Age and sex differences in bone density, height, and width: a correlational study

Introduction

Bone health is a critical component of overall well-being, particularly as individuals progress in age. Changes in bone density, height, and width are known to occur with aging, potentially leading to conditions, e.g., osteoporosis, which increases the risk of fractures [1]. Previous studies have demonstrated that bone parameters, such as density and structural dimensions can vary significantly with age and between sexes, impacting bone strength and the likelihood of age-related bone diseases [2].
Despite extensive research, the specific relationships between age, sex, and bone characteristics remain complex and contradictory at times. While some studies suggest a clear decline in bone density with age, others report minimal changes, especially in men [3, 4]. Although the influence of sex on bone height and width has been recognized, the degree of this impact and its implications for bone health need further exploration.
In dentistry, bone health is a critical determinant of successful dental implant outcomes, particularly involving bone density, height, and width [5]. These para­meters are crucial for ensuring adequate osseointegration, which is the process by which an implant integrates with the surrounding bone. The quality and quantity of bone available at the implant site directly influence the choice of implant type, size, and long-term stability of the prosthetic restoration [6, 7].
Previous studies have established that factors, such as age and sex significantly affect bone quality and structure, which in turn may impact the success rate of dental implants [8, 9]. Understanding these relationships is therefore crucial for clinicians when planning implant placement, especially in older patients or those with varying bone densities. According to the Misch classification, bone density can be divided into D1 (dense cortical bone), D2 (thick, dense to porous cortical bone on crest and coarse trabecular bone within), D3 (thin, porous cortical bone on crest and fine trabecular bone within), and D4 (fine trabecular bone) [10]. Despite its inherent subjectivity, the Misch classification was selected in this study due to its widespread clinical acceptance, ease of use, and feasibility in routine practice compared with more quantitative but resource-intensive methods, such as CT-based bone density analysis. In particular, there is limited research that integrates the Misch classification system (D1-D4), a clinically applied yet subjective method for evaluating bone density, into correlational analyses involving both structural dimensions and demographic factors [11]. A comprehensive assessment of these parameters using clinically relevant tools may help bridge the gap between research findings and real-world dental implant planning.

Objectives

This study aimed to explore the correlations between bone density according to the Misch classification and structural parameters (bone height and width) as well as demographic variables, i.e., age and sex, to provide insights, which may support evidence-based approaches in implant dentistry.

Material and methods

Study design

This study used a cross-sectional design to analyze variables associated with the quality and quantity of the edentulous region, including age, sex, bone qua­lity (classified according to Misch [10]), bone height, bone thickness, and cone beam computed tomography (CBCT) examination results. Ethical approval for this research was obtained from the Ethics Committee of the Faculty of Dentistry, Universitas Indonesia, under approval number 88/Ethical Approval/FKGUI/XII/2023 and protocol number 051211223. The study was conducted in accordance with the Declaration of Helsinki.

Sample selection

Samples were taken from edentulous sites in the posterior mandibular region, ranging from the first premolar to second molar, on either the left or right side. To ensure the independence of observations and avoid clustering effects, only one edentulous site was selected and analyzed per patient. These measurements were obtained using CBCT images to accurately assess bone conditions in these regions. Patients aged 30-49 years and 50-70 years were included. These age ranges were selected because they represent age of majority of individuals undergoing dental implant procedures. Patients aged under 30 years typically have stable bone architecture, while those aged above 70 years may present with more systemic bone changes because of aging, such as osteoporosis. The two selected groups (30-49 and 50-70) also reflect meaningful biological stages, with the latter more likely to exhibit early signs of bone resorption or density reduction. Excluding patients above 70 helped reduce variability due to age-related systemic factors. CBCT images were excluded if they presented conditions that could affect interpretation, such as patholo­gical conditions, jaw fractures, complete edentulism, or systemic diseases affecting jawbone integrity.
Minimum sample size for this correlational study was calculated using G*Power 3.1 software. Based on a two-tailed Pearson’s correlation analysis as well as an anticipated moderate effect size (r = 0.3), a statistical power of 0.80, and a significance level of α = 0.05, the minimum required sample size was 84. To account for potential data loss or participant dropout, the final sample size was increased by approximately 20%, resulting in a total of 102 subjects, equally distributed between two age groups. Dependent variables were bone quality (classified according to Misch) and bone quantity (height and thickness of bone), whereas independent variables were age (categorized into 30-49 years and 50-70 years) and sex (men or women).

