| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
r2ldere2
1 Department of Radiology, Baskent University, Alanya Research Center, Antalya,
Turkey.
2 Radiology Department, Baskent University School of Medicine, Fevzi Cakmak cad.
10. Sok, No 45 Bahcelievler 06490, Ankara, Turkey.
Received July 6, 2007;
accepted after revision April 23, 2008.
Address correspondence to N. C. Tarhan
(caglat{at}baskent-ank.edu.tr).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. The lumbar MRI findings of 372 patients (141 men, 231 women; mean age, 51.2 years) with nonradicular lower back pain and of 249 healthy persons acting as controls (126 men, 123 women; mean age, 49.3 years) were analyzed. The sagittal STIR sequence was used for all MRI examinations. Presence of interspinous ligament edema, facet joint effusion, neocysts, paraspinal muscle edema, subcutaneous edema, disk herniation, and disk degeneration was evaluated, and the incidence of each finding was determined. All findings were grouped according to age and sex. Chi-square, Fisher's exact, and independent-samples Student's t tests and Spearman's rank correlation analysis were used for statistical analysis.
RESULTS. The incidences of facet joint effusion, interspinous ligament edema, neocyst formation, and paraspinal muscle edema were found to be statistically significantly higher in patients with lower back pain than in controls. The incidences of intervertebral disk degeneration, disk herniation, and subcutaneous edema in persons with and those without lower back pain were similar. Intervertebral disk degeneration, disk herniation, subcutaneous edema, and muscle edema were found to increase with age in both persons with and those without symptoms.
CONCLUSION. Degenerative changes in the posterior paraspinal structures were found in a higher percentage of subjects with lower back pain than in controls. Use of a STIR sequence with homogeneous fat suppression facilitates visualization of these changes.
Keywords: lower back pain • MRI • posterior paraspinal soft tissue • STIR sequence
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Conventional MRI is diagnostically insufficient in 85% of cases of LBP [5]. An explanation for this diagnostic difficulty is that pain can originate from any lumbar structure. In most cases, LBP is assumed to be due to muscular strain, injury to the ligaments, and degenerative changes in the spine [6]. This idea is debatable [1], however, because all of these abnormalities are encountered in persons without symptoms [2]. A small number of reports in the medical literature describe evaluation of the posterior elements on fat-suppressed images [7, 8]. In those studies, however, the investigators did not visualize all of the posterior elements and included only small numbers of patients.
The aim of this study was to investigate the significance of the presence of degenerative changes in posterior spinal elements in patients with LBP compared with the findings in persons without such changes and to see whether the changes are a cause of LBP. Another aim was to determine among persons with and those without LBP the age- and sex-related distribution of changes in the posterior elements on MR images obtained with the STIR sequence.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The lumbar MR images of all subjects were evaluated for pathologic changes in the posterior elements of the spine. The MR images were obtained with a 1.5-T unit (Symphony, Siemens Medical Solutions) and a body coil. At conventional MRI, the following sequences were used: sagittal T1-weighted images (TR/TE, 600/10; number of signals acquired, 2; matrix size, 230 x 384; slice thickness, 4 mm; acquisition time, 2 minutes 32 seconds), sagittal T2-weighted images (3,250/92; number of signals acquired, 1; matrix size, 230 x 384; slice thickness, 4 mm; acquisition time, 1 minute 33 seconds), axial T2-weighted images (3,400/90; number of signals acquired, 2; matrix size, 230 x 384; slice thickness, 4 mm; acquisition time, 2 minutes 0 seconds). In addition to the conventional MRI sequences, a sagittal STIR sequence (6,120–7,040/60–70; inversion time, 150 milliseconds; number of signals acquired, 2; matrix size, 132 x 256; slice thickness, 4 mm; acquisition time, 2 minutes 26 seconds) was used for all MRI examinations.
|
Intervertebral disk degeneration was assumed to be a loss of signal intensity of the disk or a reduction in the height of the disk. Disk herniation was defined as focal or diffuse extension of the disk beyond its end plate. Degeneration or rupture of the interspinous ligaments was seen as high signal intensity between the spinous processes of the lumbar vertebrae. Facet joint effusion was visualized as high-signal-intensity fluid between the superior and inferior facets at lumbar levels. Cyst formation next to the facet joints and emanating from the interspinous neoarthrosis was seen as a round cystic lesion with high signal intensity on STIR images. Intrinsic spinal muscle degeneration manifested as edema of the posterior paraspinal muscles with high signal intensity on STIR images at the lumbar level. Subcutaneous edema was visualized as a triangular area of increased signal intensity of subcutaneous fat behind the lumbar vertebrae and paraspinal muscles, especially on midline images obtained with the STIR sequence.
