Introduction

The axial skeleton protects our body during daily physical movement, including the vertebral bodies (VBs), intervertebral discs (IVDs), tendons/ligaments, and muscles. The intervertebral disc is a complex structure that acts as a fibrocartilage joint connecting adjacent VBs and is responsible for flexibility, multiaxial motion, and load transmission throughout the spine, which is essential for mechanical stabilization [1]. Lumbar lordosis refers to the curvature of the lumbar spine and is shown in humans as a response to bipedal walking [2]. Lumbar lordosis is formed by the wedging of the lumbar vertebra and intervertebral disc [3]. During development, the relative contribution of the vertebral body and disc to the overall shape of the lordotic angle changes significantly [4]. Some pathological conditions can cause changes in the intervertebral disc height [5].

It is important to determine the range of LLA normal values in the clinic. Ko, K, J et al. [6] believe that poor postures in adolescents can cause changes in lumbar curvature, and this diagnosis needs to refer to the range of normal values for adolescents. Sato et al. [7] found that there is a certain relationship between the severity of spinal injury and vertebral kinematics when driving a car. Car seat posture can affect the initial spinal alignment of car occupants. Appropriate adjustment of the tilt Angle of the seat back can change the sagittal alignment of the spine, which has a protective effect on the driver. Reference will be made to the normal range of spinal curvature angles.

At present, the study on the change rule of lumbar lordosis caused by walking activities and the contribution of intervertebral disc and vertebral body factors to it is still incomplete. Therefore, this study retrospectively collected spinal MRI data of 96 children to explore the growth and evolution of LLA before and after walking activities and the change rule of the influence of vertebral body and intervertebral disc on LLA. To provide systematic and reliable data for the development and diagnosis of lumbar spine in children.

Materials and methods

Materials

A cross-sectional study. A total of 96 normal developing children aged 0 to 10 years old who underwent lumbar MRI scans in the Imaging Department of Shandong Provincial Hospital from July 2021 to July 2023 were selected. Most of the cases were from trauma or routine examinations related to hospitalization. There were 55 boys and 41 girls aged 0 ~ 10 (5.7 ± 3.0) years. They were divided into 3 groups according to walking activity: 13 cases in the Pre-walking group (0–1 years old], 14 cases in the Walking group (1–3 years old], and 69 cases in the Post-walking group (3–10 years old].

Inclusion and exclusion criteria

Inclusion criteria: (1) Children 0 to 10 years old who were not found abnormal lumbosacral vertebra by spinal MRI; (2) No previous spinal surgery and surgical history. Exclusion criteria: (1) A family history of spine-related genetic diseases and other developmental abnormalities. (2) Poor image quality affects the result analysis.

Methods

1.5T MR Scanner (Siemens Avanto) was adopted, field strength was 45 mT·m− l·s− 1, and the receiving coil was selected as a 13-channel phased array coil. Supine position, head first. Scanning sequence: T2 weighted image (T2WI) with the following parameters: TE 91ms, TR 2600.0ms, reverse Angle is 142°, layer thickness is 2.5 mm, FOV read is 320 mm, FOV phase 100.0%, scanning time is 129s/ time, acquisition times are once, spatial resolution is 0.8 mm×0.8 mm×2.5 mm.

Image analysis and data measurement

The images were processed by Siemens Syngo. via workstation and the L1 and S1 vertebrae were determined on the median sagittal image. Parameter definition: (1) LLA, The Angle between the upper endplate of L1 and the upper endplate of S1; (2) ∑D, the sum of lumbar intervertebral disc wedging angles from L1 to L5; (3) ∑B, The sum of lumbar vertebra wedging angles from L1 to L5. The above angles are positive when pointing back and negative when pointing forward. The criteria for determining the upper and lower endplates of the vertebral body is the line between the most anterior point and the last marginal point of the endplate. (4) DH, The height between the midpoint of the upper and lower margins of the sagittal position, as shown in Fig. 1. All parameter measurements were performed by the same senior imaging physician. The same sequence of images was measured three times at the same level, and then measured again after 2 weeks. The final results were averaged.

Fig. 1
figure 1

The examined child is a boy aged 5 years old. Related parameters were measured based on sagittal T2WI images of lumbosacral segment of spine. (A) Lumbar lordotic Angle (LLA) is the Angle between the upper endplate of L1 vertebra and the upper endplate of the S1 vertebra. (B) Disc wedge Angle (DL) is the Angle between the lower endplate of the lumbar vertebral body and the upper endplate of the lower vertebral body.The vertebral wedge Angle (BL) is the Angle between the upper and lower endplates of the lumbar vertebra. (C) Disc height (DH) is the height between the midpoints of the upper and lower margins of the disc

Full size image

Statistical analysis

All data were processed and analyzed using IBM SPSS (IBM Corp, Armonk, NY, USA, version 25) software, and the Shapiro–Wilk test was used to test the normality of data. Normal data is expressed as mean ± standard deviation \(\stackrel{-}{(x}\)±s). Analysis of variance (ANOVA) and Bonferroni correction were used to compare the parameters among different groups. Independent sample t-test was used to compare the genders of different groups, classified data use case (%), and χ2 test was used for inter-group comparison. Pearson or Spearman correlation analysis was used, and the correlation coefficient (r/rs) > 0.75 was a strong correlation, 0.4–0.75 was a moderate correlation, and < 0.4 was a low correlation. P < 0.05 was considered to be statistically significant.

