Introduction
The sagittal spinal balance, as defined by Dubousset [1], represents the interaction of the neuro-musculoskeletal components of the spine, pelvis, and lower extremities with gravity. This interaction enables the maintenance of a position that requires minimal energy expenditure, a concept often described as the “cone of economy” [1–3]. Achieving this balance necessitates a so-called “balance chain” comprising the feet, pelvis (pelvic vertebra), and skull (cephalic vertebra). This balance chain is a series of correlations that assess the spinopelvic relationship and regional sagittal alignment in an upright position [4–6].
Legaye et al. [5] and Legaye et al. [6] initially explored the relationship among various spinal and pelvic regions, establishing connections between the positional and morphological parameters of the hip, sacrum, and pelvis. Later, a study by Boulay et al. [7] involving 149 asymptomatic Caucasian adults reaffirmed the correlation among the pelvic parameters: pelvic incidence (PI) and sacral slope (SS) (r=0.7), as well as pelvic tilt (PT) (r=0.6). Gelb et al. [8] studied 100 asymptomatic middle-aged and older volunteers, analyzing the role of segmental spinal alignment in overall sagittal balance. They noted that despite advancing age, sagittal alignment was preserved due to the loss of distal lumbar lordosis without an increase in thoracic and thoracolumbar kyphosis [8].
Clément et al. [9] first described the importance of thoracic kyphosis (TK) as a harmonious curve in maintaining sagittal spinal balance. Recent findings indicate that TK comprises two unequal arches: a stable upper TK and a variable lower TK, with the apex typically at T8–T9, varying according to the Roussouly classification-based spinal morphotype. These considerations are crucial for analyzing sagittal compensation and planning corrections to mitigate the risk of mechanical complications [10,11].
Researchers are currently working to complete the correlation chain by investigating the influence of the cervical segment on the other spinal segments. Ames et al. [12] found correlations between pelvic and regional sagittal parameters, showing that high PI is associated with high lumbar lordosis (LL), which in turn relates to TK, and TK correlates with cervical lordosis (CL). However, they found no direct influence of PI on CL or vice versa [12]. Nunez-Pereira et al. [13] found no direct correlation between PI and cervical parameters in their study of 145 patients. Similarly, Muñoz Montoya et al. [14], in a study of 51 asymptomatic Hispanic adults, found no direct relationship between pelvic and cervical parameters, except for a weak correlation between the T1 slope and the PI–LL mismatch.
Recently, new cervical sagittal parameters, such as the C7 and C2 slopes, have emerged, but their role in sagittal alignment remains unclear. Therefore, the objective of this study was to investigate whether the previously described pelvic parameters directly influence both these new and conventional cervical parameters or if it is necessary to correlate the thoracic parameters.
Materials and Methods
Study design and population
This was an observational study of asymptomatic healthy volunteers aged between 18 and 50 years. The study was approved by the Institutional Review Board (IRB) of Ganga Medical Center and Hospital, Coimbatore (IRB no., 2020/01/02). Written informed consent was obtained from all participants before enrolment. The exclusion criteria included the presence of any spinal deformities, hip or lower limb discrepancies or diseases, recent trauma, chronic pain conditions, or musculoskeletal pathologies.
Radiographic evaluation and classification of types
Participants underwent standardized whole-spine radiographs in lateral and anteroposterior views while standing with arms crossed over their chests. Digital images were recorded on a picture archiving and communication system in digital imaging and communications in medicine format and then uploaded in the designated software program to measure the following sagittal alignment parameters (Table 1): (1) Pelvic parameters: PI, SS, PT. (2) Lumbar parameters: LL, PI–LL mismatch. (3) Thoracic parameters: TK: T1–T12; upper thoracic kyphosis (UTK): T1–T5; lower thoracic kyphosis (LTK): T5–T12. (4) Cervicothoracic parameters: thoracic inlet angle (TIA), T1 slope (T1S), neck tilt (NT), T1 slope–CL mismatch. (5) Cervical parameters: upper cervical parameters: C0–C2 angle, C1–C2 angle, C2 slope; lower cervical parameters: CL: C2–C7, C7 slope, cervical sagittal vertical axis (cSVA) (C2–C7). (6) Horizontal gaze parameter: McGregor slope (McGS).
Two spine surgery residents conducted measurements using the Surgimap program (Nemaris Inc., New York, NY, USA), which has demonstrated superior repeatability and reproducibility compared to manual radiologic measurements. The Surgimap’s specialized sagittal alignment tool was used for these assessments (Fig. 1). The user first outlined the femoral heads with two adjustable circles and then marked four segments corresponding to four key vertebral endplates (superior S1, superior L1, superior T1, and inferior C2). Once the landmarks were identified, the user adjusted the constrained spines to correctly overlay the cervical, thoracic, and lumbar curvatures. The software then automatically generated spino-pelvic parameters based on the demarcated landmarks and splines: direct parameters corresponded to angles drawn by the observer (PI, PT, SS, LL, C2C7, T1 slope, cSVA, PI–LL), while indirect parameters corresponded to angles estimated by the software based on the splines.
