How does the sagittal spinal balance of the scoliotic population deviate from the asymptomatic population?

Recently, the study of sagittal spinal balance in different populations has gained importance ever since Roussouly et al. [1] proposed a four-type classification in a study of 160 normal European adults. This novel classification is mainly determined by sacral and lumbar sagittal parameters such as sacral slope (SS), lumbar lordosis (LL), the number of lumbar vertebrae in LL, inflection point and lumbar tilt (Fig. 1). However, there is a lack of literature that comprehensively analyses and compares the deviation of sagittal standing posture patterns in the scoliotic population with their normal counterparts, though previous publications have studied the difference of some sagittal spinopelvic parameters between the two populations [2, 3]. Upasani et al. [2] compared the measurements of sagittal spinopelvic parameters between scoliotic and asymptomatic adolescent subjects and found that scoliotic patients had greater pelvic incidence (PI) and pelvic tilt (PT) values than the asymptomatic population. They also found that patients with primary thoracic scoliosis had a lower thoracic kyphosis (TK) and a larger SS than the controls, whereas patients with a primary thoracolumbar scoliosis had relatively normal TK, LL and SS values. Qiu et al. [3] compared the values of sagittal spinopelvic parameters between adolescent girls with idiopathic thoracic scoliosis and asymptomatic girls and found that TK in the former group was lower, which supported the findings of Upasani et al. [2]. Other radiographic studies in this field also reported similar results [4, 5]. However, there is no published comparative study that specifically explored the difference of the whole sagittal spinal alignments and the distribution of standing posture patterns in scoliotic patients from the norm.

Hu et al. [6] reported that 42.4% of scoliotic adolescents belonged to Roussouly Type 3. This proportion is considerably high when compared with that of the general population reported by previous publications [1, 7‐9]. Thus, a question was raised regarding how the sagittal standing posture patterns of the scoliotic population deviated from the norm. Aiming to illuminate the above question, this study was designed to describe and compare the whole sagittal spinal balance and the distribution of their patterns in asymptomatic and scoliotic young adults.

In all, 126 asymptomatic and 117 IS subjects were recruited, with an average age of 26.4 ± 3.7 and 26.3 ± 4.7 years, respectively (Table 1). The Cobb angles of coronal curves for IS adults are presented in Table 2. All sagittal parameters were normally distributed, and the reliability (R values) for all parameters was above 96%, ranging from 96.88% to 99.98%. The PI for the normal and IS groups was 41.2° ± 7.1° and 45.5° ± 10.0°, respectively. The IS group had larger PI, PT and SS but a smaller ST than the asymptomatic group (P < 0.05), while other sagittal parameters were similar (P > 0.05, Table 1). Though ST was significantly different between the two genders in the normal group (P < 0.05), other parameters were similar between males and females in either group (P > 0.05, Fig. 4).

Normal and IS groups had a comparable distribution of Roussouly classification (P > 0.05), and Type 3 was the largest for both groups, found in 49.2% and 45.3% cases, respectively (Table 1). Though the distribution of Roussouly types was similar between the two normal genders, it was different between IS males and females (P < 0.05), with more females possessing a well-balanced sagittal standing posture, namely, Type 3 (Table 3).

In all, 24.8% of IS adults belonged to the C7-anterior subgroup compared to 4.8% of normal adults (P < 0.05, Table 1). Males and females in both groups had a similar proportion of C7-anterior subgroup (P > 0.05, Table 3).

Roussouly classification has become a valuable approach for characterizing sagittal standing posture patterns. This classification system can outline four distinct types of sagittal profiles, and each type dominates in an age range, predisposes to different spinal pathologies and even indicates for a certain surgical strategy [1, 8‐10, 12, 13]. Thus, it is beneficial for spinal surgeons to be aware of the difference of sagittal spinal balance between normal and IS populations. Since the values of sagittal parameters are affected by many factors, recruiting strategies must be meticulously planned to maximally maintain consistency of these factors, such as age and the stance during filming. The two groups in this study had similar age modalities (Table 1). Moreover, this study seemed immune to imaging-relevant influence, since the aforementioned radiographic protocol has been consistently conducted in our hospital.

This study found that the values of some sagittal parameters in IS patients deviated from the norm (Table 1). Three pelvic parameters, PI, PT and SS, significantly increased in IS patients. PI is an important pelvic parameter that can represent the potential of pelvic compensation for the spinal malalignment caused by some spinal pathologies [5, 12, 14‐16]. Therefore, the increased PI and PT suggested a larger retroversion of the pelvis and thus demonstrated the existence of pelvic compensation. The pelvic compensation could influence the spinal alignment, considering the excellent linear correlation between PI and LL in the normal population, which was simply formulated by the eq. LL ≈ PI + 9° (±9°) [17‐19]. This biomechanical compensation was basically effective, which was proven by the maintenance of the distribution of Roussouly types in the IS population compared with that of the norm (Table 1).

