Characteristics of the positional and morphological parameters of sagittal spine alignment in a cohort of 623 healthy individuals aged >50 years in China
Article information
Abstract
Study Design
A retrospective radiologic study.
Purpose
To investigate the spinopelvic positional and morphological parameters of the sagittal spinal alignment in a healthy population from the community.
Overview of Literature
The existing parameters for spinal alignment based on the Cobb angle are the primary reference values for evaluating spinal alignment and pelvic morphology. However, they do not fully capture the comprehensive characteristics of sagittal spine alignment. More attention should be given to identifying the specific characteristics of sagittal spinal alignment by focusing on the positions of the kyphotic and lordotic apices.
Methods
Among 1,250 volunteers, 623 consecutive normal community volunteers aged >50 years were recruited and underwent standing postural X-ray. A customized computer application analyzed the sagittal morphological and positional parameters, examining their normal distributions and correlations.
Results
The correlation between the adjacent morphological and positional parameters was strong between the distal cervical and proximal lumbar spine. In the vertical direction, a significant association was found between the location of the thoracic kyphosis (TK) to lumbar lordosis (LL) transition point (TL point) and both the upper apex of TK (T-apex) (r=0.52) and lower apex of LL (L-apex) (r=0.64). In the horizontal direction, a moderate correlation was found between the thoracic apex offset to the femoral axis (TF) and the lumbar apex offset to the femoral axis (LF) (R2=0.314), whereas LF demonstrated a strong correlation with adjacent overhang (R2=0.685). Close correlations were observed among the morphological and positional parameters. The sacral slope exhibited significant correlations with two parameters related to the lumbar region: L-apex (r=−0.60) and LF (r=0.51).
Conclusions
This study found strong correlations between spinopelvic morphology and position, which is crucial for understanding sagittal alignment. Adjacent positional parameters showed significant compliance within the sagittal spine plane from the distal cervical to proximal lumbar regions, suggesting the necessity for additional research on its clinical relevance in spinal disease surgery.
Introduction
Several radiographic parameters have been developed as sagittal alignment guides [1–4]. The evaluation of regional alignment is crucial and can be determined by Cobb angles, which assess thoracic kyphosis (TK) and lumbar lordosis (LL). In addition, pelvic parameters such as the sacral slope (SS), pelvic tilt (PT), and pelvic incidence (PI) have been investigated to elucidate sagittal degeneration and compensatory mechanisms. Although current spinal alignment parameters based on Cobb measurements evaluate vertebral alignment and pelvic landmarks, spinal malalignment naturally triggers compensatory mechanisms to maintain a gravity line close to the coxofemoral axis sagittal coordinate [5]. The assessment of global spinal alignment commonly employs the sagittal vertical axis (SVA), a plumb line method drawn onto full-length radiographs [1,6,7]. The SVA is one of the prevalent sagittal parameters used for evaluating normal sagittal global balance in adults with an acceptable range of 0.5±2.5 cm [8–10].
The current sagittal parameters, encompassing TK, LL, and SS, are defined as the Cobb angles derived from the upper and lower endplates of the spinal anatomical segments. Recognizing that TK and LL are not circular but rather elliptical is crucial. This indicates that individuals may possess identical TK or LL while exhibiting variations in their sagittal profiles (Fig. 1). According to Tono et al. [11], even among cases with equivalent LL, the sagittal alignment can vary based on the L-apex positioning. Increased attention should be directed toward the locations of the kyphotic and lordotic apices and their inflection points as essential variables for identifying the specific characteristics of sagittal spinal alignment. Berthonnaud et al. [4] also described the TL point, which signifies a location within the spine where TK transitions into LL, thus determining the point at which the curvature direction changes and the lengths of kyphotic or lordotic vertebrae.

Illustration of subjects with identical thoracic kyphosis (TK) or lumbar lordosis (LL) values but different sagittal profiles. (A) Two sagittal spinal morphology had identical cervical lordosis (CL), T1-slope, TK, LL, sacral slope (SS), and pelvic incidence (PI). (B) Two above sagittal profiles exhibited variations when the positional parameters of T-apex, TL-point, and L-apex were altered. (C) The configuration of TK and LL represents a segmental section of an ellipse rather than a circular shape, with the apices of the spinal curve serving as markers to define the accurate arc for sagittal alignment.
