Congenital lumbar spinal stenosis: a prospective, control-matched, cohort radiographic analysis

Abstract

Background context: Degenerative lumbar spinal stenosis manifests primarily after the sixth decade of life as a result of facet hypertrophy and degenerative disc disease. Congenital stenosis, on the other hand, presents earlier in age with similar clinical findings but with multilevel involvement and fewer degenerative changes. These patients may have subtle anatomic variations of the lumbar spine that may increase the likelihood of thecal sac compression. However, to the authors' knowledge, no quantitative studies have addressed various radiographic parameters of symptomatic, congenitally stenotic individuals to normal subjects.

Purpose: To radiographically quantify and compare the anatomy of the lumbar spine in symptomatic, congenitally stenotic individuals to age- and sex-matched, asymptomatic, nonstenotic controlled individuals.

Study design/setting: A prospective, control-matched, cohort radiographic analysis.

Patient sample: Axial and sagittal magnetic resonance imaging (MRI) and lateral, lumbar, plain radiographs of 20 surgically treated patients who were given a clinical diagnosis of congenital lumbar stenosis by the senior author were randomized with images of 20, asymptomatic age- and sex-matched subjects.

Outcome measures: MRIs and lateral, lumbar, plain radiographs were independently quantitatively assessed by two individuals. Measurements obtained from the axial MRIs included: midline anterior-posterior (AP) vertebral body diameter, vertebral body width, midline AP canal diameter, canal width, spinal canal cross-sectional area, pedicle length, and pedicle width. From the sagittal MRIs, the following measurements were calculated: AP vertebral body diameter, vertebral body height, and AP canal diameter at the mid-vertebral level. On the lateral, lumbar, plain radiograph (L3 level), the AP diameters of the vertebral body spinal canal were measured.

Methods: The images of these 40 individuals were then randomized and distributed in a blinded fashion to five separate spine surgeons who graded the presence and severity of congenital stenosis utilizing a five-tier scale. Images consisting of 15 symptomatic individuals, graded definitely congenitally stenotic (mean age, 51.7 years; range, 43-65 years), and 15 asymptomatic individuals, graded definitely not stenotic (mean age, 50.7 years; range, 41-55 years), were age- and sex-matched and included for further review. From these 30 patients, a lateral, lumbar, plain radiograph and axial and sagittal MRIs (T1/T2 weighted) from L2-L5 were quantitatively analyzed. Rater reliability was assessed by Kappa coefficient testing.

Results: The cross-sectional area of the canal was significantly smaller in the congenitally stenotic patients at all lumbar levels measured (L2: 176 mm(2) vs. 259 mm(2), L3: 177 mm(2) vs. 275 mm(2), L4: 183 mm(2) vs. 283 mm(2), L5: 213 mm(2) vs. 323 mm(2), p<.05). Pedicle length was markedly shorter in the stenosis group at each lumbar level (L2: 5.9 mm vs. 8.9 mm, L3: 6.0 mm vs. 8.8 mm, L4: 6.5 mm vs. 9.2 mm, L5: 5.8 mm vs. 9.1 mm, p<.05). Furthermore, midline, axial AP canal diameter, vertebral body width, and sagittal AP canal diameter were all significantly smaller than the control patients (p<.05). A ratio of the AP diameter of the pedicle length to the vertebral body was also noted to be statistically significant on both the lateral plain radiographs (L3: 0.426 vs. 0.704) and sagittal MRI (L2: 0.343 vs. 0.461, L3: 0.361 vs. 0.461, L4: 0.362 vs. 0.481, L5: 0.354 vs. 0.452, p<.05). No difference was noted comparing the AP diameter of the vertebral body (axial and sagittal images), vertebral body height, canal width, and pedicle width. Kappa testing coefficient indicated a strong rater reliability (k=0.81, 95% confidence interval: 0.62-0.94).

Conclusions: Congenital lumbar stenosis has not been clearly defined radiographically. Clinically, congenitally stenotic patients present at a younger age with fewer degenerative changes and multiple levels of involvement. Radiographically, these patients have a shorter pedicular length and as a result a smaller cross-sectional spinal canal area (mean critical values of 6.5 mm and 213 mm(2) were observed, respectively). The mean critical ratios were 0.43 (2:1 AP vertebral body: pedicle length) on the lateral lumbar radiograph and 0.36 on the sagittal MRI. The altered canal anatomy resulting from a decreased pedicle length may anatomically predispose these patients to earlier complaints of symptomatic neurogenic claudication. Identification of the presence of congenital stenosis should increase the treating surgeon's awareness of the potential need for multilevel treatment.