Statistical analysis
Data analysis was conducted using IBM SPSS ver. 20.0 software (IBM Corp., Armonk, NY, USA), involving measures of central tendency for quantitative variables and proportions for qualitative variables. Differences between males and females were assessed using the independent t-test and Mann-Whitney U test for normally and non-normally distributed variables, respectively. Pearson’s correlation coefficient (r) was used to evaluate correlations, with strength of the correlations interpreted as very weak (0.0–0.1), weak (0.1–0.3), moderate (0.3–0.5), strong (0.5–0.7), very strong (0.7–0.9), and perfect (0.9–1.0). Intraclass correlation coefficients (ICCs) and their 95% confidence intervals (CIs) were calculated based on a mean rating (k=2), consistency, and two-way random effects for average measures to assess all sagittal parameters. ICC values were interpreted as poor (<0.50), moderate (0.50–0.74), good (0.75–0.90), and excellent (>0.90). The significance level was set at p<0.05.
Results
The study included 104 asymptomatic patients (62 female [59.6%]) with a mean age of 38.27±9.93 years. Table 2 presents the cervical and thoracic parameters expressed as mean, standard deviation, minimum, and maximum values. The mean values of various parameters were as follows: (1) Pelvic parameters: PI: 45.43°±9.717°, SS: 3.51°±7.968°. (2) Lumbar parameters: LL: −51.26°±12.456°. (3) Thoracic parameters: TK (T1–T12): 30.95°±10.928°, UTK (T1–T5): 11.133°±7.2546°. (4) Cervicothoracic parameters: TIA: 85.20°±10.646°, T1 slope: 26.26°±8.236°. (5) Cervical parameters: upper cervical parameters: C1–C2: −21.63°±11.381°, C2 slope: 2.31°±11.420°; lower cervical parameters: CL: C2–C7: −24.27°±14.902°, C7 slope: 23.0735°±8.34865°. (6) Horizontal gaze parameters: McGS: 2.118°±13.3506°. Age wise distribution of various important parameters has been mentioned in Supplement 1.
Intrarater and interrater reliability
The intrarater reliability was found to be excellent for all parameters (ICC, 0.96; 95% CI, 0.92–0.97). The interrater reliability between the two readers was also found to be excellent (ICC, 0.92; 95% CI, 0.86–0.95).
Correlation between pelvic and lumbar parameters
The correlations between pelvic and lumbar parameters are presented in Table 3. PI showed strong positive correlations with SS (r=0.602, p=0.000) and PT (r=0.613, p=0.000), and a moderate positive correlation with LL (r=−0.344, p=0.000). SS also showed a strong negative correlation with LL (r=−0.734, p=0.000). Conversely, PT showed a weak negative correlation with SS (r=−0.259, p=0.001) and a weak positive correlation with LL (r=0.312, p=0.001).
Correlation between lumbar and thoracic parameters
The correlations between lumbar and thoracic parameters are presented in Table 4. The pelvic parameters showed no direct correlation with the thoracic parameters. LL (L1–S1) showed a moderate negative correlation with TK (T1–T12) (r=−0.501, p=0.000) and LTK (T5–T12) (r=−0.499, p=0.000). Of note, LL (L1–L4) exhibited a moderate negative correlation with various thoracic segments (Table 4).
Correlation between cervical and thoracic parameters
The correlation matrix in Table 5 shows the relationships between the cervical and thoracic segmental parameters. TK (T1–T12) strongly correlated with T1S (r=0.628, p=0.000) and C7 slope (r=0.575, p=0.000). TK (T1–T12) also exhibited a weak correlation with TIA (r=0.294, p=0.002) and a negative moderate correlation with CL (C2–C7) (r=−0.321, p=0.000). Of note, TIA was the only cervical sagittal alignment parameter to correlate with all thoracic segments; the other parameters were T1S, which correlated with the LTK (T5–T12) and TK (T1–T12). Similar results were observed with the C7 slope (Table 5).
Correlation between spinopelvic and cervical parameters
Table 6 shows the relationships between spinopelvic and cervical parameters. Key findings included a weak negative correlation between PI and C7 slope (r=−0.212, p=0.05) and a weak correlation of PT with T1 slope−CL mismatch (r=−0.229, p=0.05) and C2 slope (r=−0.202, p=0.05). Additionally, PI–LL mismatch showed very weak correlations with TIA (r=−0.197, p=0.05), T1 slope (r=−0.228, p=0.05), and C7 slope (r=−0.251, p=0.05). No correlation were observed between other spinopelvic and cervical parameters (Table 6).
Discussion
The cervical spine can be divided into upper and lower cervical regions with distinct functions and responses to bodily changes. The primary roles of the cervical spine are maintaining horizontal gaze and supporting the weight of the head [12,15,16]. Recent studies have highlighted the distinction between cervical sagittal alignment and malalignment, significantly affecting patients’ quality of life [16,17]. This understanding stems from establishing normal values and efforts to standardize cervical sagittal alignment measurements, essential for planning cervical spine surgeries [18].