The “cone of economy” should be balanced over a small area between the feet without external support and with minimal effort. This means that a freestanding individual must adjust pressure transmission and bending moments to sustain an optimal gravity line [12]. As individuals age, Schwab et al. [6] revealed an increase in TK and an anterior shift in SVA through radiologic analysis; however, the location of the whole body’s gravity remained constant relative to the subjects’ feet. Therefore, the femoral head can be a radiographic indicator for approximating the gravity line. In addition, Sebaaly et al. [13] reported that restoring the LL sagittal apex reduced the incidence of proximal junctional kyphosis from 13.5% to 41.4%. These findings indicate that when assessing the biomechanical balance in spinal degeneration progression among older individuals, one should consider both spine positions based on coxofemoral joints and morphology.
Therefore, this study aimed to investigate whether and how the combined spinopelvic positional parameters and morphological parameters affect the sagittal alignment of the spine in normal Chinese community dwellers aged >50 years. In addition, this study attempted to explore the relationship between positional and morphological parameters in the older population.
Materials and Methods
Patient selection
This cross-sectional study was approved by the relevant institutional ethics committee and conducted in accordance with the Declaration of Helsinki. The study protocol was reviewed and approved by the Institutional Review Board (IRB) of Xuanwu Hospital, Capital Medical University (IRB no., 2018014). All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee. All patients provided written informed consent to participate.
In total, 1,250 normal Chinese volunteers from a single institution in the Beijing community were recruited between 2021 and 2023. Before enrollment, the volunteers received a comprehensive explanation of the study’s objectives, methods, and potential risks and provided written informed consent. The inclusion criteria were as follows: (1) age ≥50 years; (2) absence of mechanical low back or neck pain that significantly affected work, daily activities, or leisure pursuits; and (3) no significant symptoms (Visual Analog Scale [VAS] <2; Oswestry Disability Index [ODI] <20). The exclusion criteria included the following: (1) history of spinal surgery; (2) history of spinal injury, infection, tumor, or deformity; (3) apparent radiographic abnormalities such as facet hypertrophy, spondylolysis, or scoliosis; and (4) incomplete or unclear radiological records.
Radiographic measurement
Each patient was instructed to stand upright but relaxed for the radiography of the spine. Under the supervision of a professional radiologist, the hands were placed on the clavicles and shoulders to maintain a horizontal field of view, and the hips and knees were held in extension. In our hospital, long-term X-rays of the spine were captured. The distance between the radiographic source and the film was kept constant for all exposures at 180 cm. A custom computer application (Picture Archiving and Communication System; GE Electrics, Boston, MA, USA) was used to measure the sagittal parameters. The radiographic parameters were divided into morphological and positional parameters (Fig. 2). In this study, the morphological parameters were set to the Cobb angle measurements to characterize the curvature of various spinal segments (Fig. 2A), and the positional parameters were set to the distance parameters based on the coordinate system with the center of the coxofemoral heads as the origin (Fig. 2B).

(A, B) Illustration of the spine-pelvis sagittal morphology and position parameters in the standing. CL, the angle between C2 and C7; T1_slope, the angle between the upper endplate of T1 and the horizontal line; TK, the angle between T4 and T12; T-apex, the apex of thoracic kyphosis; TL-point, the inflection point from thoracic kyphosis to lumbar lordosis; LL, the angle between L1 and S1; L-apex, the apex of lumbar lordosis; SS, the angle between the upper endplate of S1 and the horizontal line; PT, the angle between the line from the center of the femoral head to the upper endplate of S1 and the plumb line; PI, draw a vertical line through the midpoint of the upper endplate of S1. The angle between the vertical line and the midpoint of the upper endplate of S1 and the femoral head center. SVA, the horizontal distance between the C7 plum line and the posterosuperior corner of S1; TF, the horizontal distance between the apex of thoracic kyphosis and the bicoxo-femoral axis; TL-width, the horizontal distance between the apices of thoracic kyphosis and lumbar lordosis; LF, the horizontal distance between the apex of lumbar lordosis and the bicoxofemoral axis; LS1-width, the horizontal distance between the apex of lumbar lordosis and the posterosuperior corner of S1; Overhang, the horizontal distance between the center of the upper sacral endplate and the bicoxofemoral axis.