Operational definitions

1. Age: The duration from birth until the CBCT exami­nation date, as recorded in medical records. Age is categorized into two groups: 30-49 years and 50-70 years (categorical).
2. Sex: Biological differentiation into men or women since birth, as recorded in medical records (catego­rical).
3. Bone density: Assessed in each edentulous area based on the Misch classification [10] by two observers using three selected sections for each implant placement (Figure 1) [12]. Sections are taken at 1, 3, and 5 transaxial-coronal areas from slices perpendicular to the patient’s edentulous region arc at regular intervals (typically 1-2 mm) (categorical) [13, 14].
4. Bone height: Measurement of vertical mandibular bone atrophy in the edentulous region using 1, 3, and 5 transaxial-coronal sections (Figure 2). Vertical lines are drawn from the alveolar crest to the most inferior point of the mandibular cortex (mm; numerical, including minimum, maximum, mean, and standard deviation) [13, 14].
5. Bone width: Measured on 1, 3, and 5 transaxial-coronal sections perpendicular to the edentulous region arc at regular intervals (1-2 mm) (Figure 2). Horizontal lines are drawn from buccal to lingual areas, approximately 3.0 mm from the alveolar bone crest (mm; numerical, including minimum, maximum, mean, and standard deviation) [13, 14].
Intra- and interobserver reliability tests were conducted on 51 samples, representing 50% of the total sample size. The intraobserver reliability test was performed to assess the consistency of measurements by the same observer at different times, while the interobserver test evaluated the consistency of measurements between different observers. Both intra- and interobserver tests were done at least two weeks after the initial measurements. The interobserver test was carried out by two dental radiology specialist students selected based on their experience. Before data collection, both observers underwent a calibration session led by a senior teaching staff member in oral and maxillofacial radiology with over 15 years of experience in the field. Observers were trained using representative CBCT images with guidance on identifying reference points for bone density classification and dimensional measurements. This training ensured consistency and reduced potential bias across all assessments. For the analysis of bone density in the edentulous mandibular posterior region using CBCT, Cohen’s κ reliability test was employed, as data were categorical. While for numerical data related to bone height and bone width measurements, the intraclass correlation coefficient (ICC) reliability test was applied.

Research tools and equipment

A computer with CBCT reconstruction software (CS 3D Imaging Software v. 7.0.23, Carestream Dental, Atlanta, USA) and a CBCT machine (type CS New 9300, 3D Digital Imaging System, Carestream Dental), with voxel sizes ranging from 90 to 500 μm and various fields of view (10 × 5, 10 × 10, 17 × 11, and 17 × 13 cm) were used.

Statistical analysis

Statistical analysis aimed to explore the relationships between bone density, height, and width with age and sex. Data were analyzed using the Spearman’s correlation test, which is suitable for assessing the strength and direction of associations between variables, particularly when data do not meet the assumptions of parametric tests, such as normality. Age was categorized into two groups: 30-49 years (coded as 0) and 50-70 years (coded as 1), whereas sex was classified as men (coded as 0) and women (coded as 1). Bone density was visually assessed and categorized according to the Misch classification (D1-D4), and bone height and width were measured in millimeters. Descriptive statistics were calculated for each variable, including means, standard deviations, and ranges for continuous variables (bone height and width) as well as frequencies and percentages for categorical variables (bone density). All statistical analyses were performed using SPSS (SPSS version 20.0; IBM Corp., Armonk, NY, USA), with a significance level set at p < 0.05 for all tests.

