Femoral Neck Width Predict Low Bone Mass in Patients with Multiple Hereditary Exostoses

Background: Recent report revealed multiple hereditary exostoses (MHE) patients showed the osteoporosis. This study aimed to determine if proximal femur deformities can indicate low bone mass in MHE patients and to investigate the correlation between Z-score or T-score and hip geometry parameters. Patients and Methods: Twenty MHE patients from unrelated families were included in this study. Bone mineral density (BMD) of both femoral neck (n = 40) was examined using dual-energy x-ray absorptiometry. We examined femoral neck axis length (FNAL), femoral head diameter (FHD), femoral neck width (FNW), femoral shaft width (FSW), and femoral neck angle (FNA) and calculated the femoral head-neck ratio (FHNR=FHD/FNW). We examined the correlation between these parameters and Z-score or T-score using linear regression analysis followed by Spearman rank correlation coecient. Results: Of the patients, 91.7% (22 of 24) male and 100% (16 of 16) female had a Z-score <0 in the femoral neck area. Results also showed that the femoral neck area of 62.5% (15 of 24) male and 56.3% (9 of 16) female patients are within the range for osteopenia. The femoral neck are of seven patients (17.5%) were found to be within the range for osteoporosis. These results suggest that MHE patients tend to have low bone mass. We found a signicant correlation between FNW and Z-score (r = -0.3924, P = 0.0123), but there was no signicant difference between Z-score and FNAL, FHD, FSW, NSA, and FHNR. We also found that there were signicant differences between T-score and FNW and T-score and FHNR (r = -0.4787, P = 0.0018 and r = 0.3636, P = 0.0211, respectively). There was no signicant difference between T-score and the other parameters. Conclusions: We found that the femoral neck width femoral head-neck

Exostosis is a cartilage-capped benign bone tumour that typically originates from the metaphysis of long bones or from the surface of at bones. The symptoms are very frequently related to the tumour size, and some patients may undergo multiple surgeries during their life in an attempt to relieve the symptoms of this disorder [9]. The risk of malignant transformation toward chondrosarcoma, which is the most serious complication, occurs in 0.38-7.0% of patients [9][10][11][12].
Osteoporosis is a disease which causes bone fragility characterized by low bone mineral density and altered bone microstructure. Previously, Lemos et al. reported an association between MHE and osteoporosis in Australian kindred with an EXT1 splice site mutation [13]. Recently, several mice models were established, and understanding of MHE has advanced. Nozawa et al. reported that heparan sulfate (HS) regulates bone mass by interacting with osteoprotegerin and found that heparan sulfate modi ed transgenic mouse have low bone mass [14]. Recently, Matsumoto et al. reported that two-thirds of MHE patients exhibit osteopenia and have low bone mass in their femoral neck area [15]. However, there were no abnormalities in their bone metabolic factors which indicates that the low bone mass in their femoral neck may be due to the deformities related to exostoses. Therefore, in this retrospective cohort study, we hypothesized that proximal femur deformities can be indicated by the low bone mass in MHE patients and investigated the correlation between osteopenia and femoral neck deformities.

Participants
All the MHE patients included in this study were from unrelated families. They were diagnosed with MHE and were followed up at the Department of Orthopedic Surgery of our University Hospital from April 2010 to March 2019. Information, such as age, ethnicity, menstrual status, medication history, and disease history, was obtained via questionnaire. The exclusion criteria were postmenopausal at the time of Bone Mineral Density (BMD) measurement, hip surgery history, and under 20 or over 50 years old. Twenty-six patients were eligible. None of the female subjects were postmenopausal at the time of BMD measurement. Patients who lacked data, including BMD measurement or hip X-ray, were also excluded.
Finally, 20 MHE patients (40 hip joints) were enrolled in this study. The Ethics Committee of our University approved all procedures (Approval No. , and all participants provided written informed consent before any research procedure.

BMD Measurement
Both femoral neck (n = 40) was examined using dual-energy x-ray absorptiometry (DEXA) (Lunar DPX-NT, GE Healthcare, Japan). BMD and T-score were obtained from the measurements using the manufacturer's software (enCORE, GE Healthcare, Japan). BMD data were compared with reference values for normal Japanese populations of corresponding age and sex published by the Japanese Society for Bone and Mineral Research [14]. The enCORE reference database was used to determine the T-score for the femoral neck. Osteopenia was de ned as a T-score between -1 and -2.5 while osteoporosis was de ned as a score of -2.5 or lower, according to the criteria of the World Health Organization (WHO) [16]. The Z-score or the number of standard deviations above or below the mean for an age-matched population, was also analysed.