The morphological parameters included cervical lordosis (CL), T1 slope, TK, LL, SS, PT, and PI. The positional parameters were measured as follows (Fig. 2): SVA, apex of TK (T-apex), thoracic apex offset to the femoral axis (TF), TL point, LL apex (L-apex), lumbar apex offset to the femoral axis (LF), width thoracic and lumbar regions (TL_width), width of lumbar and sacral regions (LS1_width), and overhang. Adjacent sagittal parameters refer to morphological or positional parameters describing adjacent spinal segments. For example, CL, TK, and LL are adjacent morphological parameters used to describe the curvature of cervical vertebrae and thoracic and lumbar vertebrae. The measurements obtained from two experienced spine surgeon observers were used for each parameter of interest. The measurement concordance was excellent, with an interclass correlation coefficient >0.88, and an intraclass correlation coefficient >0.85.
Statistical analysis
Radiographic data were analyzed using IBM SPSS Statics for Windows ver. 19.0 (IBM Corp., Armonk, NY, USA). The normal distribution of the data was assessed using the Kolmogorov-Smirnov test to determine its statistical properties. To provide a summary measure for each variable, continuous variables were presented as means and standard deviations.
This study investigated the relationships between categorical and continuous variables by employing Spearman’s and Pearson’s correlation coefficients, which are commonly used in such analyses. Correlations were classified into different categories based on their strength: Correlations were categorized as follows: very strong (r=0.80–1.00), strong (r=0.60–0.79), moderate (r=0.40–0.59), weak (r=0.20–0.39), or no correlation (r=0.00–0.19). This study did not focus on cases with weak or no correlation. Only those with strong and moderate correlations (r≥0.40) between the parameters were analyzed. When a triangular relationship existed among the parameters, the partial correlation method was used to eliminate the control variable that might influence both variables and determine their direct correlation. Radiographic parameters with strong correlations were selected for simple linear regression analysis using a custom Python program. The level of significance was set at p<0.05.
Results
Sagittal parameters
A total of 1,250 normal older participants were enrolled; however, only 623 radiographs met the eligibility criteria for inclusion. The average age of the cohort was 66.2±6.0 years (range, 50.1–88.8 years), with a sex distribution of 69.8% female and 30.2% male. The radiographic results are illustrated in Figs. 3 and 4. All morphological and positional parameters in our cohort exhibited a normal distribution pattern. The CL, T1 slope, TK, LL, SS, PT, and PI values were 10.36°±10.16°, 23.61°±7.30°, 34.35°±9.01°, 50.31°±10.84°, 34.44°±7.90°, 15.05°±6.02°, and 49.48°±9.14°, respectively (Fig. 3). On average, the TT apex, TL point, and L-apex were located in the T7, T12, and L4 vertebrae, respectively (Fig. 4A–E). The SVA, TL_width, and LS1_width were 32.22±12.56 mm, 114.94±21.33 mm, −6.36±28.11 mm, respectively (Fig. 4F–I). TF, LF, and overhang values were 117.25±25.16 mm, −2.86±14.94 mm, and 45.22±7.72 mm, respectively (Fig. 4F–O).

(A–H) Histograms showing the normal distribution of the measured morphological parameters. SD, standard deviation; CL, cervical lordosis; TK, thoracic kyphosis; LL, lumbar lordosis; SS, sacral slope; PT, pelvic tilt; PI, pelvic incidence.
