some circumstances, such as hyperparathyroidism or very obese patients. The guideline states that BMD may be measured at either hip, with femoral neck and total proximal femur being the regions of interest (ROI). It recommends against the use of other hip ROI, such as Ward’s area and greater trochanter [6]. Bonnick et al. [7] also suggested that analysis of both the right and left proximal femurs may have no benefit. Further studies have reported that right and left hip BMD shows a significant difference; thus, scanning both hips can identify more women with osteoporosis [8].

The present study aimed to compare the BMD of right and left proximal femurs in postmenopausal women and to determine if measuring BMD of both femurs would be more accurate. Furthermore, this study attempted to determine other variables that affect the differences in right and left hip BMD.

During the study period, 343 menopausal women underwent BMD measurements using DXA in both the hips and the spine. Baseline patient characteristics are described in Table 1. The average age of the participants was 62 ± 9.7 years, and the menopausal age was 50 ± 4.5 years. In the study population, 119 patients received hormone therapy after menopause, 120 patients received medication for osteoporosis, and 183 patients received calcium and vitamin D supplements.

Table 1
Basal characteristics of postmeopausal women who underwent bone mineral density measurements using dual-energy x-ray absorptiometry in both hips and spine

• Click for larger image
• Click for full table
• Download as Excel file

Comparisons between the right and left femur BMD using DXA was performed using a paired t test (Table 2). Right and left total femur BMD (0.881 ± 0.122 g/cm² and 0.889 ± 0.121 g/cm², respectively) showed a statistically significant difference (P = 0.003). The femoral neck and trochanter areas however did not show any significant differences in the right and left BMD.

Table 2
Comparison of bone mineral density between right and left proximal femur

• Click for larger image
• Click for full table
• Download as Excel file

Subgroup analysis was conducted based on the medication usage. Diagnosis of osteoporosis and osteopenia in postmenopausal women was made using the T-score if there was no previous hip or vertebral fracture history. A spinal or hip T-score of -2.5 or lower was defined as osteoporosis, and a score between -1.0 and -2.5 was defined as osteopenia. Among 343 patients, patient with osteoporosis (n = 159) showed significant right and left BMD differences in trochanter and femur total area (P = 0.041 and 0.011, respectively). In osteopenia and normal BMD group, no statistically significant differences were observed (Table 3).

Table 3
Comparison of bone mineral density between right and left proximal femur in osteoporosis, osteopenia, normal group

• Click for larger image
• Click for full table
• Download as Excel file

Osteoporosis treatment was prescribed to 118 patients. The type of medications prescribed were bisphosphonate, a selective estrogen receptor modulator (SERM), and denosumab for 97, 21, and 2 patients, respectively. The denosumab group was excluded from this subgroup analysis due to the small population size. In the bisphosphonate group, the BMD of the right and left trochanters area and femur total area showed statistically significant differences (P = 0.038 and 0.024, respectively). In SERM group, no significant differences were observed between right and left hip BMD (Table 4).

Table 4
Comparison of bone mineral density between right and left proximal femur according to osteoporosis treatment

• Click for larger image
• Click for full table
• Download as Excel file

Among 343 postmenopausal women, 119 patients received menopausal hormone therapy (e.g., livial^®, angeliq^®, climen^®, progynova^®). In hormone therapy group, no significant difference between femur neck BMD was observed. In group without menopausal hormone therapy, right and left femur total area BMD showed statistically significant difference (P = 0.013) (Table 5). However, patients who received menopausal hormone therapy showed a statistically higher BMD than those who did not (P = 0.006 in femur total area BMD). Therefore, this result supports the basic knowledge that menopausal hormone therapy prevents BMD decline, and the result from Table 2 that tells BMD difference between contralateral hips get greater in low BMD patients.

Table 5
Comparison of bone mineral density between right and left proximal femur according to use of menopausal hormone therapy after menopause

• Click for larger image
• Click for full table
• Download as Excel file

Another subgroup analysis was performed for postmenopausal women based on the prescription of calcium and vitamin D combination (calcium carbonate/cholecalciferol, Dicamax D^®). Patients who were not prescribed calcium and vitamin D supplements showed statistically significant right and left BMD differences in the trochanter and femur total area. In comparing right and left BMD using paired t test, the P value was 0.036 and 0.001 for the trochanter region and femur total area, respectively (Table 6). There was no statistically significant difference in BMD when comparing those who received calcium and vitamin D supplement with others.

Table 6
Comparison of bone mineral density between right and left proximal femur according to calcium and vitamin D supplement

• Click for larger image
• Click for full table
• Download as Excel file

We investigated whether the diagnosis of osteoporosis differed when BMD was measured on only one side (Table 7). Assuming that unilateral femur BMD and spine BMD were measured, the lowest T-score among the femur neck, trochanter, femur total region, and spine region was used to diagnose osteopenia or osteoporosis. A total of 285 patients (83.1%) showed the same diagnosis. However, 48 patients (14.0%) might have been misdiagnosed with osteopenia, and 10 women (3.0%) might have been considered normal if only unilateral hip BMD was measured.

Table 7
Concordance between the diagnosis of osteoporosis and osteopenia using unilateral femur T-score

• Click for larger image
• Click for full table
• Download as Excel file

The correlation coefficients between the right and left femoral neck, trochanter area, total femur BMD, and spine BMD were evaluated (Table 8). Overall, right and left BMD across the same regions showed high correlation (neck BMD R² = 0.925, trochanter BMD R² = 0.919, neck total BMD R² = 0.923), with spine total BMD, every hip area showing a relatively low correlation.