Hip Geometry Measurement
The hip geometry measurements are shown in Figure 1. Hip radiographs were taken for the clinical investigation of tumours around the hip to detect hip deformities caused by tumours, malignant transformation, and hip osteoarthritis. Anteroposterior radiographs were used to evaluate the shape of the proximal femur. We examined femoral neck axis length (FNAL) or the distance from the centre of the femoral head to the intersection of the neck and shaft axes, femoral head diameter (FHD), femoral neck width (FNW), femoral shaft width (FSW), and femoral neck angle (FNA) or the angle between derived axes of the neck and shaft. We also calculated the femoral head-neck ratio (FHNR=FHD/FNW) and examined the correlation between these parameters and BMD, Z-score, or T-score.

Statistical Analysis
All parameters were reported as mean ± standard deviation. Differences between the groups were compared using the Chi-square test for qualitative data. Correlations between BMD, Z-score, or T-score and hip geometry measurements and ratio were analysed using linear regression analysis followed by Spearman rank correlation coe cient. These analyses were performed using GraphPad Prism 5.0 software (GraphPad Software, Inc., San Diego, CA, USA). A P value < 0.05 was considered signi cant.

Patient Background
We examined the BMD parameters of 20 MHE patients, 12 male and 8 female, using DEXA. The mean age was 39.7 years for male (range: 29-48 years; n = 12) and 36.6 years (range: 29-45 years; n = 8) for female. The mean femoral neck BMD value was 0.725 g/cm 2 for male (range: 0.542-1.029 g/cm 2 ; n = 24) and 0.720 g/cm 2 for female (range: 0.508-0.919 g/cm 2 ; n = 16). There was no signi cant difference in the femoral neck BMD values between male and female (P = 0.315) ( Table 1). There were also no signi cant differences in the Z-scores and T-scores between male and female (Table 1).
We analysed femoral neck area Z-score and T-score and found that 91.7% (22 of 24) of male and 100% (16 of 16) of female patients had a Z-score < 0 in the femoral neck area (Figure 2A). According to the WHO criteria, 62.5% (15 of 24) of the male patients and 56.3% (9 of 16) of the female patients are within the range of osteopenia ( Figure 2B). Three male and four female patients (17.5%) were within the range to have osteoporosis in the femoral neck area. We examined the differences in Z-score and T-score of the male and female patients but found no signi cant differences (Figure 2).
Correlation between Z-score or T-score and Hip Geometry Parameters We examined the correlation between Z-score and the hip geometry parameters. The mean value of FNAL was 99.4 ± 10.1 mm and there was no signi cant difference between FNAL and Z-score (r = -0.0334, P = 0.8374, Figure 3A). The mean value of FHD and FNW were 51.3 ± 4.6 mm and 48.4 ± 11.7 mm, respectively. There was no signi cant difference between Z-score and FHD (r = 0.0977, P = 0.5488, Figure  3B), but there was a signi cant difference between Z-score and FNW (r = -0.3924, P = 0.0123, Figure 3C).
The mean value of FSW and NSA were 33.5 ± 3.3 mm and 136.4 ± 7.5 degrees, respectively. There was no signi cant difference between Z-score and FSW and between Z-score and NSA (r = 0.0936, P = 0.5658 and r = -0.1047, P = 0.5203, respectively) ( Figure 3D-E). The mean value of FHNR was 1.11 ± 0.3 and there was no signi cant difference between Z-score and FHNR (r = 0.1640, P = 0.3120, Figure 3F).
We further analysed the correlation between T-score and the hip geometry parameters. There was no signi cant difference between T-score and FNAL and between T-score and FHD (r = -0.0088, P = 0.9569 and r = -0.1234, P = 0.4482, respectively) ( Figure 4A-B). There was a signi cant difference between Tscore and FNW and between T-score and FHNR (r = -0.4787, P = 0.0018 and r = 0.3636, P = 0.0211) ( Figure 4C, 4F). There was no signi cant difference between T-score and FSW and between T-score and NSA (r = 0.0273, P = 0.8668 and r = -0.2074, P = 0.1992) ( Figure 4D-E).