|
The study data were entered into a statistical program (SPSS 11.0, SPSS) for data control. Chi-square and Fisher's exact tests were used to compare nominal data variables, such as sex and pathologic findings, in the two subject groups, the sex groups, and the age groups. The independent-samples Student's t test was used for finding the concordance of ages between the case and control groups. Correlation between grouped age-related pathologic findings was tested with Spearman's rank correlation analysis. A value of r < 0.4 was accepted as weak correlation, 0.4 < r < 0.7 as moderate correlation, and r > 0.7 as strong correlation. Values of p < 0.05 were accepted as statistically significant.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients with LBP without radiculopathy were given the diagnosis of pathologic change in at least one of the posterior elements stabilizing the vertebral column. The largest percentage of these patients (25%) were 41–50 years old. The distribution of the pathologic findings according to sex and age as percentages in the group with symptoms is shown in Figures 1 and 2. Posterior element abnormalities in order of frequency were facet joint effusion (85.5%) (Figs. 3, 4, 5), interspinous ligament degen eration or rupture with edema at L4–L5 (80.6%) and L5–S1 (79.8%) (Figs. 6 and 7), neocyst formation and synovial cysts (62.4%) (Figs. 3 and 4), subcutaneous edema (27.2%) (Fig. 7), and intrinsic spinal muscle degen eration (24.7%) (Fig. 5). Intervertebral disk herniation was encountered in 41.7% of all cases and degeneration in 65.1%.
|
|
|
|
Degenerative changes in at least one posterior element were found in only 76.3% of the control group (190 subjects); 23.7% (59 subjects) had no changes. The frequency of pathologic changes in the posterior elements of the control group was as follows: facet joint effusion, 45.8%; interspinous ligament degeneration with edema at L4–L5, 32.5%; subcutaneous edema, 19.7%; neocyst formation and synovial cysts, 15.3%; and intrinsic spinal muscle degeneration, 4.8%. Intervertebral disk herniation and disk degeneration were encountered in 39.8% and 60.2% of this group, rates very close to the frequencies in the group with symptoms. The distribution of frequencies in the control group according to sex and age is summarized in Figures 8 and 9.
|
|
Most patients had interspinous ligament degeneration or rupture at more than one level, mostly at the L4–L5 and L5–S1 levels and in the 15- to 30-year and the 31- to 40-year age groups. In the 15- to 30-year group with symptoms, the incidence was 88.4% at the L4–L5 and L5–S1 levels; in the 31- to 40-year group, the incidence was 89.8% at the L4–L5 level and 73.5% at the L5–S1 level. Interspinous ligament edema was encoun tered mainly in the 41- to 50-year (41.8% at L4–L5) and 61- to 70-year (41.9% at L5–S1) age groups without symptoms. Seventy-eight percent of the patients with interspinous ligament edema had concomitant interver tebral disk degenera tive changes at the same level. The case–control group comparison of interspinous ligament edema based on age group and spinal level showed this finding to be more frequent in the case than in the control group.
Neocyst formation and synovial cysts were encountered mainly in the 15- to 30-year (69.8%) and 31- to 40-year (69.4%) age groups with symptoms. In the subjects without symptoms, these findings were made mostly in the 71-year and older group (22.2%). In the age-group comparison, these findings were more frequent in the case than in the control group. In most of the cases of symptomatic facet joint cysts (95%), there was concomitant effusion in the neighboring facet joint; effusion was present in only 69% of the subjects without symptoms.
Intrinsic spinal muscle injury was encountered most frequently in the 51- to 60-year-old (32.2%) and 61- to 70-year-old (32%) subjects in the case group. In the control group, this finding was most frequent in the 71-year and older age group (14.8%). The comparison of signal intensity increase in the intrinsic spinal muscles according to age group showed a higher frequency at all ages in the case group than in the control group. All intrinsic muscle injuries were accompanied by interspinous ligament edema in subjects with and those without symptoms.
