Lumbar X-rays are usually preferred in patients with lower back pain, but lumbar spinal stenosis (LSS) cannot be directly observed on lumbar X-ray films.
OBJECTIVE:
The purpose of this study is to explore the correlation between the degree of single-segment central LSS and lumbar X-ray measurements.
METHODS:
The data of 60 male patients aged 39–78 years with single-segment central LSS were analyzed. Linear correlation analysis was used to determine the correlation between the single-segment central LSS and the various measurement parameters. Multiple linear regression analysis was used to analyze the factors affecting single-segment central LSS.
RESULTS:
There were significant differences in S/S, E, B, LCobb, and M among the three groups ( 0.05). S/S was positively correlated with E, B, LCobb, and M ( 0.05), but was not correlated with D ( 0.66). After multiple linear regression analysis, B, LCobb, and M were independently associated with S/S.
CONCLUSIONS:
The B, LCobb, and M parameters were independently associated with single-stage central LSS, and would likely be of particular value in evaluating the degree of single-segment central LSS; B, LCobb, and M served as independent predictors of the degree of LSS. These findings will guide clinicians’ decision-making in the future.
Lumbar spinal stenosis (LSS) is one of the most prevalent lumbar degenerative diseases [1, 2], is a common cause of lower back pain and radiating lower extremity pain, and is the main cause of lumbar surgery in patients aged over 65 years [3]. Symptomatic LSS is often described as neurogenic claudication and is characterized by pain, weakness, numbness, and/or fatigue arising from the back and radiating to the buttocks, thighs, or calves [4]. Diagnostic results are often inconsistent due to a lack of reliable clinical or X-ray criteria [5, 6, 7, 8, 9, 10]. Follow-up studies showed that according to the natural history of LSS, early interventions for mild and moderate cases would not significantly aggravate the clinical symptoms [11]. Early diagnosis and intervention are particularly important. At present, the clinical diagnosis of LSS depends on magnetic resonance imaging (MRI) and computed tomography (CT). However, MRI is time intensive and CT exposes patients to ionizing radiation [12, 13]. Moreover, they are not the first choice for initial diagnosis and are unsuitable for patients with early lower back pain. Lumbar X-rays are often used as a basic imaging screening test for patients with lower back pain. However, X-ray films cannot directly confirm narrowing of the spinal canal. Amundsen et al. [14] reported the changes of sagittal diameter of spinal canal in flexion and extension X-ray films of patients with LSS, and found obvious pain in extension position, but did not explore the diagnostic value of X-ray film-related indicators for LSS. If the initial screening of LSS can be carried out through X-ray film, it can simplify the treatment process of a large number of patients with low back pain, which is beneficial to the early detection and treatment of LSS. In this study, our aim was to investigate whether or not plain film radiographs can be used to assist in confirming the presence of central canal stenosis.
Material and methods
Study population
The 60 patients enrolled in this study were males aged 39–78 years (mean age, 59.18 10.06 years), who were treated in the Department of Spinal Surgery of Lanzhou University Second Hospital from June 2017 to May 2019. The sample size required for our study is 20–40 cases. Our study was completed with 60 cases, which meets the statistical requirements [15, 16]. The inclusion criteria were as follows: (1) imaging (MRI/CT) examination showed single-stage central LSS; (2) clinical neurological symptoms and signs of LSS, and finally undergoing surgical treatment; and (3) complete clinical and imaging data. The exclusion criteria were: (1) spinal stenosis caused by trauma, tumor, etc.; (2) spinal stenosis secondary to inflammatory lesions (tuberculosis, brucellosis, etc.); (3) cervical and/or thoracic spinal stenosis; (4) multi-stage LSS; (5) incomplete clinical and imaging data; and (6) simple lumbar disc herniation. The time from onset to diagnosis and treatment in the 60 patients was 0.33–96.0 months (mean, 25.8 20.9 months). There were 32 cases of L4/L5 stenosis and 28 of L5/S1 stenosis. The severity of LSS was evaluated according to the preoperative Japanese Orthopaedic Association (JOA) score.