Correlation analysis
The relationships between the morphological and positional parameters are depicted in Fig. 5. In our cohort. most sagittal parameters exhibited varying degrees of correlation. The partial correlation analysis results revealed a direct correlation between the variables after excluding the influence of the control variables. Fig. 6 demonstrates a strong correlation (r=0.60–1.00, p<0.001). SVA demonstrated a moderate negative correlation with the TL_width (r=−0.49, p<0.001). Both the adjacent morphological and positional parameters showed strong correlations from the distal cervical to the proximal lumbar spine. There were moderate or strong correlations between CL and T1-slope (r=0.66, p<0.001), T1 slope and TK (r=0.56, p<0.001), TK and LL (r=0.57, p<0.001), LL and SS (r=0.87, p<0.001), SS and PI (r=0.76, p<0.001), and PI and PT (r=0.52, p<0.001). Similarly, TF moderately correlated with the LF (r=0.51, p<0.001), and LF had a strong correlation with the adjacent overhang (r=−0.81, p<0.001). Furthermore, the TL point was significantly linked to the upper T-apex (r=0.52, p<0.001) and lower L-apex (r=0.64, p<0.001).

Graphic depiction of the correlation coefficients between the sagittal parameters. SVA, sagittal vertical axis; CL, cervical lordosis; TK, thoracic kyphosis; LL, lumbar lordosis; SS, sacral slope; PI, pelvic incidence; PT, pelvic tilt; TL_width, width thoracic and lumbar regions; LS1_width, width of lumbar and sacral regions; TF, thoracic apex offset to the femoral axis; LF, lumbar apex offset to the femoral axis.

Graphic depiction of the strong correlation between the sagittal positional and morphological parameters. (A) Morphological parameters. (B) Positional parameters of the apex and its inflection. (C) Positional parameters of the horizontal distance between the apices and the posterosuperior corner of S1. (D) Partial correlation coefficients between the positional and morphological parameters. CL, cervical lordosis; TK, thoracic kyphosis; LL, lumbar lordosis; SS, sacral slope; PI, pelvic incidence; PT, pelvic tilt; TF, thoracic apex offset to the femoral axis; LF, lumbar apex offset to the femoral axis; SVA, sagittal vertical axis; TL_width, width thoracic and lumbar regions; LS1_width, width of lumbar and sacral regions.
In comparison with the morphological and positional parameters, TK exhibited a moderate correlation with TF (r=0.46, p<0.001), whereas SS showed moderate correlations with LF (r=0.51, p<0.001, and L-apex, r=−0.60, p<0.001). PI was moderately correlated with the LS1_width (r=0.47, p<0.001). Notably, overhang displayed a very strong positive correlation with PT (r=0.91, p<0.001).
Linear regression analysis
Linear regression modeling revealed a relationship between the SVA and TL_width (R2=0.226, p<0.001), TF and LF (R2=0.314, p<0.001), and LF and overhang (R2=0.685, p<0.001) (Fig. 7A–C). In addition, close correlations were found between the morphological and positional parameters: TK and TL width (R2=−0.428, p<0.001); LF and SS (R2=0.279, p<0.001), and overhang and PT (R2=0.939, p<0.001) (Fig. 8A–C). Furthermore, the LS1_width had a link with PI (R2=0.225, p<0.001) (Fig. 8D).

Scatter diagram on correlation between the sagittal position parameters. (A) Sagittal vertical axis (SVA) was negatively correlated with width thoracic and lumbar regions (TL-width). (B) Thoracic apex offset to the femoral axis (TF) was positively correlated with lumbar apex offset to the femoral axis (LF). (C) LF was negatively correlated with Overhang. ***p<0.001.

Scatter diagram on correlation between the sagittal positional and morphological parameters. (A) Thoracic kyphosis (TK) was positively correlated with width thoracic and lumbar regions (TL-width). (B) Sacral slope (SS) was positively correlated with lumbar apex offset to the femoral axis (LF). (C) Pelvic incidence (PI) (D) was positively correlated with width of lumbar and sacral regions (LS1_width). PT, pelvic tilt. ***p<0.001.
