Radiological Analysis of Sagittal and Cross-sectional Morphology of Congenital Lumbar Spinal Stenosis

Background Purpose This retrospective study was applied to investigate the morphology characteristics of the spine and pelvis in patients with congenital spinal stenosis, to explore the effect of morphological parameters in the pathogenesis and development of the disease. Methods The analysis is based on data of a case - control study, including 40 patients (19 females/21 males) with congenital lumbar spinal stenosis, 40 patients (17females/23males) with age - 、 Sex - and the waist and leg pain score - matched acquired lumbar spinal stenosis and 40 age - 、 Sex - matched normal volunteers(controls). Lumbar MRI, lumbar computerized tomography （ CT ） and full - length radiographs were used to obtain sagittal and cross - sectional parameters. Parameters including pelvic incidence （ PI ） , sacral slope (SS), pelvic tilt (PT), lumbar lordosis (LL), sagittal vertical axis （ SVA ） , and thoracic kyphosis （ TK ） on the sagittal plane were measured on full - length radiographs and analyzed. The anteroposterior (AP) bone canal diameter and spinal canal area of L4 were collected and analyzed on Lumbar CT. Lumbar MRI was taken to evaluate the angle of the ligamentum flavum at the level of L4/5 intervertebral space ， and the Lumbar disc degeneration degree was calculated. Results the acquired group (p = 0.041 ＜ 0.05). There is no statistically significant difference in other parameters.The correlations between LL and PI are well in the congenital group （ r=0.336 ； P=0.034 ） , acquired group （ r=0.464 ； P=0.003 ） and control group （ r=0.584 ； P=0.000 ） . However, the trend line of LL/PI in the acquired group was drawn below the control population. Also, the trend line of LL/PI in the congenital group was below the waist and leg pain score - matched acquired group with lower lumbar degeneration. In addition to bony structural stenosis, the smaller angle of the ligamentum flavum may be an anatomical factor that causes the smaller effective area of the spinal canal in patients with congenital lumbar spinal stenosis. Patients with congenital lumbar spinal stenosis show a significant reduction in the physiological curvature of the thoracic and lumbar spine, and the trunk leans forward. In addition to intervertebral disc degeneration and pain factors, bony spinal stenosis is also a possible factor leading to smaller LL in patients with congenital lumbar spinal stenosis. LL less than 41 ° can be used as the initial screening standard for congenital lumbar spinal stenosis among patients with lumbar spinal stenosis. acquired group was higher than that of the normal control group( 39.28mm±36.57mm VS 7.60mm±24.99mm P=0.002 ).Other spinal and pelvic parameters were not statistically significant. This shows that in most cases of acquired lumbar spinal stenosis, the backward rotation of the pelvis and TK reduction （ Thoracic spine straightened ） are not enough to compensate for the loss of LL.

the acquired group (p = 0.041＜0.05). There is no statistically significant difference in other parameters.The correlations between LL and PI are well in the congenital group(r=0.336；P=0.034), acquired group(r=0.464； P=0.003) and control group(r=0.584；P=0.000). However, the trend line of LL/PI in the acquired group was drawn below the control population. Also, the trend line of LL/PI in the congenital group was below the waist and leg pain score-matched acquired group with lower lumbar degeneration.
Conclusion In addition to bony structural stenosis, the smaller angle of the ligamentum flavum may be an anatomical factor that causes the smaller effective area of the spinal canal in patients with congenital lumbar spinal stenosis. Patients with congenital lumbar spinal stenosis show a significant reduction in the physiological curvature of the thoracic and lumbar spine, and the trunk leans forward. In addition to intervertebral disc degeneration and pain factors, bony spinal stenosis is also a possible factor leading to smaller LL in patients with congenital lumbar spinal stenosis. LL less than 41° can be used as the initial screening standard for congenital lumbar spinal stenosis among patients with lumbar spinal stenosis.

