Introduction
The increasing aging population and extended life expectancies have promoted a significant increase in the prevalence of adult spinal deformity, making it a growing concern in global healthcare [1–3]. This condition is particularly common among individuals aged >65 years, with prevalence rates ranging from 32% to 68% in this demographic [1]. In clinical practice, evaluating sagittal spinal alignment is crucial for effective surgical planning and understanding disease pathology. Numerous studies have shown a strong correlation between spinal alignment parameters and clinical outcomes, emphasizing the need for a comprehensive assessment strategy that includes spinal and pelvic parameters [4,5]. A comprehensive assessment of the spinopelvic alignment, particularly in the coronal plane, is crucial because it directly affects the overall balance and quality of life of patients [6,7]. However, evaluating sagittal alignment can be difficult in patients with scoliosis because their pelvic rotation hinders the accurate identification of the center of the femoral heads on sagittal radiographs [8]. This challenge is further complicated in patients with obesity in whom obtaining a clear view of the femoral heads on sagittal radiographs is difficult. Therefore, a method that allows for the evaluation of the pelvic tilt (PT) from the frontal view is desirable.
The sacrofemoral pubic (SFP) angle is a valuable parameter for estimating PT from coronal radiographs and has been used effectively in various clinical scenarios to assess pelvic alignment from the frontal view [9–12]. Despite its utility, the SFP angle has some limitations, particularly among patients with hip deformities or advanced osteoarthritis in whom visualization of the femoral heads is challenging. Moreover, the SFP angle does not fully capture the relationship between lower limb alignment and overall body alignment [12,13]. To address these limitations, a new indicator for evaluating PT from the frontal view was proposed, called the pelvic ring ratio. This new parameter aims to provide a more accurate and comprehensive assessment of pelvic alignment, particularly in complex cases where traditional methods are inadequate. EOS imaging technology (EOS Imaging, Paris, France) allows for obtaining detailed whole-body alignment, enabling a thorough evaluation of the PT in relation to the overall body alignment.
This study aimed to evaluate the utility of the pelvic ring ratio by determining its effectiveness as an indicator of PT assessment and its potential applications in clinical practice. This new indicator may enhance diagnostic accuracy and improve treatment planning for patients with spinal and hip disorders.
Materials and Methods
Participants
In this study, 104 healthy individuals (28 men and 76 women; mean age, 52.8±12.3 years) were proposed, all of whom were verified to have no previous or present medical history of spinal disorders through a comprehensive questionnaire designed to evaluate clinical complaints. The age distribution of the participants was as follows: 12, 30, 32, 16, and 14 individuals were in their 20s–30s, 40s, 50s, 60s, and 70s, respectively. Individuals with neurological conditions or joint disorders that could affect standing posture assessment were excluded from this study.
This study was conducted with the approval of the institutional review boards of the participating institutions (approval no., 2022–012 (0478)). All participants were provided with detailed information about the study, including its purpose and potential risks, and gave their written informed consent prior to participation.
Radiographical examination
Participants underwent whole-body radiography using an EOS imaging system (EOS Imaging). During the examination, the participants adopted a “hands-on-cheek” position while standing horizontally, gently touching their fingers and remaining as relaxed as possible throughout the imaging process.
Radiographical measurements
Whole-body sagittal alignment (WBSA) was analyzed from the craniocervical junction to the knee joint. The specific measurements were as follows: occipito–C2 angle (O–C2 angle: McGregor line to the C2 endplate), C2–7 lordotic angle (C2 endplate to C7 caudal endplate), T1 slope, thoracic kyphosis (T1–12), lumbar lordosis (L1–S1), sacral slope (SS), PT, pelvic incidence (PI), average knee flexion (average of the left and right knee flexion angles), and average ankle flexion (average of the left and right knee flexion angles). Kyphosis and lordosis were defined as the angles between the upper endplate of a selected vertebra and the lower endplate of another vertebra.
Evaluation of the pelvic ring ratio
The pelvic ring ratio was calculated using measurements taken from the frontal radiographs obtained during EOS imaging. Specifically, the vertical distance was defined as the distance between the line connecting the lower ends of the sacroiliac joints and the cranial edge of the pubic symphysis. The horizontal distance was defined as the maximum transverse diameter of the pelvic ring, which was measured at its widest point (Figs. 1–3). The pelvic ring ratio was then calculated as the ratio of these two measurements: pelvic ring ratio=(vertical distance/horizontal distance)×100%
The postural assessment workflow provided by EOS imaging was used to measure the WBSA parameters. Participants were categorized into five age groups (20–30s, 40s, 50s, 60s, and ≥60s) and compared according to the results. To elucidate the relationship between the pelvic ring ratio and WBSA parameters, correlations between these parameters were evaluated. In addition, the pelvic ring ratio was divided into three categories (<30%, 30%–60%, and >60%) and investigated whether significant differences in WBSA parameters were present among these groups.