The PI–LL mismatch is a widely used parameter in the context of adult spinal deformity. It serves as a key objective for correcting the deformity, significantly impacting the patient’s quality of life. It is measured using a formula that categorizes results as either less than 10° or greater than 10°, illustrating the relationship between PI and LL, in addition to PT and SS [19]. While the influence of this parameter on cervical alignment is largely unknown, our study revealed a weak correlation between pelvic and cervical parameters. Specifically, we found connections between PI and the C7 slope, PT and T1 slope, CL mismatch, and C2 slope. Additionally, a weak inverse correlation was observed between PI–LL mismatch and TIA, T1S, and C7 slope. These findings align with Muñoz Montoya et al. [14], who also reported a relationship between PI–LL mismatch and T1S.
Lee et al. [20] in 2012 introduced several parameters like TIA, NT, and T1S, advancing the study of cervical sagittal alignment. These parameters have facilitated research into direct and indirect correlations between the cervical, thoracic, lumbar, and pelvic regions [20]. Current and previous research suggest a minimal or weak direct influence of pelvic parameters on cervical sagittal alignment, which does not follow the expected correlation chain [21].
The results of our study are consistent with those of Nunez-Pereira et al. [13], Lee et al. [22], and Shao et al. [23], who found very weak correlations between CL and LL (r=0.1, p<0.01), as well as between C0–C2 and sacral slope (SS) (r=−0.1, p<0.05) in asymptomatic healthy volunteers. Similarly, the study by Alijani and Rasoulian [21] among the Iranian population found a weak correlation between the C1–C2 angle and LL (r=0.1, p=0.01) in symptomatic individuals. Additionally, CL (C2–C7) showed weak correlations with PI (r=0.1, p=0.04) and PT (r=0.12, p=0.003) [21]. Our study also found weak negative correlations between PI and C7 slope, PT and T1 slope−CL mismatch, and PT and C2 slope (r=−0.202, p=0.05). Additionally, PI–LL mismatch showed weak correlations with TIA, T1S, and C7 slope. These findings highlight the importance of evaluating global spinal sagittal alignment as a whole, considering correlations between different regions, rather than in isolation or only among non-contiguous segments (Supplement 2A, B).
The chain of correlations indicates that increased CL leads to increased T1S, leading to an increase in TK and LL, all related to PI [24]. However, studies on cervical alignment, including pelvic and thoracolumbar alignment, showed no direct correlation between LL and CL, emphasizing the complex interconnections between cervical, thoracic, lumbar, and pelvic regions [25,26]. LL and PI directly influence TK, which in turn indirectly affects cervical sagittal alignment [22,27]. This suggests an interdependence among cervical parameters. However, the impact of LL and pelvic parameters on the cervical spine remains unclear [22,28].
The concept of the C7 slope, introduced by Le Hue et al. [18], is the foundation for understanding global cervical alignment. It is defined as the angle of the C7 vertebra, with values categorized as either less than or greater than 20° [18]. Another study utilized a T1S value greater than 26° as a threshold. However, due to difficulties in visualizing T1, the C7 slope is a more practical surrogate variable for studying cervical sagittal alignment [29]. The study by Nunez-Pereira et al. [13] identified the C7 slope as a crucial link between pelvic, spinal, and cervical parameters, showing it is an important indicator of global sagittal balance. Furthermore, their study underscored the direct relationship between the thoracolumbar and cervical spine. Notably, an altered C7 slope on cervical X-ray may signal potential global sagittal imbalance, warranting a full spinal evaluation, which can inform and refine the surgical plan [13].
The study’s limitations include its focus on asymptomatic patients within a specific age group, not accounting for cervical pathology, and not using advanced imaging techniques to rule out degenerative pathology (e.g., magnetic resonance imaging). The asymptomatic cohort limits the clinical significance of the findings, as the correlations might be more relevant in patients with cervical malalignment or degenerative conditions. Additionally, the study is constrained by the small sample size and retrospective design, highlighting the need for a larger prospective study to obtain more robust findings.
Conclusions
Our study highlights that spinopelvic and cervical alignment are significantly interdependent in asymptomatic individuals. Key spinopelvic parameters (PT, PI, and PI–LL mismatch) correlate with cervical parameters (C2 slope, C7 slope, T1 slope, and TIA), highlighting the interconnectedness of the pelvis, lumbar spine, thoracic spine, and cervical spine in determining global spinopelvic sagittal alignment.
Key Points
Cervical sagittal parameters significantly correlate with spinopelvic parameters.
Pelvic tilt, pelvic incidence, and pelvic incidence–lumbar lordosis mismatch exhibit significant correlation with C2 slope, C7 slope, T1 slope, and thoracic inlet angle.
The study highlights the interconnectedness between the pelvis, lumbar spine, thoracic spine, and cervical spine in determining global spinopelvic sagittal alignment.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