Table 8
Correlation coefficient between bone mineral density of different hip and spine region^*

• Click for larger image
• Click for full table
• Download as Excel file

The aim of this study was to verify whether there is a difference in the BMD of both femurs in postmenopausal women. This study included 343 postmenopausal women who underwent BMD measurement using DXA in both the hips and spines. In postmenopausal women, the BMD of the bilateral hip can be different, especially when the femur total area is assessed. In addition, the BMD difference of right and left femur increased more in the osteoporosis group than in the normal BMD group. In the osteoporosis group, the trochanter area and total neck BMD showed statistically significant differences. On the other hand, in the normal group, neither the femoral neck, trochanter, nor the femur total BMD showed any meaningful differences. Therefore, it is crucial to assess both the right and left femoral BMD in case of suspected osteoporosis.

Several studies have debated the efficacy of analyzing BMDs in both the hips. Hwang et al. [9] analyzed the differences in BMDs of both the hips in 384 postmenopausal women. The study showed differences in the BMDs of the femoral neck and trochanter area (P < 0.05) and exhibited the lowest T-scores for both hips (P < 0.05). Taddei et al. [10] used computed tomography (CT)-derived finite element models to compare mineral density of contralateral proximal femurs in 198 postmenopausal women. The right and left femurs had statistically significant differences, with the right hip being 4% stronger than the left on an average. Other studies demonstrated that even though statistically significant differences exist between the BMDs of the right and left hip, only 2.2% of patients show discrepancies in osteoporosis diagnosis, which is quite lower compared to our study [11]. The study of Bonnick et al. [7] also demonstrated that possibility of misclassification was low, even though their result showed BMD differences in right and left trochanter and ward’s area. This study did not include femur total BMD in analysis, which is important in diagnosing osteoporosis. In our study, femur total area BMD showed significant differences, so this could have made the different conclusions.

The strength of our study is that it is the first to reveal differences in the BMDs of contralateral femur, especially in osteoporotic postmenopausal women. In the small group analysis, this study suggested that the BMDs of both hips showed more differences in the osteoporosis group, which is not well documented in other similar studies. In women with osteoporosis, we conducted an analysis based on the osteoporosis medications they used. Women who used SERM for osteoporosis treatment showed similar contralateral hip BMD, while the bisphosphonate group and no-treatment group showed statistically different contralateral hip BMD. However, due to the small sample size of the SERM group and the fact that follow-up was not conducted after using osteoporosis medication, these data are difficult to interpret.

Postmenopausal hormone therapy seemed to decrease the difference between the BMD of the right and left femur. However, the data showed that postmenopausal women who received hormone therapy had an overall higher BMD. This is consistent with the findings that patients with osteoporosis or low BMD have differences in their right and left hip BMDs, which is described in Table 3. Vitamin C and calcium supplementation did not increase the hip BMD significantly, but decreased the differences between the right and left hip BMD.

Overall, the left femoral BMD was higher than that of the right. Interestingly, in the osteoporosis group, the left femoral BMD was higher than the right femoral BMD, whereas in the normal group, the right was higher than the left, although the difference was not statistically significant. In this study, the leg dominance of the participants was not included, but most people had right leg dominance [12]. In young sports athletes, differences in BMD were observed based on the intensity of exercise. Higher-impact activities were associated with higher BMD in the non-dominant hip, while lower-intensity exercises showed a tendency for higher bone density in the dominant hip [13]. A previous study by Yang et al. [14] reported that leg dominance had little effect on proximal femur BMD. However, the population of the study included only normal Chinese women, and whether they have osteoporosis or not was unclear. In other paper, they compared dominant and non-dominant femur BMD in soccer players and non-athlete group. Soccer players tend to had higher BMD in non-dominant leg, which can be explained by more frequent engagement of non-dominant leg in take-off, landing and stance of shooting. In non-athlete group, no significant difference was observed [15]. Although the study population of our study mostly would be non-atheletes, this can explain why left femur BMD was higher than right.

The primary limitation of this study is attributed to the retrospective nature of the study. If the individual patients’ BMD had been followed up in a longitudinal study type, we could have observed BMD changes based on the osteoporotic change, hormonal therapy usage, and vitamin D supplement usage. In addition, after assessing each patient’s dominant leg, verifying the change in hip BMD after menopause can be useful. Furthermore, this study did not consider patients with unilateral hip disorders, such as arthritis, hemiplegia, or prior fracture. Low BMI is also an important factor which affects BMD, but the participants of our study had few patients with low BMI, less than 18.5, making difficult to analyze according to BMI. Considering the potential impact of various factors on bone mineral density [16], it is imperative to conduct a more extensive longitudinal study in the future, incorporating comprehensive individual patient medical histories.

This study compared the contralateral femur BMD in 343 Korean postmenopausal women. Statistically significant disparities in bone mineral density were observed between the right and left femurs, particularly in femur total. Especially among postmenopausal women with osteoporosis noteworthy variations were observed not just in the femur total but also in the trochanter area BMD, underscoring clinical significance. Therefore, osteoporosis screening in patients based only on the unilateral femoral scanning can lead to underdiagnosis. Further large-scale, longitudinal studies are required to further assess the effect of dominant leg, osteoporosis treatment, and menopausal hormone therapy.