Imaging Studies
Routine thoracolumbar radiography in an athlete suspected of having a lumbosacral spondylolysis (lumbar spondylolysis) should include anteroposterior (AP), lateral, and oblique views. Flexion and extension lateral radiographs can be added if spondylolisthesis is present or stability is in question.
AP radiographic findings are usually normal in spondylolysis. [31] The oblique views are particularly screened for the "Scotty dog" lesion in the pars interarticularis (see the image below). [12] The pars defect is represented by the collar on the Scotty dog.
Keep in mind that the oblique view must be of good quality and read by someone experienced in making the diagnosis of lumbosacral spondylolysis (lumbar spondylolysis). Considerable interobserver variation in making this diagnosis has been noted. [30]
The 30º cranially angulated AP view and the coned lateral view of the lumbosacral junction are proposed to display the majority of defects. Because the 30º cranially angulated AP view clearly depicts the pars interarticularis of L5 in the frontal plane, it also allows the distinction between unilateral and bilateral defects. [32] Suffice to say, plain radiography is used more to exclude unusual lesions and spondylolisthesis than to make the diagnosis of lumbosacral spondylolysis (lumbar spondylolysis). [15]
An athlete who is symptomatic with hyperextension and whose plain radiography results are normal requires additional investigations. A 99-m technetium (99m Tc) ethylene diphosphonate bone scintigraphy identifies pars interarticularis stress fractures that were missed on oblique plain radiography. Those patients who have had recent trauma or strenuous activity and who are symptomatic show increased activity on the spondylolytic area. On the other hand, those with chronic low back pain show normal scans. [33]
Technetium bone scanning has its limitations. [34] When compared to single-photon emission computed tomography (SPECT) scanning, the SPECT scanning was found to be more sensitive and better at localization of abnormalities in those athletes with radiographically obvious lumbosacral spondylolysis/spondylolisthesis. [35]
Bone scanning has a false-positive rate of 15% when using computed tomography (CT) scanning as the standard to establish the diagnosis of lumbosacral spondylolysis (lumbar spondylolysis). [15] Therefore, SPECT scanning is needed to identify the more subtle stress reactions. Moreover, these scans can aid in establishing the acuity of the lesion or in identifying the site of the problem in an athlete with negative plain radiography results but whose clinical course is suggestive of a pars interarticularis fracture. [33] The main value of SPECT scanning lies in the identification of an acute stress reaction of the pars before its manifestation radiographically. [32]
For more information about the defect in the bony structure involved, performing CT scanning through the fracture provides excellent detail (see the image below).
Sagittally reconstructed computed tomography scan of the lumbar spine shows a defect of the pars interarticularis on the left at L5.
CT scanning also provides information regarding other features of spondylolysis, such as facet joint changes, spondylolisthesis, and disc herniations, and it also helps prognosticate the potential for a lesion to heal. [36, 37]
The technique for CT scanning is crucial, because a lesion is optimally demonstrated when the plane of scanning is perpendicular to the plane of fracture. This is achieved if the CT scan gantry is reversed so that the scanning plane is perpendicular to the fracture. [32] CT scanning is not useful for impending stress fractures. [6]
MRI is inferior to CT scanning for direct visualization of defects of the pars interarticularis. [38] However, MRI has utility to rule out disc herniations. (See the images below.)
Long TR (T2-weighted) fat suppressed sagittal magnetic resonance image shows increased signal in the pars interarticularis on the left at L5 (same patient in Images 3-4). This is an acute stress reaction.
Sagittal short TR (T1-weighted) magnetic resonance image shows decreased signal in the pars interarticularis on the left at L5 (same patient in Images 3-4).
Sairyo et al evaluated how the stage of a pars defect on CT scan and the presence or absence of high signal change in the adjacent pedicle on T2-weighted MRI were related to bony healing following conservative treatment of lumbar spondylolysis in childhood and adolescence. [39] Twenty-three children (19 boys, 4 girls; mean age 13.5 y) with 41 pars defects were conservatively for at least 3 months, in which they were asked to refrain from sporting activity and to wear a Damen soft thoracolumbosacral type brace.
The pars defects were classified as an early, progressive or terminal stage on CT scans. Early-stage lesions were defined as those that had a hairline crack in the pars interarticularis, which became a gap in the progressive stage. Terminal-stage defects were considered equivalent to a pseudarthrosis. [39] On T2-weighted MRIs, the presence or absence of high signal change in the adjacent pedicle were evaluated; then, the defects were divided into high signal change-positive or -negative. Healing of the defect was assessed by CT scanning. [39]
The investigators found 13 (87%) of the 15 early defects healed, but of 19 progressive defects, only 6 (32%) healed. None of the seven terminal defects healed. In addition, of the 26 high signal change-positive defects, 20 (77%) healed after conservative treatment, but none of the high signal change-negative defects did. [39] Sairyo et al concluded that early-stage defects on CT scans and high signal changes in the adjacent pedicle on T2-weighted MRIs may be useful predictors of bony healing of a pars defect in children following conservative treatment.
In another study, Zehnder et al performed a retrospective radiographic review to assess the relationship between interfacet spacing and pediatric spondylolysis. [40] The investigators compared the anteroposterior lumbar spine radiographs of 41 children with spondylolytic defects with 41 unaffected controls. The vertebral body width and interpedicular distance were recorded, and statistical analysis of the relationships of interfacet distances between the affected and unaffected groups was performed.
Zehnder et al noted that the absolute increase in interfacet distance between adjacent levels was significantly smaller at the L4/L5 level in spondylolytic individuals; however, when interpedicular distance was used to standardize for vertebral body size, a significantly smaller increase in the interpedicular distance was noted at the L4/5 level in spondylolytic individuals. The investigators demonstrated similar results when body width was used to standardize for vertebral body size, with a similar, but not significant, trend at the L3/4 level when standardizing with interpedicular distance. Zehnder et al indicated that a likely explanation for the etiology of lumbar pars defects is insufficient caudal increase in lumbar interfacet spacing but noted that further prospective studies are needed to determine if unaffected individuals with a narrowed interfacet spacing are at increased risk of developing spondylolytic defects later in life. [40]
-
Radiograph of L4 defect in the pars interarticularis.
-
Computed tomography scan demonstrating defects in the left and right pars interarticularis.
-
Long TR (T2-weighted) fat suppressed sagittal magnetic resonance image shows increased signal in the pars interarticularis on the left at L5 (same patient in Images 3-4). This is an acute stress reaction.
-
Sagittal short TR (T1-weighted) magnetic resonance image shows decreased signal in the pars interarticularis on the left at L5 (same patient in Images 3-4).
-
Lateral radiograph of the lumbar spine shows spondylolysis at L5 with spondylolisthesis at L5 through S1. On this single view, it is not possible to determine if these pars defects are unilateral or bilateral. Oblique views may help resolve this issue.
-
Sagittally reconstructed computed tomography scan of the lumbar spine shows a defect of the pars interarticularis on the left at L5.
-
Lumbar oblique radiograph showing the "Scotty dog." A pars defect is seen at L5.