Keywords Congenital lumbar spinal stenosis ； Spine and pelvis parameters；Sagittal plane； cross-sectional morphological Background
Lumbar spinal stenosis is one of the most common diseases which needs surgical intervention in the elderly, and is also one of the most likely causes of low back pain in patients aged over 50 1

Visual Analogue Scale (VAS)
The pain is represented by 11 numbers from 0 to 10, 0 means no pain, and 10 means the most pain. The higher the score, the stronger the pain. It was recommended to use the following cut-off points on painful VAS: ≤2, 3-5, 6-8, 8-10，Representing no pain, mild pain, moderate pain, and severe pain. According to the individual's pain experience, the subjects pointed out the degree of pain that suits them on the ruler at the time of admission.

Radiological assessment
All subjects' full-length radiographs were taken in the neutral position at admission and were measured the following indicators. Two independent measurements of the above sagittal spine-pelvic parameters were conducted by the same researcher, and the mean of parameters was calculated.

Spine parameters(Fig1
A)：All the spine parameters are masured as described below 16,17 .

Patients with congenital lumbar spinal stenosis have a more upright spine and a more forward lean in the sagittal plane.(Fig3)
Comparing the general indices between these three groups, no significant differences in age, gender were found. It showed that patients were matched in age, the gender ratio in these three groups. Also, there was no significant difference in the VAS scores of lower limb pain and low back pain between the development group and the acquired group total scores for lumbar disc degeneration and the angle of the ligamentum flavum were significantly lower in the congenital group than in the acquired group(P=0.020 and P=0.012，respectively).
Through CT measurement of the lumbar spine, we can find that the Spinal canal area and anteroposterior (AP) bone canal diameter of L4 in the congenital group are smaller than the acquired group, and the difference is statistically significant.(P=0.000 ，all)( Table 1) The values of LL, PT, SVA, and TK were significantly different, while the SS and PI were not among the acquired, congenital, and control groups. (Table 2) A post hoc test (Table 3) showed that TK was significantly lower in the congenital group than in the acquired and control groups(P=0.024 and P=0.006 respectively＜0.05).
Both the congenital and acquired groups had significantly lower LL than the control group (p = 0.000 and 0.041, respectively＜0.05). Also, LL was significantly lower in the congenital group than in the acquired groups(p=0.000

Bony spinal stenosis leads to lower LL in patients with lumbar spinal stenosis.
We speculated that bony spinal canal size such as spinal canal area and anteroposterior (AP) bone canal diameter may be a factor affecting the LL of patients with lumbar spinal stenosis (congenital lumbar spinal stenosis and acquired lumbar spinal stenosis). Therefore we conducted bivariate correlation and linear regression models to clarify the relationships between LL of the 80 patients with lumbar spinal stenosis and spinal canal area, as well as the relationships between LL of patients with lumbar spinal stenosis and anteroposterior(AP) bone canal diameter.

LL less than 41° can be used as the initial screening standard for congenital lumbar spinal stenosis
Because the LL of patients with congenital lumbar spinal stenosis was smaller than that of acquired lumbar spinal stenosis, and the difference was statistically significant (P＜0.05). Therefore, we drew the ROC curve for LL and determined the optimal critical value of 41° according to the maximum Youden index. The area of the ROC curve line is 0.735. (Fig5) Youden index is 0.55. According to this standard, parallel diagnostic experiments were performed. When the LL is less than 41° on the whole spine radiograph, the sensitivity is 90% and the specificity is 65%. LL less than 41° can be used as the initial screening standard for congenital lumbar spinal stenosis among patients with lumbar spinal stenosis. There are also studies showing that lumbar isthmic spondylolisthesis is complicated by lower PI 22,23 .All these studies showed that sagittal parameters were involved in the pathogenesis of lumbar degenerative diseases.PI was first proposed by Legaye et al. It can objectively reflect the anatomical positional relationship between the upper endplate of S1 and the femoral head, and reach a stable state in adulthood without being affected by subjective symptoms and changes in body position 9 .Previous studies have shown that the shape of the spine in low PI and low SS tends to be more linear, which leads to an increase in the vertical stress of the lumbar intervertebral disc, which is likely to cause early disc degeneration and herniation. Therefore, low PI may be a factor leading to lumbar disc herniation. At the same time, PI is the main risk factor for adolescent lumbar spondylolisthesis in Chinese Han adolescents. High PI means high risk 10,23 .In this study, the PI value of the congenital group was 48.40°±9.07°, the PI of the normal control group was 49.08°±11.76°, and the PI of the acquired group was 45.08°±9.17°.