Discussion
In this study, we examined the BMD and the correlation between Z-score or T-score and geometry parameters of the proximal femur. We found that three-fourths of MHE patients show a range of osteopenia or osteoporosis and that femoral neck width signi cantly correlates with Z-score and T-score. These results suggest that femoral neck width may be a reliable predicting factor of bone mineral density.
Several mice models for this disease were established recently and understanding of this disease has been gradually advancing [3,17,18]. Nozawa S et al. reported that heparan sulfate (HS) regulate bone mass by interacting with osteoprotegerin and reduced bone mineral density in mice model [17]. However, there is few information about bone mineral density in MHE patients. Recently, Matsumoto et al. reported that MHE patients have low bone mass in the lumbar spine and femoral neck area [15]. In this study, the bone mineral density of most MHE patients in their femoral neck area (31 out of 40 hips) were within osteopenia or osteoporosis range.
Osteoporosis is a common disease characterized by reduced bone mass. The primary pathogenic mechanism of osteoporosis is an imbalance between bone formation and bone resorption, which are mediated by osteoclasts and osteoblasts, respectively. Recently, Matsumoto et al. reported that serum osteocalcin and urine NTx, the markers of bone formation and bone resorption, respectively, are within normal range in almost all MHE patients [15]. These results suggest low bone mass in this disease might not be due to the alterations in bone metabolism. Therefore, we hypothesized that the low bone mass in MHE patients may be due to developmental abnormalities, such as proximal femur deformities, caused by the development of exostoses.
Hip exostoses are present in a high percentage of MHE patients, mainly in the proximal femur, which they have been reported in 30% to 90% of patients [19][20][21]. Furthermore, Duque Orozco et al. reported that hip exostoses were found in 90% of patients, and their most common location was the femoral neck [22]. In this study, we focused on the hip geometry parameters and examined the correlation between these parameters and Z-score or T-score. Interestingly, there was no signi cant correlation between Z-score or T-score and femoral neck axis length, femoral head width, femoral shaft width, and neck shaft angle. Only femoral neck width was found to have a signi cant correlation with Z-score and Tscore. As we have mentioned above, femoral neck is the frequent site for exostoses. Taken together, these facts suggest that the low bone mass of MHE patients is not due to abnormal bone metabolism, but to the development of exostoses.
Although exostoses around the hip are often asymptomatic, they can produce various deformities including hip dysplasia, subluxation, impingement, and premature osteoarthritis [20,[23][24][25]. In this study, we focused on these characteristic deformities and examined the correlation between various hip geometry parameters and BMD. We found that the femoral neck width may be one reliable predictor of low bone mass. These results suggest that the wider the femoral neck is, the lower the bone mass. These data support the hypothesis that the low bone mass in MHE patients may be due to the deformities caused by exostoses. However, MHE patients show various deformities including hip dysplasia and subluxation. These deformities may alter normal weight bearing and may affect bone mineral density.
The mechanisms that lead to low bone mass in MHE patients are still unclear and more evidence are needed to elucidate them.
There are several limitations in this study. First, there is a limited number of MHE patients from only one institution. Second, we did not evaluate the deformities in the lower legs which may affect the degree of low bone mass in these patients. However, this is the rst report where the relationship between BMD and hip geometry parameters in MHE patients was examined in detail. When a wide femoral neck is observed, we should suspect low bone mass in the patients. These results could provide useful information in understanding the mechanisms of MHE.

Conclusions
We found that the femoral neck width signi cantly correlates with Z-score or T-score. These results suggest that the femoral neck width may be a reliable predicting factor of bone mineral density in MHE patients.

Declarations
All methods were carried out in accordance with relevant guidelines and regulations.

Ethics approval and consent to participate
The Ethics Committee of our University approved all procedures (Approval No. , and all participants provided written informed consent before any research procedure.

Consent for publication
All participants accepted the publication.

Availability of data and materials
The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
All authors have no con ict of interests.

Funding
No fundings.
Authors' contributions KM, and HA prepare the manuscript, and KM, HO, and SN collected and analysed the data.  Comparison of (A) Z-score and (B) T-score of the femoral neck area in MHE patients. There were no signi cant differences between male and female. A T-score between -1 to -2.5 indicates osteopenia while a T-score <-2.5 indicates osteoporosis.

Figure 3
Correlation between Z-score and the hip geometry parameters. FNW was signi cantly correlated with Zscore.

Figure 4
Correlation between T-score and the hip geometry parameters. FNW and FHNR were signi cantly correlated with T-score.