There are few previous articles exploring the sagittal standing postures in a scoliotic population. Yu et al. [20] reported that 42.5% of the Caucasian IS population belonged to Roussouly Type 3, but his study recruited both adolescents and young adults. Therefore, the result might deviate from either the exact proportion found in adolescent IS or adult IS, considering that some sagittal parameters, including PI, PT and LL, increase with age in the growing child [21, 22]. Hu et al. [6] reported that the proportion of Roussouly Type 3 in scoliotic adolescents was 42.4%, which is lower compared to the proportion of 45.3% in scoliotic adults found in the present study (Table 1). We speculate that this difference is age-associated. As a child ages, his/her pelvis gains pelvic retroversion, which can provide a higher compensatory capacity and prevents forward displacement of the gravity axis in the case of a spinal deformity such as scoliosis. Therefore, the proportion of the well-balanced sagittal standing posture, namely, Roussouly Type 3, is maintained.

This study showed IS adults had a higher SS than normal adults (Table 1). According to Roussouly’s reports [1], higher SS in IS adults should be accompanied with higher proportions of Roussouly Type 3 and Type 4 (Fig. 1). However, this study displayed a similar distribution of Roussouly types between IS and normal populations (Table 1). We speculated that this similarity was related to less coherence of lumbar alignment, another important determinant of Roussouly classification, to the increased PI and SS in the IS population [5, 15]. Another underlying reason in addition to PI-based regulation might be the complex inter-dependence of spinal three-dimensional malalignments in IS, which was presumably associated with the joint effects of “anterior overgrowth”, vertebral rotation and spinal buckling [2, 23, 24]. Therefore, the transition of the whole spine-pelvis complex in IS is more complex than that of the normal population. In other words, higher SS is not always accompanied with higher proportions of Roussouly Type 3 and Type 4 in the IS population but is comparable or has slightly lower proportions, as shown by the findings of this study (Table 1) given the joint effects of the aforementioned factors. However, the above conclusion is not definitive and needs further study for verification.

Asymptomatic males and females demonstrated a similar distribution of Roussouly types, whereas a significant difference existed between IS males and females (Table 3). In all, 55.1% of female IS subjects belonged to Type 3, which was higher than that of IS males. IS females were prone to standing neutrally, which was somehow unexpected since female IS patients tend to have an unfavourable prognosis [5]. Whether this divergence indicated any clinical advantage for IS females could not be arbitrarily asserted, especially in the absence of longitudinal follow-ups. This study also found that normal females had increased absolute values for all parameters except TK compared with normal males, although without statistical significance, corresponding to the previous studies in normal populations [4, 21, 22]. This means that normal females tended to possess prominent lordosis and/or kyphosis, which are usually seen in Roussouly Type 3 and Type 4. This might explain the finding of higher proportions of Type 3 and Type 4 in normal females than in males, which were 68.2% versus 58.4%, respectively (Table 3). However, this was not the case for IS males and females, in whom the total proportions of Type 3 and Type 4 were quite close (64.4% versus 62.3%), although IS females were predominantly Roussouly Type 3 (Table 3). This phenomenon might be related to the aforementioned ill-matched transition of spine-pelvis complex in IS adults. Correspondingly, we found that the divergence of sagittal parameters was more trivial between the two IS genders (Fig. 4), but this finding was not unique [6‐8]. Therefore, the sagittal balance in the IS population was more complex and affected by the two related factors, namely, the inherent malalignment of this entity and the regulation of the PI-based system.

Mac-Thiong et al. [4] stated that the relative position of C7 plumbline with respect to the sacrum and HA was an important indicator for categorizing the global sagittal balance (Fig. 3). Generally, the forward displacement of the C7 plumbline was associated with a higher risk of developing spinal pathology [4, 9, 21]. This study found that 24.8% of the IS population stood with forward displacement of the C7 plumbline, which was significantly higher compared to 4.8% in asymptomatic adults (Table 1). In other words, the anterior shifting of the whole spine was more common in the IS population. This conclusion was reinforced by the finding of a lower ST in the IS population (Table 1). In addition, the finding that the proportion of the C7-anterior subgroup was similar between males and females in both groups implied that the occurrence of the forward displacement of the C7 plumbline might not be correlated with gender.

In summary, this study was primarily dedicated to defining the deviation of sagittal standing postures in an IS population from the norm and found that there was no apparent correlation between specific Roussouly types and IS. However, there were some limitations. First, the source of the controls was restricted and was not from general society (see “Materials”), and these subjects had higher homogeneity in habits, lifestyle, etc., which might compromise the reliability of this study. Second, coronal deformities may affect the sagittal spinal profile, as previously reported by Hu et al. [6]. Thus, correction surgery has to restore both coronal and sagittal alignments. For example, low TK values occur in IS patients with thoracic curves, and so an effective correction of coronal curves should be accompanied with good restoration of TK. Secondary adaptation and adjustment of LL as well as pelvic sagittal parameters are then performed, and the overall sagittal balance is regained. The present study, however, mainly focused on the research of the deviation of the whole sagittal spinal balance in IS patients from the norm and did not include research and discussion regarding the effect of coronal deformity on the sagittal spinal profile.