Grouping
The proportion of the dural sac undergoing compression was calculated on MRI images, acquired with the patient placed in the sagittal position (E) as the sagittal diameter of the compressed dural sac L/the sagittal diameter of the spinal canal L 100% (Fig. 1). The 60 patients were divided into three groups according to the value of E: E 66% for group I (20 cases), 33% E 66% for group II (27 cases), and E 33% for group III (13 cases). There were no significant differences in age or disease duration among the three groups ( 0.05). The JOA scores of the preoperative group III were lower than those of groups I and II without significance ( 0.05; Table 1).
General data of the three patient groups
Group
Number
Age (years)
Course (months)
JOA score
I
20
62.10 9.56
21.99 17.67
13.35 2.59a
II
27
57.07 9.82
32.29 24.06
12.96 2.12a
III
13
59.08 10.95
18.46 15.51
9.15 1.86
F
0.922
2.541
19.36
0.24
0.08
0.05
Analysis of variance. Group I: spinal cord compression ratio (E) 66%; Group II: 33% E 66%; Group III: E 33%; JOA: Japanese Orthopaedic Association; SNK- test vs. Group III, 0.05.
A 63-year-old male patient with an L4/5 lesion: lumbar magnetic resonance imaging (MRI) sagittal T2-weighted image of the dural sac. (A) shows the sagittal diameter (L) of the spinal canal, and (B) shows the sagittal diameter (L) of the dural sac in the most affected section.
MRI scan and ratio calculation
A GE 1.5T MR scanner (GE Healthcare, USA) was used to scan the intervertebral space. The patient was placed in the supine position and the scanning angle was adjusted to be parallel to the intervertebral space. Each intervertebral space was scanned in two to four layers around the intervertebral disc. The thickness and spacing of the scanning layers were both 2 mm. The image showing the most severe level of compression on MRI cross-section was selected, and a picture archiving and communication system (PACS; Shenzhen, China) was used to demarcate the cross-section of the cauda equina and the effective spinal canal. The irregular area measurement function of the system was then used to measure the cross-sectional area (S) of the dural sac and the effective lumbar spinal canal area (S) accommodating the spinal cord (or cauda equina). The ratio (S/S) of these values to a certain extent reflects the severity of LSS (Fig. 2).
A 63-year-old male patient with L4/5 single-stage central lumbar spinal stenosis (LSS). (A) The white dotted line denotes the cross-sectional area of the dural sac (S); (B) The white dotted line denotes the effective area of the spinal canal (S).
X-ray film measurement
X-ray measurements were taken of each subject. The centerline of the X-ray machine aligned with L4/5. The projection distance was 1.0 m, and the lateral radiographs were taken with the subject in positions of lateral overextension and hyperflexion, as well as a natural erect. The average height of the intervertebral space (D) is given as the mean value [D (D1 D2)/2] of the leading edge height (D) and the trailing edge height (D) of the superior vertebral intervertebral space on the lateral radiograph. The area of the bilateral intervertebral foramen (M) is the stenotic intervertebral foramen measured by PACS on the natural erect lateral radiograph. The L1–5 Cobb angle (LCobb) was that between the parallel line formed by the end plate of the L1 vertebral body and the vertical line of the L5 vertebral body perpendicular to this. The measurement method for B was the same as that for LCobb, i.e., the difference between the Cobb angle value and the hyperflexible Cobb angle value (B ) (Fig. 3).
(A) D is the height of the anterior border of the vertebral body, D is the height of the posterior margin of the vertebral body (D D D/2); (B) is the area of the intervertebral foramen of L4/5 (M); (C) The LCobb angle; (D) is the overextension position LCobb angle; (E) is the overflexion position L Cobb angle.