Discussion
Sagittal spine alignment is essential for understanding the biomechanics underlying spinal pathologies and effectively managing spinal disorders, particularly in older people. This study examined morphological and positional parameters of spinal alignment in a cohort of 623 healthy Chinese individuals from the community. A strong correlation was identified between the adjacent morphological parameters from the distal cervical to proximal lumbar regions (including CL, T1 slope, TK, LL, SS, PI, and PT) within this healthy older group. Our findings are consistent with the results of previous research investigating the relationships among thoracic, lumbar, and pelvic Cobb angle parameters [2,4,14,15]. In the present study, adjacent positional parameters demonstrated significant compliance within the sagittal plane of the spine from the distal cervical to proximal lumbar regions—similar to what was observed with morphological parameters. Furthermore, in our cohort, spinopelvic positional and morphological parameters were intricately linked and collectively defined sagittal alignment.
Recent research has emphasized that the success of spinal surgery is closely linked to the restored sagittal alignment and balance. The existing sagittal alignment parameters, including TK, LL, and SS, do not sufficiently capture the unique sagittal profile characteristics of individual spines. This is primarily due to the recognition that TK and LL are elliptical rather than circular. Consequently, individuals may present with identical TK or LL while demonstrating distinct differences in their sagittal profiles (Fig. 1). A study indicated variations in sagittal alignments based on the positioning of the L-apex, even among cases exhibiting comparable LL values [11]. Pan et al. [16] identified a correlation between the PI and the L-apex in asymptomatic adults, suggesting that higher PI values are linked to a more proximal positioning of the apex. In line with this, we did not find a strong relationship between PI and the L-apex (r=−0.58). The subanalysis revealed that when the L-apex did not align with its theoretical position, patients were more prone to experiencing significant sagittal imbalance [16,17]. Our results revealed a significant correlation between SS and two parameters (L-apex and LF) to determine the lumbar apex positioning. To address the limitations of existing sagittal spinal morphological parameters, positional parameters such as LL, position of the apex of TK, and its inflection point, is essential to accurately identify the specific characteristics of the sagittal spinal alignment and its balanced state, is essential to investigate.
The sagittal position of the apex of the spine has been identified as a crucial parameter influencing the morphological characteristics of the lumbar spine. In healthy adults, Roussouly et al. [3,7,18] established sagittal classification criteria based on SS and PI, each having distinct sagittal inflection points and lumbar apex positions. This study determined the kyphosis or lordosis position using the longitudinal level (T-apex and L-apex) based on the spinal vertebrae segments and the horizontal distances (TF and LF) to the coxofemoral heads. On average, the T-apex was located at the T7 vertebrae, TL point at the T12 vertebrae, and L-apex at the L4 vertebrae, consistent with previous studies [3,19]. The TL point indicated a strong positive correlation between the upper T-apex and lower L-apex. The findings suggest conformity in the adjacent positional parameters of the spine in the sagittal plane, which aligns with the observed morphological parameters. Moreover, the L-apex is critical in determining lumbar sagittal alignments associated with LL, PI, and SS. Adjusting the L-apex to its optimal position can help minimize mechanical complications following surgical intervention for adult spinal deformity [20]. In planning lumbar surgery, it is essential to ensure that the apex position of the LL conforms to its parameter range while restoring the required lordosis curvature.
Moreover, positional parameters such as TK and its inflection points further to accurately identify the specific characteristics of the spinal sagittal alignment in our investigate. In this study, the SVA correlated more closely with TL_width than all other parameters. The distance between the apex of TK and the apex of LL plays a significant role in assessing SVA. The linear relationship between the TL_width and SVA pre- and postsurgery must be considered to determine the optimal position of the LL apex during the restoration of sagittal balance in lumbar surgeries. In addition, a consistent PI was significantly and positively correlated with LS1_width, suggesting that the horizontal distance from the apex of LL to the posterosuperior edge of S1 should maintain a stable value aligned with PI. The stable LS1_width may provide a reference to identify the relative position of the sacrum and lumbar spine apex while correcting lumbar spine deformities. However, this study only provides preliminary insights into the effect of positional parameters on sagittal spinal alignment, highlighting the need for further investigations to comprehensively understand its clinical significance for the surgical management of adult spinal diseases.