Discussion
There was no significant difference between the three groups (P=0.168>0.05). It shows that the anatomical positional relationship between the upper endplate of S1 and the femoral head of patients with lumbar spinal stenosis is similar to that of normal people.
Many authors have been revealed that the sagittal profile was characterized by lower LL and TK, lower SS, and greater SVA( a straight spine, vertical sacrum and an anterior displaced C7 plumb line )for patients with lumbar disc herniation.

21,24-27
The results of this study showed that the average TK (  .This sagittal plane change is a combined effect of degeneration and pain compensation. The first is the degeneration of the lumbar intervertebral disc and the decrease in the height of the intervertebral disc leads to a decrease in LL. The onset of congenital lumbar spinal stenosis is the combined effect of lumbar disc degeneration and bony spinal stenosis. Under the condition that the bony area of the spinal canal is constant, the patient can further reduce the LL to expand the length of the spinal canal to achieve a larger effective volume of the spinal canal, so as to relieve neurological symptoms such as low back pain 21,25 .In addition, PI is also an innate anatomical factor that affects LL. A lower PI means a lower LL. The decrease in LL is followed by an increase in SVA, which means that the body leans forward. The result is an unstable center of gravity and a sagging field of vision. For this reason, the body will reduce TK and increase the cervical lordosis angle and bend the hips and knees to maintain a level of vision and a stable center of gravity.

Conclusion
In addition to bony structural stenosis, the small included angle of the ligamentum flavum may be an anatomical factor that causes the small effective area of the spinal canal in patients with congenital lumbar spinal stenosis. Compared with patients with acquired lumbar spinal stenosis and normal people, patients with congenital lumbar spinal stenosis show a significant reduction in the physiological curvature of the thoracic and lumbar spine, and the trunk leans forward. In addition to intervertebral disc degeneration and pain factors, the degree of spinal stenosis is also an important factor leading to smaller LL in patients with congenital lumbar spinal stenosis. The reason may be that the narrower the spinal canal, the stronger the pain sensitivity, and the more eager to relieve pain. But it may also be because patients with congenital lumbar spinal stenosis have a smaller LL before the onset.
This needs to be further confirmed by relevant prospective studies. In addition, we define a cut-off value based on the characteristics that patients with congenital lumbar spinal stenosis are generally smaller than those with acquired lumbar spinal stenosis. A LL which is less than 41° can be used as the initial screening standard for congenital lumbar spinal stenosis among patients with lumbar spinal stenosis.

Availability of data and materials
The data analysed during the current study are available from the corresponding author on reasonable request.

Figure3
Comparison of sagittal plane morphology of the CG, AG, NG groups. Arranged from left to right are the whole spine X-rays of patients with congenital lumbar spinal stenosis, acquired lumbar spinal stenosis and normal people. Compared with acquired lumbar spinal stenosis patients group and normal people group, the spine of patients with congenital lumbar spinal stenosis is more vertical and the trunk is more forward.

Figure4
Correlation between PI and LL of normal group (NG， red label) , acquired group (AG， blue label) and congenital group (CG， green label) .The red line, blue line, and green line increase sequentially, which means that the LL/PI trend line of the control group, acquired group, and control group gradually rises.

Figure5
ROC curve of the LL for preliminary screening of patients with congenital lumbar spinal stenosis. The area under the curve (AUC) is :0.735. ROC: receiver operating characteristic. Table captions   Table 1： Normal information in the CG, AG, and NG. Table 2 Comparisons of spinopelvic parameters in the CG, AG, and NG.    Comparison of sagittal plane morphology of the CG, AG, NG groups. Arranged from left to right are the whole spine X-rays of patients with congenital lumbar spinal stenosis, acquired lumbar spinal stenosis and normal people. Compared with acquired lumbar spinal stenosis patients group and normal people group, the spine of patients with congenital lumbar spinal stenosis is more vertical and the trunk is more forward.

Figure 4
Correlation between PI and LL of normal groupNGred label, acquired groupAGblue labeland congenital group CGgreen label.The red line, blue line, and green line increase sequentially, which means that the LL/PI trend line of the control group, acquired group, and control group gradually rises.