The measurement indices (D, M, B, LCobb) we selected can be measured on lumbar X-ray films. The relationship between D and M and LSS is easy to understand. When more advanced degenerative changes in the spine occurs, the disc becomes more narrow (D value decreases) [17], and soft tissues such as ligamentum flavum and posterior longitudinal ligament fold, resulting in LSS, and accordingly the area of intervertebral foramen (M) decreases. We chose to measure the difference between the Cobb angle and the Cobb angle in the flexion position of the lumbar spine (B ), considering that instability caused by degeneration of lumbar spine can lead to wider range of motion of lumbar spine, and instability of lumbar spine can easily lead to the growth of soft tissue such as ligamentum flavum, and then cause LSS. The Cobb angle (LCobb) of the lumbar spine in the natural upright position was chosen to take into account that the LSS patients suffer from back pain, because LSS patients suffer from back pain primarily in the upright (lordotic) position.
Observation indicators by lesion site (mean standard deviation)
Lesion site
Number
E (%)
S1/S0 (%)
B ()
D (mm)
L1–5 Cobb ()
M (mm)
L4/5
32
55.06 16.83
73.69 14.32
11.38 2.66
9.34 2.64
19.97 5.14
111.80 16.36
L5/S1
28
53.31 20.14
72.93 16.72
11.29 2.68
9.04 2.58
19.46 6.35
112.39 19.02
t
0.367
0.191
0.119
0.444
0.340
0.128
0.367
0.71
0.84
0.90
0.65
0.73
0.89
0.71
Student’s test. E: ratio of compression of the dural sac; S/S: ratio of the cross-sectional area of the dural sac and the effective cross-sectional area of the spinal canal; B: difference between the overextension and hyperflexion L1–5 Cobb angles; D: mean height of the anterior and posterior margin of the intervertebral space on natural erect lateral radiographs; M: area of the intervertebral foramen.
Observation indicators of the three patient groups
Group
Number
E (%)
S1/S0 (%)
B ()
D (mm)
L1–5 Cobb ()
M (mm)
I
20
73.85 4.48
88.35 6.73
13.74 1.97
10.13 2.35
25.50 3.38
128.83 12.74
II
27
49.49 8.95
73.76 4.89
10.85 1.89
8.50 2.46
19.37 2.32
109.81 6.65
III
13
30.96 17.5
73.33 15.36
8.64 1.61
8.39 2.71
11.62 1.66
91.00 13.89
F
63.300
176.222
31.089
3.404
111.362
50.111
0.001
0.001
0.001
0.04
0.001
0.001
Analysis of variance. Group I: E 66%; Group II: 33% E 66%; Group III: E 33%; SNK- test vs. Group I, 0.01; and vs. Group II, 0.01 (D: SNK- test, no significant difference among groups); E: ratio of compression of the dural sac; S/S: ratio of the cross-sectional area of the dural sac and the effective cross-sectional area of the spinal canal; B: difference between the overextension and hyperflexion L1–5 Cobb angles; D: mean height of the anterior and posterior margin of the intervertebral space on natural erect lateral radiographs; M: area of the intervertebral foramen.
Measurement parameters and assessment criteria
The E, S/S, B, D, LCobb, and M values were recorded for all three groups, and the clinical indexes of each group were subject to statistical analysis. Linear regression analysis was used to investigate the correlation between the S/S values and E, B, LCobb, and M. A multiple linear regression equation with B, LCobb, and M as the independent variables and S/S as the dependent variable was established.
Statistical analysis
The data were processed using SPSS statistical software (ver. 22.0; IBM Corp., Armonk, NY, USA). Variance, test, and multivariate linear regression were performed for data analysis. A value 0.05 was considered statistically significant.
Results
There were no significant differences in measurement parameters by lesion site
There were no significant differences in E, S/S, B, D, LCobb, or M by lesion site ( 0.05; Table 2).
Measurement parameters were significantly different among subgroups
Except for D, the measurement parameters (S/S, B, M, and LCobb) were significantly different among groups I, II, and III, ( 0.05; Table 3).
E, B, LCobb, and M were positively correlated with S/S
There were no correlations between age (Y) and duration (C) and S/S, in any group (R 0.032, 0.81; and R 0. 017, 0.90, respectively). This suggests that these two indicators do not modulate the relationships between the degree of LSS and the other measurement parameters. Lumbar X-ray measurements of E, B, LCobb, and M were positively correlated with S/S(R 0.762, 0.00; R 0.688, 0.00; R 0.867, 0.00; and R 0.720; 0.00, respectively). However, there was no correlation of S/S with the mean height of the anterior and posterior margin of the intervertebral space (D) (R 0.056, 0.66) (Fig. 4).