The comparison of pathologic changes in intervertebral disks (degeneration, herniation) and structures other than disks (signal intensity increase in the intrinsic spinal muscles and subcutaneous edema) according to age within the case and control groups revealed a statistically significant relation. A weak positive correlation between age and pathologic condition was found for disk herniation (control group, r = 0.348, p = 0.0001; case group, r = 0.299, p = 0.0001), intrinsic spinal muscle injury (control group, r = 0.208, p = 0.001; case group: r = 0.159, p = 0.002), and subcutaneous edema (control group, r = 0.325, p = 0.0001; case group, r = 0.327, p = 0.0001). A statistically moderate positive correlation between age and intervertebral disk degeneration was found in both the case and control groups (case group, r = 0.462, p = 0.0001; control group, r = 0.508, p = 0.0001).
There was a statistically significant difference in interspinous ligament degeneration or rupture at L3–L4 and L4–L5 only between age groups of patients with symptoms (p < 0.05). No statistically sig nificant relation was found between age and interspinous ligament edema at T12–L1, L1–L2, L2–L3, or L5–S1, neocyst formation, or facet joint effusion.
Case–control comparison showed a statistically significant difference with regard to the following pathologic conditions: facet joint effusion (p = 0.0001), interspinous ligament edema at all lumbar levels (p = 0.001 for T12–L1, p = 0.0001 for all other levels), neocyst formation and synovial cysts (p = 0.0001), and intrinsic muscle degeneration (p = 0.0001). A similar relation was found with regard to subcutaneous edema (p = 0.03). Because, however, there was a significant sex difference between the case and control groups with regard to subcutaneous edema, which was statistically more frequent among women, a separate evaluation was performed for men and women. Thus within-sex comparison showed subcutaneous edema more frequent in patients with LBP than in those without LBP, but the difference was not statistically significant (men, p = 0.289; women, p = 0.180). Similarly, no statistically significant difference was found in a comparison of case and control groups with regard to both intervertebral disk degeneration and disk herniation (p > 0.05).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The higher rates of facet joint edema, interspinous ligament injury, neocyst formation and synovial cysts, and intrinsic spinal muscle edema among persons with symptoms than among those without symptoms show that pathologic changes in the posterior osteoarticular elements and soft tissues of the spinal column may be the cause of or a manifestation of the cause of LBP. On the other hand, the similar rates of intervertebral disk degeneration, disk herniation, and subcutaneous edema in the case and control groups indicate that these findings may be found even if LBP is not present. The age-related increase in intervertebral disk degeneration, disk herniation, subcutaneous edema, and intrinsic spinal muscle edema observed in both groups indicates that those degenerative changes may develop with age. Conventional MRI is insufficient for visualizing these paraspinal changes [5]. On STIR images, however, these degenerative changes can be clearly visualized owing to homogeneous fat suppression and better depiction of edema than on conventional MR images.
In a study by Jensen et al. [3], disk herniation (diffuse or focal protrusion), being a frequently encountered asymptomatic finding, was assumed to be an incidental finding on MR images of patients with LBP. Savage et al. [9] conducted a study with 149 patients with and without LBP and found similar ratios and no statistically significant difference between the case and control groups regarding disk herniation (excluding extruded or sequestered disks) and disk degeneration. Similar findings were made in other studies [10, 11] and in ours. As a result, we conclude there is a poor correlation between LBP and disk degeneration or herniation.
The posterior paravertebral elements (facet joints, ligamentum flavum, interspinous ligaments, supraspinous ligaments, and para spinous muscle structures) have a large number of innervations. The interspinous ligament, which is one of the most important structures rendering stability to the spinal column, is innervated by the medial branch of the dorsal ramus [12, 13]. Acute trauma sprains the interspinous ligament, and chronic repetitive microtrauma usually results in subtotal degenerative rupture of the ligament. Degenerative changes in interspinous ligaments begin early in the second decade of life in healthy persons. Rupture usually is present in 20% of persons older than 20 years, especially at L4–L5 and L5–S1 [14]. In our study, interspinous ligament rupture at L4–L5 was found in 32% and at L5–S1 in 31% of healthy subjects older than 20 years. Before this study, only a limited number of studies reported in the literature evaluated whether an observed increase in interspinous ligament signal intensity indicates the presence of changes causing LBP. In a study conducted with a small number of subjects with LBP who had a mean age of 56 years, Jinkins [7] found an increase in interspinous ligament signal intensity on the fat-saturated T2-weighted images of 71% of the subjects. The control group was quite small, however, and age distribution was not reported. Our study revealed similar results with regard to presence of interspinous ligament edema in the 51- to 60-year age group with symptoms; this finding was most frequent at L4–L5 (75.9%) and L5–S1 (82.8%). Our study was conducted with a larger control group than Jinkins used, and the results revealed a significantly higher ratio of interspinous ligament edema in the case group than in the control group, indicating that the edema may be a cause of or a manifestation of the cause of LBP.
