Table 5. Interobserver agreement regarding characteristics of the impaired disc level.
| A vs B (n = 343) | A vs C (n = 329) | B vs C (n = 327) | All observers (n = 321) | |||||
| %agreement | kappa | %agreement | kappa | %Agreement | kappa | %agreement | multiraterkappa | |
| Disc contour ‡ | 95.9 | * | 98.2 | * | 95.1 | * | 95.0 | * |
| Loss of disc height ∫ | 97.9 | 0.86 | 72.2 | 0.26 | 72.4 | 0.26 | 71.5 | 0.31 |
| Signal intensity of nucleus pulposus on T2 images ¶ | 95.3 | 0.75 | 90.4 | 0.64 | 90.7 | 0.57 | 88.6 | 0.61 |
| Type of vertebral endplate signal changes upper endplate∥ | 75.8 | * | 83.4 | * | 84.5 | * | 72.6 | * |
| Type of vertebral endplate signal changes lower endplate∥ | 81.1 | * | 83.7 | * | 84.8 | * | 75.4 | * |
| Spinal canal stenosis ∫ | 63.3 | 0.21 | 57.4 | 0.10 | 91.3 | ** | 55.1 | 0.08 |
| Absence of epidural fat adjacent to the dural sac or surrounding the nerve root sheath Ψ | 74.0 | 0.52 | 74.1 | 0.54 | 73.6 | 0.54 | 61.7 | 0.50 |
| Place of absence of epidural fat § | 94.4 | 0.70 | 96.5 | 0.72 | 96.7 | 0.75 | 95.3 | 0.75 |
| Impaired discs on other disc levels ∫ | 93.2 | 0.79 | 85.5 | 0.62 | 85.4 | 0.62 | 82.3 | 0.68 |
| Nerve root thickness distal to the site of compression|-- | 93.5 | *** | 93.5 | *** | 97.5 | *** | 92.1 | 0.40 |
| Flattening of the ventrolateral angle of the dural sac or the emerging root sheath ∫ | 84.3 | 0.60 | 78.7 | 0.51 | 78.3 | 0.46 | 70.9 | 0.50 |
The number between brackets on the first row is the number of patients of which the observers suggested the same disc level as the symptomatic disc level. A en B represent the two neuroradiologists, while C represents the neurosurgeon.
Categories were: bulging disc versus disc herniation.
Categories were: yes versus no.
Categories were: 1) Hypointensity 2) Normal 3) Hyperintensity.
Categories were: 1) No vertebral endplate signal changes (VESC) 2) VESC type I 3) VESC type II.
4) VESC type III 5) Mixed VESC type I/II 6) Mixed VESC type II/III.
Categories were: 1) Yes, completely disappeared 2) Yes, partly disappeared 3) No disappearance.
Categories were: 1) Sub-articular zone 2) Foraminal zone 3) Extra-foraminal zone.
|--Categories were: 1) Normal 2) Thickened 3) Narrowed.
Prevalence of findings too low (<10% of the reports) to calculate kappa values.
Prevalence of spinal canal stenosis too low (<10% of the reports) to calculate kappa values.
Prevalence of thickened nerve roots too low (<10% of the reports) to calculate kappa values.