Multiple linear regression analysis of the parameters of interest
Parameter
Non-standardized coefficient
Standardized coefficient
B (X)
1.028
0.177
2.229
0.03
LCobb (X)
1.654
0.613
6.913
0.00
M (X)
0.183
0.208
2.529
0.01
X, difference between the overextended and flexion L1–5 Cobb angle; X, natural upright posture LCobb; X, area of the intervertebral foramen in the natural upright posture.
Correlation between lumbar X-ray film measurements and the cross-sectional area of the cauda equina/effective spinal canal area (S/S. (A) The proportion of the dural sac undergoing compression (E) is positively correlated with S/S; (B) The overextension position L1–5 Cobb angle (B) is positively correlated with S/S; (C) The natural upright position LCobb angle is positively correlated with S/S; (D) M in the natural erect posture is positively correlated with S/S.
B, LCobb, and M independently influence S/
Using S/S as the dependent variable, multiple linear regression analysis with B, D, LCobb, and M as the independent variables showed that B (X), LCobb (X), and M (X) were independently associated with S/S. The resulting regression equation was y 0.138 1.028X 1.654X 0.183X (complex correlation coefficient, R 0.896; adjustment coefficient, R 0.796), and the regression model was statistically significant (F 76.076, 0.05; Table 4).
Discussion
LSS is a common senile lumbar degenerative disease and its definitive diagnosis relies on CT/MRI. To the best of our knowledge, no previous study has explored the relationship between LSS measurements obtained in different positions and LSS severity. We analyzed the preoperative radiographic measurements of 60 male patients with single-stage central LSS. We found that S/S, E, B, LCobb, and M were positively correlated (Fig. 4).
Patients with LSS often show hypertrophy of the ligamentum flavum, bone hyperplasia, ossification of the posterior longitudinal ligament, superior facet hypertrophy, disc bulging and pedicular kinking, which can cause direct compression of the dural sac [18]. Loss of intervertebral space and lumbar ligament relaxation during lumbar degeneration, as well as a rapid increase in the small joint load, can easily lead to facet joint damage [19, 20, 21]. Takashima et al. [22] demonstrated that the small joints are important for lumbar stability. In addition, paravertebral muscles play an important role in maintaining lumbar stability [23]. When the lumbar spine is in a state of degeneration, the paravertebral muscles also change. Fortin et al. [24] found that the cross-sectional area of the multifidus and erector spinae gradually decreased with age, while fat infiltration increased and muscle composition changed. Yarjanian et al. [25] found that the cross-sectional area of the lumbar dorsal extensors gradually decreased, relative to normal people, in lower back pain patients, and even more so in LSS patients. Patients with spinal stenosis were more likely to exhibit atrophy of the dorsal extensors of the lumbar spine than those with lower back pain. The multifidus changes contribute to a loss of segmental stability, and change lumbar stability to segmental stability throughout, which may in turn be responsible for changes in B and LCobb. Scale et al. posited that increased instability in lumbar degenerative patients will lead to a cyclical angular change in the lumbar spine [26], similar to our results. The positive correlation between M and S/S in cases with lateral stenosis of the lumbar spine may be related to an increase in lumbar motion and bone degeneration. The detailed mechanism underlying this relationship remains to be determined.
The ligamentum flavum is one of the tissues adjacent to the spinal cord and nerve. It consists of elastic fibers, collagen fibers, reticular fibers and matrix, of which elastic fibers and collagen fibers account for 80% and 20% respectively [27]. Collagen fiber ensures the stability of ligament and elastic fiber ensures the elasticity of ligament. Both of them can maintain the stability of lumbar spine and are the key to the physiological function of ligamentum flavum. When the lumbar spine degenerates and stabilizes, the ligamentum flavum gradually proliferates and thickens, reduces elasticity, calcification and ossification [16]. During extension, the ligamentum flavum folds and protrudes into the spinal canal, compressing the dural sac of the spinal canal, resulting in spinal canal stenosis and increasing S/S, which is one of the main reasons for the compression of the spinal cord or corresponding nerve roots. Usually, the low back pain that follows is taken as the initial symptom, and fatigue, pain, numbness and weakness of the lower limbs will occur over time [4].
