Introduction
Recently, much work has been done to elucidate the global sagittal postural patterns of different populations [
123456]. Roussouly et al. [
1] employed a four-type classification in a study of 160 asymptomatic European adults, of which type 3 was perceived as a well-balanced neutral position, and the other three types were non-neutral. The classification was mainly based on sacral and lumbar alignments such as sacral slope (SS), lumbar lordosis (LL), numbers of lumbar vertebrae in lordosis, inflection point and lumbar tilt (
Fig. 1). Roussouly et al. [
1] found that only 37.5% of asymptomatic Caucasian adults were of type 3. Afterwards, Mac-Thiong et al. [
2] characterized the global sagittal balance in a larger cohort of asymptomatic Caucasian adults, but using a new six-type classification system, which was based on the relative position of C7 plumbline with respect to the midpoint of upper sacral endplate and hip axis (HA, the midpoint of the line connecting bicoxofemoral centers), and found that 14.2% of asymptomatic Caucasian adults stood with C7 plumbline ahead of HA and midpoint of superior endplate of S1.
By categorizing the global sagittal patterns according to Roussouly classification, Hu et al. [
6] found that 42.4% of the scoliotic Chinese adolescents belonged to type 3, in a study among 184 patients with adolescent idiopathic scoliosis. However, there has been a lack of studies exploring the global sagittal postural patterns in asymptomatic Chinese adults, though Zhu et al. [
7] once characterized the norms of each sagittal parameter without the involvement of the global sagittal patterns. Thus, we designed this prospective imaging study, aiming to characterize the global sagittal postures of asymptomatic Chinese adults, and meanwhile, compared it with the Caucasian counterparts [
1,
2].
Results
No difference was detected in average age between the two genders (
p=0.917). There were significant differences in the values of LL, PI, SSA and SVA between males and females (
p<0.05), but no differences in TK, SS and PT (
p>0.05) (
Table 2).
From the Chinese subjects, 47.8% belonged to Roussouly type 3 (
Table 3). There were also 63, 38 and 41 subjects in type 1, 2 and 4 groups, respectively. Roussouly type 3 subjects were 44.7% of males and 52.3% of females (
Table 3). No difference was found between males and females in the distribution of Roussouly types (
p=0.130) (
Table 3), or in the proportions of the neutral postures (
p=0.222) (
Table 4).
All sagittal parameters except PT were significantly different among four Roussouly types (
p<0.05). LL, SS, PI and SSA parameters were different for all the Roussouly groups with the exception of Roussouly type 1 and type 2. There were no significant differences in PT between any two different Roussouly types. There were also none in SVA except between type 1 and type 4 groups (
p<0.05) (
Table 5).
Besides the Roussouly classification, Mac-Thiong's classification was also applied and 4.4% of the Chinese subjects belonged to the C7-anterior subgroup (
Table 6).
Discussion
This prospective study focused on the categorization of the global sagittal postures of asymptomatic Chinese adults in order to provide some references for future studies. For a comparison between the Caucasian and Chinese ethnicities, we deliberately adopted a radiographic protocol as closely matched to that of Roussouly's as possible (see "Imaging protocols"). In addition, this study bifurcated the cohort according to gender in order to reveal the divergence in the distribution of the sagittal postural patterns between males and females. To the best of our knowledge, this study was the first to thoroughly explore the distribution of the global sagittal postural patterns in Chinese adults and compare it with that with the Caucasian counterparts.
In total, 272 asymptomatic adult were enrolled, whose average PI value was 46.4°±9.6° (
Table 1), lower than that of Caucasian adults (51.7° to 55.0°) [
128]. According to various publications, this phenomenon might be related to the smaller average body size of Chinese population than that of the Caucasian adults [
79]. Our study found that 47.8% of asymptomatic Chinese adults possessed a neutral sagittal posture, namely Roussouly type 3, which was higher than 37.5% of Caucasian adults, as reported by Roussouly et al. [
1]. As we adopted an identical imaging protocol reported by Roussouly et al. [
1], this difference between the two ethnics could not be attributed to nuances of imaging protocols and measurements. It implies that Chinese adults might have a lower susceptibility to clinical entities like low back pain, according to the previous studies on the relationship between Roussouly types and low back pain [
410].
In our study of asymptomatic individuals, females had higher SVA and SSA values than males (
Table 2) as females had a spine more prone to a backward drift, compared with males, and this was reflected by a higher proportion of Roussouly type 3 and type 4's in females (
Table 3). According to Araujo et al. [
4] study, increased SVA in Roussouly type 4 were associated with pain and decreased quality of life measures. This finding implied that females were at a higher risk to develop certain spinal pathologies. However, our study also found that Chinese females possessed a favorable distribution of neutral sagittal postural pattern (
Table 4), a pattern that is inversely related to low back pain as demonstrated by Chaleat-Valayer et al.'s [
10] study. To resolve these conflicting claims, future studies may have to revisit the topic.
Nearly all the sagittal parameters were significantly different between Roussouly types except PT (
Table 6), and this is a reflection of the effectiveness of Roussouly classification on sagittal alignments. However we observed that in addition to PT, SVA was similar between any two Roussouly types except between type 1 and type 4 groups (
Table 6). According to a previous publication [
4], these two parameters, PT and SVA, have a stronger association with quality of life measures than other parameters. As Roussouly's classification cannot stratify PT and SVA, it implies a limited applicability of this metric for quality of life measures.
Within the Mac-Thiong's six-type classification [
2], the first three types were defined as the midpoint of upper sacral endplate behind HA, whereas type 4, 5, and 6 were defined as the midpoint ahead of HA. Type 1 and 4 referred to subjects with C7 plumbline behind both the midpoint and HA, type 2 and 5 between the midpoint and HA, and type 3 and 6 ahead of both the midpoint and HA. Unfortunately, the classification does not address which types are the neutral or balanced global sagittal alignments. The subjects of our study were divided into six types accordingly, and the result was different from that of Caucasian adults [
2]. There were 78.7% of Chinese adults in type 1 versus 55% for Caucasian adults. Since there have been no risk or correlation analyses regarding the relationship of this classification with clinical pathologies and/or future prognoses, this distributive difference between the two ethnicities could not provide for a direct clinical conclusion on any posture benefits for sagittal balance. Nevertheless, the forward displacement of C7 plumbline relative to sacrum may be thought of an increased risk in developing a spinal pathology [
24]. Therefore with 4.4% of Chinese adults standing with the forward displacement of C7 plumbline compared with 14.2% of Caucasian adults [
2], it might indicate a lower risk of spinal pathology for asymptomatic Chinese adults. To verify this hypothesis, it is necessary to design an appropriate correlation study with risk analysis and longitudinal follow-ups.
There were some limitations in our prospective study. First, the recruitment bias seemed inevitable, considering the geographical origin of the subjects and the younger mean age of the enrolled subjects in this study, which could possibly compromise the validity of the comparison between the two ethnicities. Our cohort, however, had a comparable subject source to that of Roussouly's [
1] study, and it partially justifies a cross comparison analysis of the two ethnicities. Finally, as this was not a confirmatory study, any clinical interpretations need to be supported by future studies.