Statistical analysis
Data are presented as means±standard deviations. To assess the strength of the correlations between the parameters and between age and each parameter, Pearson’s correlation coefficient (r) was employed. To assess the reliability of the pelvic ring ratio measurements, both intraobserver and interobserver variability were evaluated using intraclass correlation coefficients (ICCs). Three spine surgery specialists independently measured the pelvic ring ratio. For the intraobserver variability, each specialist repeated the measurements after a 2-week interval, and the average ICC from the three specialists was calculated.
In the development of a predictive model for PT, multiple linear regression analysis was performed using the pelvic ring ratio and sex as predictors. All statistical analyses were performed using IBM SPSS Statistics ver. 29.0 (IBM Corp., Armonk, NY, USA), with p-value <0.05 indicating significance.
Results
The intraobserver ICC was 0.89 (average of the three specialists), and the interobserver ICC was 0.87. The mean pelvic ring ratio was 53.3%±11.5%. A histogram summarizing the results is presented in Fig. 4. The pelvic ring ratio according to the age group was as follows: 58.1%±10.0% (20–30s), 57.8%±9.7% (40s), 52.1%±9.5% (50s), 51.8%±11.3% (60s), and 43.7%±14.8% (70s). Significant differences in the pelvic ring ratio were observed between the 20–30s and 70s groups (p=0.008) and between the 40s and 70s groups (p=0.001), with lower ratio in the 70s group. No significant differences were found between the other age groups. Furthermore, the pelvic ring ratios according to sex were 48.6%±10.8% for men and 55.0%±11.3% for women, with women having a significantly higher pelvic ring ratio (p=0.01).
Correlation between the pelvic ring ratio, age, body mass index, and various WBSA parameters
A significant negative correlation was observed between the pelvic ring ratio and age (r=−0.387, p<0.05), indicating a decreased pelvic ring ratio with age (Table 1). No significant correlations were found between the pelvic ring ratio and height (r=0.023, p=0.815) or body mass index (BMI) (r=−0.113, p=0.252), O–C2 angle (r=0.049, p=0.623), and TK (r=−0.074, p=0.453). However, significant negative correlations were observed between the pelvic ring ratio and the C2–7 angle (r=−0.202, p<0.05), T1 slope (r=−0.206, p<0.05), PT (r=−0.598, p<0.05), knee flexion angle (r=−0.453, p<0.05), ankle angle (r=−0.547, p<0.05), and SVA (r=−0.218, p<0.05). Conversely, a significant positive correlation was noted between the pelvic ring ratio and LL (r=0.393, p<0.05), whereas no significant correlation was observed between the pelvic ring ratio and PI (r=−0.139, p=0.159).
Comparison among the three groups based on their pelvic ring ratio: <30%, 30%–60%, and ≥60%
The <30% group (n=5; mean age, 68.2±6.1 years) had an O–C2 angle, C2–7 angle, TK, LL, PI, PT, knee flexion angle, and SVA of 18.0°±11.5°, 7.0°±22.9°, 31.7°±20.3°, 38.0°±13.5°, 55.1°±15.2°, 27.6°±8.8°, 10.2°±10.9°, and 2.6±5.0 cm, respectively (Table 2).
The 30%–60% group (n=73; mean age, 53.4±11.5 years) had an O–C2 angle, C2–7 angle, TK, LL, PI, PT, knee flexion angle, and SVA of 14.5°±8.5°, 2.9°±13.1°, 30.7°±11.3°, 47.2°±11.0°, 49.1°±10.4°, 14.6°±7.6°, 2.1°±4.0°, and 0.5±2.7 cm, respectively (Table 2).
The ≥60% group (n=26; mean age, 48.2±5.9 years) had an O–C2 angle, C2–7 angle, TK, LL, PI, PT, knee flexion angle, and SVA of 15.6°±5.9°, 1.3°±13.1°, 32.4°±11.3°, 55.5°±11.0°, 50.7°±10.4°, 10.3°±7.6°, 0.2°±4.0°, and 0.5±3.0 cm (Table 2).