Lumbar X-rays are the most commonly used screening tests in spinal surgery clinics. The information provided by X-ray films is important for the diagnosis and treatment of lumbar diseases. In this study, we used multiple linear regression analysis to determine that B, LCobb, and M were independently associated with S/S (Table 4). This indicates that the degree of stenosis of the single-stage central lumbar spinal canal can be estimated and evaluated by measuring B, LCobb, and M, where higher values thereof increase the likelihood (or severity) of spinal stenosis. There was no significant difference in the measurement parameters by lesion site (Table 2), which indicates that the above is applicable to the prediction of spinal stenosis.
Narrowing of the intervertebral space is common in lumbar degeneration [18, 20, 28]. Surprisingly, our correlation analysis showed that D was not significantly related to the degree of spinal stenosis (R 0.056, 0.66). Moreover, there was no significant difference in the average height of the intervertebral space among the groups (Table 3). No study has been published on the relationship between narrowing of the intervertebral space and spinal stenosis; thus, this relationship requires further study.
In this study, we focused on male patients with single-stage central LSS, and therefore limited the scope of clinical guidance, which is also a limitation of this study. Besides, we only studied clinically symptomatic, pre-operative patients, with an average age of 57–62 years between the three groups. The application to pre-clinical patients is unknown. Further research requires attention to multiple stages LSS, female patients and pre-clinical patients of different ages in order to make the findings more generalizable.
Conclusions
In single-stage central LSS patients with lumbar hyperextension and hyperflexion, B, LCobb, and M were independently associated with the degree of LSS. Therefore, measuring these three indicators could be useful to evaluate single-stage central canals severity. The degree of stenosis has diagnostic and predictive value, which could guide clinical decision-making.
Footnotes
Conflict of interest
The authors have no conflict of interest to report.
References
1.
HaigAJTomkinsCC. Diagnosis and management of lumbar spinal stenosis. JAMA.2010; 303: 71-2.
WeinsteinJNTostesonTDLurieJDTostesonANBloodEHanscomB, et al. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med.2008; 358(8): 794-810.
4.
BarzTMellohMLordSJKaschRMerkHRStaubLP. A conceptual model of compensation/decompensation in lumbar segmental instability. Med Hypotheses.2014; 83(3): 312-6.
5.
AmundsenTWeberHLilleåsFNordalHJAbdelnoorMMagnaesB. Lumbar spinal stenosis. Clinical and radiologic features. Spine (Phila Pa 1976).1995; 20(10): 1178-86.
6.
BarzTMellohMStaubLRoederCLangeJSmiszekFG, et al. The diagnostic value of a treadmill test in predicting lumbar spinal stenosis. Eur Spine J.2008; 17(5): 686-90.
7.
ChiodoAHaigAJYamakawaKSQuintDTongHChoksiVR. Magnetic resonance imaging vs. electrodiagnostic root compromise in lumbar spinal stenosis: a masked controlled study. Am J Phys Med Rehabil.2008; 87(10): 789-97.
8.
KannoHOzawaHKoizumiYMorozumiNAizawaTKusakabeT, et al. Dynamic change of dural sac cross-sectional area in axial loaded magnetic resonance imaging correlates with the severity of clinical symptoms in patients with lumbar spinal canal stenosis. Spine (Phila Pa 1976).2012; 37(3): 207-13.
9.
KonnoSHayashinoYFukuharaSKikuchiSKanedaKSeichiA, et al. Development of a clinical diagnosis support tool to identify patients with lumbar spinal stenosis. Eur Spine J.2007; 16(11): 1951-7.
10.