Results

Parameter description of 96 normal children aged 0 to 10 years

See Table 1.

Table 1 Parameter description of children aged 0–10 years
Full size table

Comparison of parameters between different groups

The LLA and DH values of children in different groups were significantly different (P < 0.05). According to Bonferroni’s correction, there were statistically significant differences in LLA values between the Pre-walking group and the Post-walking group (P = 0.001), and there were statistically significant differences in DH values between the three groups (P < 0.001), as shown in Table 2.

Table 2 Comparison of parameters of children in different groups
Full size table

The comparison of the gender differences within different groups and the differences among groups of the same sex

In the Post-walking group, the differences in LLA, DHL2 − 3, DHL3 − 4, and DHL4 − 5 in different genders were statistically significant(P = 0.031, 0.040). Comparison between groups of girls: The difference in DHL1 − 2 between the Pre-walking group and the Post-walking group was statistically significant (P < 0.001); There were significant differences in DHL2 − 3 and DHL3 − 4 values among the three groups (all P < 0.05); The differences of DHL4 − 5 and DHL5-S1 between the Post-walking group and the Pre-walking group and between the Pre-walking group and the walking group were statistically significant (all P < 0.05). Comparison between boys groups: There were significant differences in DHL1 − 2, DHL3 − 4, and DHL5-S1 between the Post-walking group and the Walking group and between the Pre-walking group and the Post-walking group (P < 0.05); There were significant differences in DHL2 − 3 and DHL4 − 5 values among the three groups (all P < 0.05), as shown in Table 3; Fig. 2.

Table 3 The comparison of the gender differences within different groups and the differences among groups of the same sex
Full size table
Fig. 2
figure 2

The differences in lumbar disc height (DH) among groups of the same sex. The figure shows the results of ANOVA with a Bonferroni correction

Full size image

The correlation between parameters and age and between parameters was analyzed

Age had a low positive correlation with LLA and ∑D and a moderate to strong positive correlation with DH; LLA showed a moderate positive correlation with ∑D, and a low positive correlation with ∑B and DH. ∑D is negatively correlated with ∑B. ∑B showed a low positive correlation with DHL1 − 2 and DHL2 − 3. There was a strong positive correlation between DH, as shown in Table 4.

Table 4 The correlation between parameters and age and between parameters
Full size table

Discussion

In this paper, the data of sagittal plane LLA, ∑D, ∑B, and DH before and after walking were retrospectively measured by MRI. On lateral radiographs, measuring the Cobb Angle is considered the gold standard in radiological evaluation of lumbar lordosis, and the Cobb Angle of L1-S1 is superior to MATLAB software, showing the highest diagnostic accuracy [8]. In this study, the Cobb Angle of L1-S1 with gold standard was adopted, and it was finally concluded that the LLA value was 33.2°±8.7°, ∑D value was 14.1°±8.6°, ∑B value was 11.9°±8.6°, and DH showed an upward trend in L1-L2-L3-L5-L4 in normal children aged 0 to 10 years. That is, DHL4 − 5 is the thickest lumbar disc. Under ideal conditions, the LLA value =∑D+∑B, but the LLA value in this study is about 7.2° larger than (∑D+∑B), which may be due to the small vertebral wedge Angle and disc wedge Angle, and manual measurement will cause certain errors. Mekhael et al. [9] found that the average LLA value was 60° (29 ± 7 years old), and Park et al. [10] found that the LLA values were 53.8 ± 9.5° (< 50 years old), 50 ± 10.7° (50–59 years old), and 48.5 ± 13.0° (≥ 60 years old), respectively, revealing that LLA peaked at a certain age. Then there was a decreasing trend with the increase in age, and the peak age was preliminarily determined to be about 20 years old. This study confirmed that the contribution of ∑D to LLA in children aged 0–10 years was about 54%, and Kalichman et al. [3] found that the contribution of ∑D to LLA in children aged 40–49 years increased to 85%, which also indicated that the contribution of ∑D to LLA would increase significantly with age.