After the group comparisons, the <30% group had the highest mean age and PT, whereas the ≥60% group had the lowest knee flexion angle. Significant differences in the pelvic ring ratio were noted between the groups, indicating variations in the pelvic and spinal alignment parameters across different categories of the pelvic ring ratio.
Regression model to predict PT from the pelvic ring ratio
To develop a predictive model for PT, a regression model including the pelvic ring ratio, age, and sex as predictors was initially constructed: PT=35.81−0.48×pelvic ring ratio−0.0089×age+6.21×sex (where sex is coded as 0 for males and 1 for females). However, age was subsequently excluded from the model because of its lack of significance and minimal effect on the model’s explanatory power. The updated regression model, excluding age as a predictor, led to the following equation for predicting PT: PT=35.13−0.48×pelvic ring ratio+6.24×sex.
Discussion
This study evaluated the utility of the pelvic ring ratio as a new indicator for assessing PT from the frontal view. The pelvic ring ratio was calculated by measuring the vertical distance from the line that connects the lower ends of the sacroiliac joints to the cranial edge of the pubic symphysis, along with the horizontal diameter of the pelvic ring, on a frontal radiograph. This study primarily aimed to determine the effectiveness of the pelvic ring ratio in relation to the WBSA parameters. Using the EOS imaging technology, this study obtained detailed whole-body alignment data, allowing for a thorough evaluation of PT in relation to overall body alignment.
The regression model highlights the significance of considering both the pelvic ring ratio and sex in PT assessment. By integrating these two variables, the model provides a more comprehensive assessment of pelvic alignment, which can be useful for diagnosing and planning treatment for patients with spinal and hip disorders in clinical settings. In this study, the pelvic ring ratio demonstrated high reliability, with intraobserver and interobserver ICCs of 0.89 and 0.87, respectively. These values indicate excellent agreement, as ICCs >0.75 are generally considered indicative of excellent reliability. Comparatively, studies evaluating spinopelvic parameters have reported ICCs ranging from 0.75 to 0.90, reflecting good to excellent reliability [14]. For example, a study assessing spinopelvic measurements using the EOS imaging system reported ICCs between 0.7 and 0.9 [15], further supporting the robustness of our results. These findings indicate that the pelvic ring ratio is a consistent and reproducible measurement, comparable with the established spinopelvic alignment parameters. This reliability level is crucial for clinical applications because it ensures pelvic alignment assessments are dependable across different observers and over time. Such consistency enhances the utility of the pelvic ring ratio in both diagnostic evaluations and treatment planning for patients with spinal and hip disorders.
The pelvic ring ratio exhibited a strong negative correlation with PT (p<0.05), as indicated by the coefficient of −0.48. This finding indicates that an increase in the pelvic ring ratio corresponds to a decrease in PT. Specifically, for every unit increase in the pelvic ring ratio, PT decreases by approximately 0.48°. This inverse relationship emphasizes the role of the pelvic ring ratio in capturing variations in pelvic alignment, which is crucial for understanding the compensatory mechanisms that balance the musculoskeletal system.
This study showed that the coefficient for sex (i.e., 6.24) was significant (p<0.05). This positive coefficient indicates that, on average, women tend to have a PT that is 6.24° higher than that of men, holding the pelvic ring ratio constant. This finding aligns with known anatomical differences between the sexes, wherein women generally have a wider pelvic inlet and outlet [8,16], contributing to higher PT values. Thus, the model effectively captures these sex-related anatomical PT variations.
The potential influence of sex differences in the pelvic ring ratio is an important consideration. Anatomically, men and women have different pelvic morphologies, with women generally having a wider pelvic inlet and outlet to accommodate childbirth [8,16]. These structural differences could potentially affect the measurements of the pelvic ring ratio. This study found that women had a significantly higher pelvic ring ratio than men, which is likely attributable to the inherent anatomical variations in the pelvic structure between the sexes. Although sex differences are significant, their effect may be less pronounced than that of age and overall body alignment. In fact, age showed a strong negative correlation with the pelvic ring ratio, indicating a decrease in the pelvic ring ratio with age. This decrease likely reflects age-related changes, such as increased knee flexion and pelvic retroversion [17], which are compensatory mechanisms that maintain balance and posture as individuals age. Although age was excluded from the final regression model because of its lack of significance and minimal effect on the explanatory power of the model, its indirect effects through other variables should not be overlooked. These age-related changes can overshadow the sex differences observed, emphasizing the importance of considering both age and sex when assessing pelvic alignment.