SirvanciMBhatiaMGaniyusufogluKADuranCTezerMOzturkC, et al. Degenerative lumbar spinal stenosis: correlation with oswestry disability index and MR imaging. Eur Spine J.2008; 17(5): 679-85.
11.
WessbergPFrenneredK. Central lumbar spinal stenosis: natural history of non-surgical patients. Eur Spine J.2017; 26(10): 2536-2542.
12.
BekerKGarces-DescovichAMangosingJCabral-GoncalvesIHallettDMorteleKJ. Optimizing MRI logistics: prospective analysis of performance, efficiency, and patient throughput. AJR Am J Roentgenol.2017; 209(4): 836-844.
13.
NelsonEMMonazzamSMKimKDSeibertJAKlinebergEO. Intraoperative fluoroscopy, portable X-ray, and CT: patient and operating room personnel radiation exposure in spinal surgery. Spine J.2014; 14(12): 2985-2991.
14.
AmundsenTWeberHLilleåsFNordalHJAbdelnoorMMagnaesB. Lumbar spinal stenosis. Clinical and radiologic features. Spine (Phila Pa 1976).1995; 20(10): 1178-1186.
15.
SunZQXuYY. Medical Statistics: Graduate Teaching Books. 4rd ed. Beijing: People’s Health Press. 2016 [in chinese].
16.
CohenJ. Statisticalpower analysis for the behavioral sciences. 2rd ed. New York: A cademic. 1988.
17.
PfirrmannCWMetzdorfAZanettiMHodlerJBoosN. Magnetic resonance classification of lumbar intervertebral disc degeneration Spine. (Phila Pa 1976).2001; 26(17): 1873-1878.
18.
ParkJOLeeBHKangYMKimTHYoonJYKimH, et al. Inflammatory cytokines induce fibrosis and ossification of human ligamentum flavum cells. Journal of Spinal Disorders & Techniques.2013; 26(1): E6-E12.
19.
SplendianiAFerrariFBarileAMasciocchiCGallucciM. Occult neural foraminal stenosis caused by association between disc degeneration and facet joint osteoarthritis: demonstration with dedicated upright MRI system. Radiol Med.2014; 119(3): 164-74.
20.
GuermaziARoemerFWBursteinDHayashiD. Why radiography should no longer be considered a surrogate outcome measure for longitudinal assessment of cartilage in knee osteoarthritis. Arthritis Research & Therapy.2011; 13(6): 247.
21.
DuCFYangNGuoJCHuangYPZhangC. Biomechanical response of lumbar facet joints under follower preload: a finite element study. BMC Musculoskelet Disord.2016; 17: 126.
22.
TakashimaHTakebayashiTYoshimotoMTerashimaYIdaKShishidoH, et al. Investigation of intervertebral disc and facet joint in lumbar spondylolisthesis using T2 mapping. Magn Reson Med Sci.2014; 13(4): 261-6.
23.
KäserLMannionAFRhynerAWeberEDvorakJMüntenerM. Active therapy for chronic low back pain: part 2. Effects on paraspinal muscle cross-sectional area, fiber type size, and distribution. Spine.2001; 26(8): 909-919.
24.
FortinMVidemanTGibbonsLEBattiéMC. Paraspinal muscle morphology and composition: a 15-yr longitudinal magnetic resonance imaging study. Med Sci Sports Exerc.2014; 46(5): 893-901.
25.
YarjanianJAFetzerAYamakawaKSTongHCSmuckMHaigA. Correlation of paraspinal atrophy and denervation in back pain and spinal stenosis relative to asymptomatic controls. PM R.2013; 5(1): 39-44.
26.
ScaleDZichnerL. Natural course in lumbar disk prolapse. Orthopade.1994; 23(3): 236-42.
27.
CheungPWHTamVLeungVYLSamartzisDCheungKMLukKD, et al. The paradoxical relationship between ligamentum flavum hypertrophy and developmental lumbar spinal stenosis. Scoliosis Spinal Disord.2016; 11(1): 26.
28.
StokesIAGardner-MorseMChurchillDLaibleJP. Measurement of a spine motion segment stiffness matrix. Biomech.2002; 35(4): 517-521.