In patients with interspinous ligament injuries, the ligament becomes visible as an area of high signal intensity on T2-weighted images owing to inflammatory exudation and inflammation of the fibrous bands. This high signal intensity, however, resembles the fat tissue surrounding the ligament on conventional sequences, making it difficult to differentiate. This problem necessitates use of STIR [4] or fat-suppressed T2-weighted sequences, both of which are effective at showing inflammatory soft-tissue lesions, for correct diagnosis [15, 16]. In their study correlating imaging and surgical findings in the cases of thoracolumbar trauma patients, Lee et al. [17] found that fat-suppressed T2-weighted imaging was highly specific and sensitive for visualization of interspinous ligament injury and correctly showed the presence of the lesions. In our study, the edematous and degenerative changes in the posterior spinal elements not seen on conventional T1- and T2-weighted images—especially when the interspinous ligaments, synovia, neocysts, and muscles were surrounded by fat tissue—were easily visualized on STIR images. Although it took approximately 20 seconds longer than the fat-suppressed T2-weighted sequence, the STIR sequence was especially preferred in our study because it provided homogeneous fat suppression with out magnetic field inhomogeneity and with better depiction of edema.
It is thought [1] that degenerative changes in the facet joint and an increased amount of intraarticular fluid are among the important overlooked causes of LBP. Study results [18, 19] support this idea in that LBP in some patients was relieved after selective facet joint block or facet denervation. Our study revealed a higher frequency of facet joint effusion among subjects with symptoms (85.5%) than among subjects without symptoms (45.8%), supporting the hypothesis that effusion may be a possible cause or a manifestation of the cause of LBP. Most degenerative changes in the facet joint can be diagnosed with conventional MRI [1]. However, joint effusion can be clearly visualized with the STIR sequence, as found in our study. Our study showed that 78% of cases of facet joint effusion were accompanied by intervertebral disk degeneration at the same level. The explanation may be degenerative height reduction of the intervertebral disk leading to partial subluxation of the posterior facet joint, which is usually accompanied by intraarticular effusion [12].
Synovial cysts and neocysts of the posterior spinal facets are rare lesions related to degenerative lumbosacral changes [20]. These lesions may extend into the spinal canal or be located in the posterior paraspinal soft tissues. The number of studies in the literature in which the frequency of synovial cysts and neocysts was evaluated is small. In a study of conventional sequences for MRI of patients reporting back pain or sciatica, Doyle and Merrilees [21] found a 7.3% frequency of posterior synovial cysts. In our study group, the rate of these cysts was 62.4%. We consider the difference due to our use of the STIR technique, which can clearly depict even small cysts that are barely discernible from the fat tissue in which they are embedded.
It is well known that synovial cysts extending into the spinal canal can cause neurologic symptoms. The clinical findings of extraspinal cystic lesions, however, are debated unless they cause a pressure effect on the basic neurologic structures [1]. The contribution to LBP of synovial cysts and neocyst formation observed more frequently in persons with symptoms (62.4%) than in those without symptoms (15.3%) was probably due to degenerative and inflammatory changes in the corresponding facet joints. This supposition is supported by the results of our study revealing that most of the synovial cysts and neocyst formations in the group with symptoms, unlike the subjects without symptoms, were accompanied by effusion in the neighboring facet joint.