Results

The κ values for intra- and interobserver reliability ranged from 0.873 to 0.892, respectively, indicating almost perfect agreement. The ICC analysis, based on a two-way mixed-effects model with absolute agreement (single measures), demonstrated high reliability. For bone height, ICC values ranged from 0.946 to 0.994, indicating almost perfect agreement. Regarding bone width, ICC values ranged from 0.877 to 0.916, also reflecting a high level of agreement. These values were consistent across both intra- and interobserver assessments. The interpretation was based on the Landis and Koch criteria, where values above 0.81 denote almost perfect agreement [15].
Table 1 presents the distribution of the subjects based on age and sex, along with the corresponding p-values for various bone density types, bone height, and bone width. The data showed no significant difference in bone density across different age groups (30-49 years and 50-70 years), or between men and women. Bone height demonstrated no significant difference across the two age groups. When analyzed by sex, men had a significantly higher mean bone height (28.83 ± 3.6 mm) compared with women (27.37 ± 14.29 mm), with a p-value of 0.000, indicating a statistically significant difference. Bone width showed no significant difference across the two age groups. However, when comparing sexes, men presented a significantly greater bone width (9.2 ± 2.22 mm) compared with women (7.84 ± 1.97 mm), with a p-value of 0.003, signifying a significant sex-based difference. Overall, while bone density did not vary significantly by age or sex, bone height and width revealed statistically significant differences between men and women.
Table 2 provides the results of the Spearman’s correlation test, examining the relationship between bone density, bone height, and bone width with age and sex. The correlation between bone density and age was weak and not statistically significant (r = 0.158, p = 0.113). Similarly, the correlation between bone density and sex was weak and nonsignificant (r = –0.120, p = 0.228). This indicated that bone density was not significantly influenced by age or sex in this study population. The correlation between bone height and age was weak and not significant (r = 0.090, p = 0.368). However, there was a strong negative correlation between bone height and sex (r = −0.424, p = 0.000), signifying that sex was a significant factor in bone height, with men generally having a greater bone height than women. The correlation between bone width and age was weak and nonsignificant (r = 0.115, p = 0.249). However, the negative correlation between bone width and sex was moderate and statistically significant (r = –0.296, p = 0.003), suggesting that sex significantly affected bone width, with men generally having wider bones than women.

