The Adams classification for cadaveric discograms: inter- and intra-observer error in the clinical setting.

The Adams classification for discogram morphology is based on a cadaveric study. It provides the basis for several subsequent classifications proposed in the literature. However, little or no attention has been paid to its reproducibility in the clinical setting. The authors assessed the reliability of this classification using three independent observers of differing experience. One hundred and thirty-three discograms belonging to 71 patients with chronic low back pain were reviewed in a randomised and blinded manner. The morphological appearance at each discogram level was assessed and assigned a type according to the Adams classification. The exercise was repeated 3 weeks later. Respective inter- and intra-observer agreements were calculated in the standard fashion using the kappa statistic. Both inter- and intra-observer agreements were excellent (kappa= 0.77-0.85). The Adams grading system for discogram morphology is consistently reproducible amongst observers with differing levels of experience. It can be safely recommended in the clinical setting as a reliable classification.