The aim of this study was to evaluate the diagnostic value of the MR Imaging findings with provoked discography used as the standard for painful lumbar disc derangement.

Two hundred patients (412 discs), (age range: 21-77 years), with chronic low back pain underwent MRI and provoked discography. We evaluated the MRI T2-WI findings such as disc degeneration, high-Intensity zones and endplate abnormalities. Subsequently, provocative discography was independently performed with using MR imaging, and a painful disc was defined when moderate to severe and concordant pain was provoked. We calculated the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the MRI findings with using provoked discography as the standard.

400 discs showed abnormal findings such as disc degeneration, HIZ and endplate abnormalities on the T2-WI images. 12 discs showed normal findings. HIZ or endplate abnormalities were always combined with disc degeneration. The prevalence of each findings were disc degeneration (400 discs: 97.1%), HIZ (111 discs: 26.9%), type I endplate abnormalities (34 discs: 8.3%), type II endplate abnormalities (75 discs: 18.2%), the combined findings of HIZ and type I endplate abnormalities (2 discs: 0.5%) and the combined findings of HIZ and type II endplate abnormalities (7 discs: 1.7%). The disc degeneration showed high sensitivity (99.5%) and low specificity (5.0%), so only the NPV (91.7%) was significant, and not the PPV (47.8%). Each findings of HIZ (sensitivity, 36.5%; specificity, 81.4%; PPV, 63.18%; NPV, 59.5%), type I endplate abnormalities (11.0%, 94.1%, 61.8% and 54.8%, respectively), type II endplate abnormalities (19.8%, 83.2%, 50.7% and 54.3%, respectively), the combined findings of HIZ and type I endplate abnormalities (0.5%, 99.6%, 50.0% and 53.4%, respectively) and the combined findings of HIZ and type II endplate abnormalities (26.0%, 99.1%, 71.4% and 53.8%, respectively) show high specificity, but low sensitivity, so the PPV and NPV were also not significant.

For diagnosing painful lumbar disc derangement, the MR imaging findings seem to be inadequate as predictive factors when provoked discography was used as the standard.