Intrinsic spinal muscle changes can have a direct (intrinsic spinal muscle strain or rupture) or indirect (intrinsic spinal muscle spasm) role in the pathogenesis of LBP [7]. Bennett and associates [8] used fat-suppressed T2-weighted imaging to examine 20 gymnasts with a mean age of 16 years with and without LBP and visualized muscle strain only in those with LBP (19%). In our study, increased intrinsic spinal muscle signal intensity was observed in 12% of the subjects in the 15–30 year age group with LBP. In the other age groups, the rate of intrinsic spinal muscle edema was higher in subjects with symptoms (24.7%) than in those without symptoms (4.8%). Thus we believe is it highly probable that muscle strain is a cause of LBP. Our study also revealed a statistically significant age-related increase in intrinsic spinal muscle edema. The age-related statistically significant in crease in inter vertebral disk degeneration and disk herniation and of intrinsic spinal muscle edema shows that degenerative changes in these structures may appear with aging.
Acute or subacute degeneration of the intrinsic spinal muscles, which originate from and insert into the vertebrae (multifidus and interspinales), usually develops owing to intersegmental hypermobility, which leads to degenerative rupture of the interspinous ligament [22]. Jinkins [7] observed intrinsic spinal muscle (interspinales and multifidus) degeneration (abnormal signal intensity) in only 7% of patients, but all patients with increased muscle signal intensity had an increase in interspinous ligament signal intensity. In our study, the explanation for the lesser degree of degenerative changes observed in the intrinsic muscles than in the interspinous ligaments is probably that muscular tissue is stronger than ligamentous tissue. Both muscular strain (which has high signal intensity owing to muscle edema or hemorrhage) and interspinous ligament injury, can be overlooked, especially when STIR sequences are not used.
The number of reports of evaluation of edema of the subcutaneous soft tissues in the posterior lumbar vertebral column is limited [23, 24]. Furthermore, to our knowledge, there are no reports of evaluation of the presence of such edema in patients with LBP. Increased signal intensity, which is assumed to be subcutaneous tissue edema or fluid collection, is believed to be an incidental finding. The causes can be infectious, inflammatory, traumatic, hydrostatic, and even neoplastic. In their study involving obese and nonobese persons without LBP, Shi and associates [23] found that the rate of the high signal intensity on fat-suppressed T2-weighted images increased with age and weight and among women. Similarly, in our study, the incidence of subcutaneous edema significantly increased with age and among women. No statistically significant difference, however, was encountered between persons with and those without LBP with regard to sex distribution of subcutaneous edema. The higher ratio of subcutaneous edema in persons with than those without LBP might have been be due to poorer lymphatic drainage in patients with LBP as the result of pain-related restriction of physical activity.
Findings on MR images obtained with conventional sequences in the evaluation of patients with LBP who do not have evidence of disk herniation usually do not contribute to a definitive diagnosis, and patients receive symptomatic treatment with no pathoanatomic diagnosis [25]. An accurate diagnosis is important because it contributes to the patient's reassurance and trust in the clinician [26]. If these changes are found, patients can be treated and followed up accordingly. Thus STIR sequences are impor tant for clear evaluation of the posterior spinal structures as the possible cause of LBP and for determination of the therapeutic approach.
Despite the large population, our study had limitations. First, there was no correlation between the lesions and the intensity of symptoms (degree of pain). The lack of control MRI examinations for evaluation of the clinical and radiologic response to therapy was another limitation.
The results of our study suggest that, with high probability, degenerative changes in the posterior paraspinal soft-tissue structures, especially interspinous ligament edema, facet joint effusion, neocyst formation, and intrinsic spinal muscle edema, cause LBP in some patients. Because of homogeneous fat suppression and better depiction of soft-tissue edema, the STIR sequence is the best imaging technique for visualizing the aforementioned changes, and it adds only 2 minutes to the imaging examination. Therefore, we suggest that for patients with LBP without other obvious pathologic findings, the STIR sequence be added to the MRI evaluation to visualize degenerative changes in posterior spinal structures as a possible cause of pain.
Acknowledgments
We thank Metin Karatas and Co
kun Bakar for their help in performing
the statistical analysis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. N. Kraft, P. H. Pennekamp, U. Becker, M. Young, O. Diedrich, C. Luring, and M. von Falkenhausen Magnetic Resonance Imaging Findings of the Lumbar Spine in Elite Horseback Riders: Correlations With Back Pain, Body Mass Index, Trunk/Leg-Length Coefficient, and Riding Discipline Am. J. Sports Med., November 1, 2009; 37(11): 2205 - 2213. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
