The aim of this work was to study the influence of the slice orientation of T2 axial images in numerical measurements of DSCA and study the effect that this change of slice angle would have on the morphological grading assessment. TSE T2 three dimensional aquisition MRI studies reconstructed with OsiriX DICOM viewer from 32 patients were used. Patients included were a series of consecutive cases with either suspected spinal stenosis or low back pain. A total of 97 disc levels were studied and axial reconstructions were made at 0°, +10°, +20°, +30° relative to the disc space orientation. For each image, DSCA was digitally measured and a severity grade was assigned by two observers according to the recently-published 4-point (A-D) morphological grading system. Interobserver kappa score was 0.71. Statistical analysis of DSCA measurements was performed using kappa and t-tests. Comparing DCSA between 0° at each level and +10°, +20° and +30° slice orientation, a significant increase in surface area was found in each case (P<0.0001). % change in DSCA combining all disc levels comparing 0° and +10°: range -15.48% to +31.89% (SD 18.40%); 0° and +20°: range -24.00% to +143.82% (SD 20.45%); 0° and +30°: range -29.35% to +231.13% (SD 26.52%). At 13 disc levels, DSCA was <100mm2 at 0°, but changed to >100mm2 in three cases by a +10° increase, in five cases by a +20° increase and in 10 cases by a +30° increase. In only two out of 97 levels studied did the morphological grading change as the angle increased, one of which was not amongst those above (change in DSCA from <100mm2 to >100mm2).Purpose of the study
Methods and Results
Surgical decision-making in lumbar spinal stenosis
involves assessment of clinical parameters and the severity of the
radiological stenosis. We suspected that surgeons based surgical
decisions more on dural sac cross-sectional area (DSCA) than on
the morphology of the dural sac. We carried out a survey among members
of three European spine societies. The axial T2-weighted MR images
from ten patients with varying degrees of DSCA and morphological
grades according to the recently described morphological classification
of lumbar spinal stenosis, with DSCA values disclosed in half the
assessed images, were used for evaluation. We provided a clinical
scenario to accompany the images, which were shown to 142 responding
physicians, mainly orthopaedic surgeons but also some neurosurgeons
and others directly involved in treating patients with spinal disorders.
As the primary outcome we used the number of respondents who would
proceed to surgery for a given DSCA or morphological grade. Substantial
agreement among the respondents was observed, with severe or extreme
stenosis as defined by the morphological grade leading to surgery.
This decision was not dependent on the number of years in practice, medical
density or specialty. Disclosing the DSCA did not alter operative
decision-making. In all, 40 respondents (29%) had prior knowledge
of the morphological grading system, but their responses showed
no difference from those who had not. This study suggests that the
participants were less influenced by DSCA than by the morphological
appearance of the dural sac. Classifying lumbar spinal stenosis according to morphology rather
than surface measurements appears to be consistent with current
clinical practice.