Discussion

This study explored the relationships between bone density, height, and width with age and sex in a cohort aged 30-70 years. The findings provide valuable insights into how these bone parameters vary according to demographic factors. A key observation is the significant nega­tive correlation between bone height and sex, indicating that men generally exhibit greater bone height than women. Notably, the standard deviation of bone height among female participants (± 14.29 mm) was substantially higher than that of males (± 3.60 mm), reflecting genuine intragroup variability. This may be attributed to a broader age distribution among female participants, postmenopausal bone loss, and a higher prevalence of edentulism-related resorption, all of which contribute to differences in alveolar ridge dimensions. These findings align with established evidence that men typically have longer and denser bones, largely due to hormonal influences, such as testosterone, which promotes bone growth and mineralization [16]. This strong correlation emphasizes the importance of considering sex-specific strategies in managing bone health and preventing conditions, such as osteoporosis. Similarly, the study found a significant negative correlation between bone width and sex, with men displaying greater bone width than women. This finding aligns with the existing literature on sex-related differences in bone structure, with men generally having thicker and more robust bones [17]. These structural differences are likely to contribute to the lower incidence of osteoporotic fractures in men compared with women, who typically experience more bone loss with age due to declining estrogen levels [18].
The analysis revealed no statistically significant correlation between bone density and age or sex. This suggests that within the age range studied, bone density remains relatively stable, contrary to the general assumption that bone density uniformly decreases with age. One possible explanation is that the selected age groups (30-49 and 50-70 years) may not adequately capture age, at which significant bone loss typically occurs, particularly in populations aged over 70 years. In addition, bone remodeling with age can vary greatly between individuals depending on hormonal status, genetics, and systemic conditions, which were not controlled in this study. Several studies have shown that bone density loss becomes more pronounced in postmenopausal women [19, 20], yet this study did not find a significant difference. This discrepancy may be due to methodological differences. While DEXA and CT imaging provide precise and quantitative measurements, such as bone mineral density (BMD) or Hounsfield units (HU), our study employed CBCT with the Misch classification (D1-D4), which relies on grayscale values and subjective visual interpretation. Although this method is widely used in clinical settings due to its practicality, it lacks the sensitivity to detect subtle age-related changes in bone density. These limitations may have contributed to the absence of a significant correlation in our findings. However, it is important to note that while HU values in CBCT can offer approximate assessments of bone density, they are not standardized across different CBCT devices. This is due to variations in machine calibration, field of view, exposure parameters, and reconstruction algorithms. As a result, HU values obtained from CBCT should be interpreted with caution and are generally considered less reliable than those from conventional medical CT. Furthermore, software-derived HU values in CBCT often reflect relative grayscale values rather than true tissue density [21]. While the Misch classification offers practical clinical utility, it lacks the quantitative accuracy of HU-based or densitometric methods. In this study, to minimize variability, all scans were performed using the same machine and protocol, and observers were trained and calibrated before data collection. As a result, the use of CBCT in this study might have contributed to the lack of a significant correlation between bone density and age. The Misch classification is primarily designed to assess bone quality in a categorical manner, rather than to provide continuous and quantifiable data on bone density. Consequently, while CBCT-based assessment is valuable for clinical decision-making, it may not fully capture the nuances of age-related changes in bone density, which might be detected using more precise methods.
The findings of this study carry significant implications for dental implantology. The weak correlation between bone density and age suggests that bone density in the age range studied may not deteriorate significantly, which is a positive indication for older adults seeking dental implants. However, the method of assessing bone density according to the Misch classification (D1-D4) could influence these findings, as it may not capture subtle variations in bone density. The strong negative correlation between bone height and sex highlights that men typically have greater bone height, which could be advantageous in implant planning, as sufficient bone height is essential for the stability and success of dental implants [5, 8]. The moderate negative correlation observed between bone width and sex indicates that men generally have wider alveolar ridges, which may facilitate the placement of standard or wider-diameter implants, and potentially enhance osseointegration and long-term outcomes. Given these findings, sex-specific considerations should inform dental implant treatment planning [17]. For male patients, standard implant protocols can often be appropriate due to their generally greater bone dimensions. In contrast, for female patients, particularly those with reduced bone height and density, clinicians should employ a more tailored approach. Preoperative CBCT imaging is recommended to assess bone volume and quality more thoroughly. When bone dimensions are insufficient, treatment strategies may include utilization of narrow or short-diameter implants, staged implant placement, or bone augmentation techniques, such as guided bone regeneration (GBR) and sinus lift procedures in the maxillary posterior region. These individualized approaches can help improve primary stability and long-term implant success in female patients.
This study is limited by its cross-sectional design, which precludes the establishment of causality. Although CBCT provides useful structural information for clinical implant planning, it does not yield absolute Hounsfield unit (HU) values, and its use in bone density analysis is inherently subjective. Future studies could benefit from integrating CBCT with quantitative imaging modalities, such as dual-energy X-ray absorptiometry (DEXA) or CT-based HU analysis, which offer more precise and objective measurements of bone density, and may better elucidate its relationship with age. Another limitation lies in the selected age range of 30–70 years. While this age range reflects typical demographics of implant patients, it may not capture more advanced age-related bone loss. Including individuals over 70 years old may provide a more comprehensive understanding of skeletal changes with aging. Additionally, although this age range was chosen to reduce the influence of age-related variability, systemic conditions, such as poorly controlled diabetes mellitus, long-term corticosteroid use, or bisphosphonate therapy, may still impact bone metabolism and density within this group. Future studies should incorporate detailed systemic health screening to control for these potential confounding factors. Furthermore, the study did not account for individual-level variables, including body mass index, physical activity, hormonal status, or nutritional intake, all of which may influence bone density, height, and width. Lastly, the study was conducted in a relatively homogenous population. Including participants from diverse ethnic, dietary, and geographic backgrounds in future studies could enhance understanding of the broader biological and environmental factors affecting bone characteristics relevant to implantology. Future research should consider these variables to improve clinical relevance and applicability.

Conclusions

The findings suggest that sex plays a significant role in determining bone height and width, which are crucial for dental implant planning. These results underscore the need for sex-specific considerations in implantology, particularly in bone augmentation strategies for women.

Disclosures

1. The approval of the Bioethics Committee for the research: This study was approved by Ethics Committee of the Faculty of Dentistry, Universitas Indo­nesia, under approval number 88/Ethical Approval/FKGUI/XII/2023 and protocol number 051211223.
2. Assistance with the article: None.
3. Financial support and sponsorship: None.
4. Conflicts of interest: None.

References