We concluded that there were statistically significant differences in LLA and DH values among different groups (P < 0.05), and LLA in the Post-walking group was significantly higher than that in the Pre-walking group, with a difference (95%CI) of 9.2° (3.1°,15.3°). The Pre-walking group was less than 1 year old and basically could not walk. The walking activity was stable and the activity intensity increased in the Post-walking group, which indicated that walking activity could promote the increase of lumbar lordosis and thus maintain body balance. This study found that the contribution of ∑D to LLA increased slightly from the Pre-walking group to the Post-walking group, but the difference was not statistically significant. In this study, the contribution of ∑B was slightly greater than that of ∑D in the pre-walking group (0–1 years old), which is consistent with Williams et al. [4] that the ability to bear significant vertebral wedging during early development explains the large lumbar lordosis that humans can achieve. Shefi et al. [11] found that the contribution of ∑D to LLA increased from 47 to 91% in children aged 2–4 years old to adults aged 17–60 years old, and 53% in children aged 3–10 in this study, revealing that with the increase of age, the contribution of the vertebral body (∑B) to lumbar lordosis gradually decreased, while the contribution of intervertebral disc (∑D) gradually increased. DH in the three groups showed an increasing trend in the order of L1-L2-L3-L5-L4, and the average value of DH in the latter group was higher than that in the adjacent former group, respectively: 1.05 mm (L1) -1.35 mm(L2)-1.55 mm(L3)-1.55 mm(L5)-1.75 mm(L4), indicating that DHL4 − 5 was the thickest lumbar intervertebral disc with the fastest growth among groups. The average increase of five DH within the groups was 0.175 mm (Pre-walking group), 0.425 mm (Walking group) and 0.45 mm (Post-walking group), which revealed that the average growth rate of lumbar intervertebral disc was the fastest in the Post-walking group.

In the study, the LLA value of girls was significantly higher than that of boys in the Post-walking group(P = 0.008), and the difference (95%CI) was 5.1° (1.4° ~ 8.9°), which was consistent with previous studies [12, 13]. The gender difference in lumbar lordosis may be related to the need for females to compensate for the increased obstetric load on both feet during pregnancy [2, 14]. However, Ma H et al. [15] did not find gender differences in LLA. The effect of gender on lumbosacral curvature has not been uniformly concluded, possibly because lumbar lordosis is influenced by ethnicity [16]. Aoki et al. [17] found that the LLA value in women > 80 years old was 32.7°±13.5°, which was significantly lower than the LLA value in < 59 years old, which was 38.3°±9.8°, and the LLA value of girls aged 3–10 years in this study was 38.1°±7.9°, which also increased the evidence of the loss of lumbar lordotic Angle in women with age. In this study, DHL3 − 4 and DHL4 − 5 in boys were significantly larger than those in girls, with a difference (95%CI) of 0.6 (0.06,1.2) and 0.6 (0.03,1.2), respectively. L3-L4 was located in the anterior part of lumbar lordosis and had greater motor flexibility. It is speculated that the gender difference in disc thickness may be related to higher height and stronger back muscle strength in boys than in girls. Abnormal disc thickness also indicates the presence of certain lesions, such as the disc height in patients with degenerative spondylolisthesis being lower than that in the normal population [5]. Figure 2 shows that most of the peak DH values in girls are located in the Walking group, while most of the peak values in boys are located in the Post-walking group, which indicates that the growth rate of the intervertebral disc in boys is slower than that in girls, but the DH value is always greater than that in girls.

In this study, there were low positive correlations between age and LLA, age and ∑D (rs=0.337, 0.221,P = 0.001, 0.030), and moderate to strong positive correlations between age and DH (rs=0.635–0.803, P < 0.001). The correlation between LLA and ∑D was moderately positive (rs =0.599, P < 0.001), and the correlation between LLA and ∑B, LLA and DH was low positive (r = 0.277–0.362, P < 0.05). ∑D and ∑B showed a low negative correlation (r=-0.312, P = 0.002). There was a high positive correlation within DH (r = 0.774–0.941, P < 0.001). Bailey et al. [18] concluded that LLA was significantly positively correlated with age (2–9 years old), which was consistent with the conclusion of this study, indicating that age has a certain impact on the LLA value. The correlation between LLA and ∑D is stronger than that between LLA and ∑B, indicating that the influence of ∑D on LLA of children is greater than that of ∑B [11]. DHL4 − 5 was the thickest lumbar intervertebral disc and showed a strong positive correlation with age (rs=0.803,P < 0.001), suggesting that DHL4-5 thickens significantly with age. In the sagittal position, L4 was located at the front of the spine, and DHL4-5 showed a low positive correlation with LLA (r = 0.362, P < 0.001), showing the highest correlation among the five DHs. It is speculated that the L4 vertebral body has a larger vertebral body and disc wedge Angle, which has a more significant impact on LLA than the rest of the vertebral body.

There are some limitations in this study. First, it is a retrospective study, which inevitably leads to selection bias in case selection, and the data collection methods and measurement methods in various studies are not completely uniform. Secondly, the influence of each lumbar vertebra and intervertebral disc factor on LLA was not evaluated separately. Finally, the sample data of children were limited and the possible effects of BMI on lumbar curvature were not included. Future studies will continue to increase the sample size and expand the age range, refine the specific effects of the single vertebral body and intervertebral disc wedging Angle on LLA, improve image resolution, and reduce measurement errors.

Conclusion

The contribution of vertebral body and intervertebral disc wedging to lumbar lordosis and the change of disc thickness before and after walking were evaluated on T2WI. We confirmed that walking activity significantly increased LLA, that age was the main influencing factor for disc thickening, and the contribution of the vertebral wedge before walking was slightly greater than that of the intervertebral disc wedge.