This study found no significant correlation between BMI and the pelvic ring ratio (r=−0.113, p=0.252), suggesting that the pelvic ring ratio is less affected by obesity. This finding is important because it indicates that the pelvic ring ratio can be a reliable measure of PT even among patients with obesity in whom traditional measures such as PT may be challenging. Visualizing the femoral heads can be difficult in patients with obesity because of excessive soft tissue, which leads to higher interobserver variability in PT measurements [18]. In contrast, the pelvic ring ratio was less likely to be influenced by these factors, providing a more consistent and reliable assessment.
The SFP angle, a valuable tool for estimating PT from coronal radiographs in clinical practice [9–12], has been used effectively in various scenarios to assess pelvic alignment. However, it has its limitations, particularly in patients with hip deformities or advanced osteoarthritis in whom visualization of the femoral heads is difficult. In addition, the SFP angle does not fully reflect the relationship between the lower limb alignment and overall body alignment. In contrast, the pelvic ring ratio offers a more comprehensive assessment by incorporating measurements that reflect the entire pelvic ring structure. This new indicator offers advantages in evaluating complex cases where traditional methods such as the SFP angle are inadequate. The ability to measure PT from frontal radiographs is crucial for hip surgeons [10], as it enables them to reliably and reproducibly assess sagittal pelvic alignment without needing lateral radiographs [10]. PT influences the orientation of the acetabular component, thereby reducing the dislocation risk. This is particularly beneficial for the preoperative planning and intraoperative assessment of patients scheduled to undergo total hip arthroplasty (THA), given its ability to enhance surgical outcomes.
This has several limitations. First, the cohort consisted exclusively of healthy volunteers without spinal or hip disorders, which limits the generalizability of the findings to clinical populations where pelvic alignment issues are most relevant. Although the pelvic ring ratio demonstrated potential as a supplementary indicator, its utility in patients with spinal deformities, hip osteoarthritis, or those undergoing THA remains to be validated. Accordingly, we are currently conducting additional studies in these patient populations, and the results will be reported in future publications.
Second, the relatively small sample size within specific subgroups, such as age and sex distributions, may affect the robustness of the subgroup analyses. For instance, most participants were in the middle-aged group (40s–50s), with fewer participants in the younger (20s–30s) and older (70s) age groups. In addition, the cohort had a higher proportion of female participants (73%) than male participants (27%). These imbalances may influence the representativeness of the findings and should be considered when interpreting the results.
Finally, the reliance on the EOS imaging system, while advantageous for obtaining detailed whole-body alignment data, may limit the widespread adoption of this method in clinical practice because of its availability and cost. Future studies should explore alternative methods to assess the pelvic ring ratio, which are more accessible across diverse healthcare settings.
Conclusions
This study introduces the pelvic ring ratio as a novel indicator for assessing pelvic alignment from the frontal view. Our results demonstrate that the pelvic ring ratio correlates significantly with PT and other WBSA parameters, highlighting its potential as a reliable and reproducible measurement for understanding the spinopelvic balance. However, the inclusion of only healthy individuals limits the generalizability of these findings to clinical populations where pelvic alignment issues are most relevant.
This study was conducted to initially investigate and establish baseline data and validate the measurement methodology. Expanding the sample to include individuals with spinal deformities, hip osteoarthritis, or those undergoing THA is a critical next step to confirm the clinical utility of the pelvic ring ratio. Future studies should also examine specific subgroups, including varied age ranges, sex distributions, and patients with other musculoskeletal conditions. Such research will offer a deeper understanding of how the pelvic ring ratio interacts with diverse clinical presentations and enhance its utility in diagnosis and treatment planning.
Key Points
The study introduces the pelvic ring ratio as a reliable and reproducible method for assessing pelvic tilt (PT) from frontal radiographs, addressing limitations of traditional sagittal measurements, especially in patients with pelvic rotation or obesity.
The pelvic ring ratio demonstrated significant negative correlations with PT and other whole-body sagittal alignment parameters, such as the C2–7 angle, T1 slope, and knee and ankle flexion angles.
The study found that females had a significantly higher pelvic ring ratio compared to males, which is attributed to anatomical differences, with a stronger impact on PT observed in females.
A regression model developed in the study showed that the pelvic ring ratio, along with sex, significantly predicts PT, offering a useful tool for diagnosing and planning treatments for patients with spinal and hip disorders.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