In addition, positional parameters are the significance of sagittal balance, which entail achieving equilibrium between the responses of the trunk muscles and external forces, primarily gravity, acting on the spine [5,6]. Ideally, the gravity line should pass through the center of the coxofemoral heads in healthy adults and be positioned slightly anterior to the vertebral centers for each segment [21,22]. From the distal cervical to proximal lumbar spine segments, there should exist an excellent linear regression relationship between adjacent positional parameters: TF=−115.07+0.76×LF (R2=0.314) and LF=28.10−0.96×overhang (R2=0.685). Although not equivalent to the gravity line, the positional parameters of the kyphotic and lordotic apices may provide valuable insights into the local stress distribution in the sagittal plane. Surgical interventions should aim to alleviate discomfort and restore normal sagittal stability under gravitational influence, particularly for older individuals suffering from spinal degeneration and imbalance. Several studies have suggested that the evaluation of apexes and inflection points of reconstructed spinal curves should be included in surgical planning [23,24]. Thus, patients were more likely to maintain long-term sagittal balance when the theoretical positions and relationships of TF, TF, and overhang were aligned postsurgery. However, further academic research is required to elucidate this field.
Several limitations should be mentioned. First, the female-to-male ratio was approximately 3.2 because women were more willing to participate in the community flow study, and sex was not a screening criterion during enrollment. This study provided the normal reference values for spinal sagittal alignment in individuals aged ≥50 years. Owing to spatial constraints, a detailed examination of sex differences was not included; however, the parameter ranges for different sexes are provided in Supplementary Materials (Supplements 1, 2). Consequently, this study did not radiograph the entire body, including the lower limb. Subsequent studies must examine the compensatory effect of knee and ankle flexion on the overall spinal alignment. In addition, the study only considered the effects of sagittal parameters on sagittal spinal alignment; therefore, other tissues should be thoroughly investigated in subsequent studies. Despite the limitations, this study provides a new perspective for an in-depth understanding of sagittal balance and alignment compliance in functional spinal segments by combining morphological and positional parameters.
Conclusions
The sagittal alignment was characterized by the spine-pelvic morphological and positional parameters, which were closely related to normal Chinese elders. Our findings indicate that positional parameters are critical in determining the spine sagittal alignments associated with TK, LL, SS, and P. Adjusting the apex position of thoracic kyphosis can help restore the necessary spinal curvature and maintain long-term sagittal balance following surgical intervention. These results initially reveal the influence of positional parameters combined with morphological parameters on sagittal spinal alignment, and their clinical significance in the surgical treatment of spinal diseases requires further investigation.
Key Points
Reference values of the morphological and positional parameters of the spinal alignment were investigated in asymptomatic Chinese individuals from the community.
In healthy older individuals, morphological and positional parameters were in harmony from the distal cervical vertebrae to the proximal lumbar vertebrae, indicating a well-balanced sagittal plane.
Spinopelvic positional and morphological parameters were intricately linked and collectively defined as sagittal alignment in our cohort.
Notes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
A nonauthor writer who has contributed to the manuscript should be credited in an acknowledgment footnote.
Funding
This study was supported by Beijing Natural Science Foundation (7244356), National Natural Science Foundation of China (No.12402376), R&D Program of Beijing Municipal Education Commission (22JG0059), and the key Science and Technology Project of Beijing Municipal Education Commission (KZ20231002537).
Author Contributions
Conceptualization: DW. Methodology: FP. Data curation: CL. Investigation: WZ, YW. Visualization: SZ. Validation: XC. Supervision: SL. Writing–original draft: WW. Writing–review & editing: ZW. Final approval of the manuscript: all authors.
Supplementary Information
Supplementary materials can be available from https://doi.org/10.31616/asj.2024.0466.
Supplement 1. The average values of all measured sagittal parameters of the female.
Supplement 2. The average values of measured sagittal parameters of the male.
asj-2024-0466-Supplementary-Tables.pdf