Discography was a commonly used diagnostic test until a 2009 study by Carragee et al. suggested that that procedure resulted in accelerated degeneration of discs subjected to it. Enthusiasm for discography had been waning due to its relatively poor ability to predict the results of fusion, and the Carragee study was essentially the final nail in the discography coffin. In order to further evaluate the question of the effect of discography on disc degeneration, Dr. McCormick and colleagues designed a retrospective cohort using a consecutive series of low back pain patients who underwent low-pressure discography with a 22 gauge needle between 1998 and 2007. They identified 585 patients who underwent the procedure, 49 of whom had an MRI 7 to 10 years after their discogram. Of these, 23 were excluded as they underwent fusion, one was excluded due to a significant scoliosis, and one was unable to be matched to control patients. They identified 82 patients who had a baseline lumbar MRI during the same time frame and also had a follow-up MRI 7 to 10 years later to serve as controls (matched for age, BMI, Pfirrmann score, and number of disc herniations present). This resulted in cohorts of 77 discs that had been subjected to discography and 260 control discs. The groups were well-matched on age and BMI, but the discography group was 100% male and the control group was 55% male. At baseline, the discography group had a greater proportion of Pfirrmann grade III or IV discs (82% vs. 70%) and a significantly lower disc to CSF signal intensity ratio (indicating more severe disc degeneration). There were no significant differences in proportion of disks that progressed in Pfirrmann grade (17% discography vs. 21% controls), though the mean loss in disc height was greater for the discography group (1.1 mm vs. 0.5 mm). An analysis comparing the non-punctured discs in the discography group to the corresponding discs in the control group showed no differences. The authors concluded that discography did not increase the risk of disc degeneration.
The authors have done a nice job assembling a well-controlled retrospective cohort study in an effort to determine if discography increases the rate of disk degeneration. Prior to drawing any strong conclusions, the limitations of the study need to be considered. While the authors were effective at matching on age and BMI, the fact that 100% of the discography group were males compared to 55% of the control group calls into question how similar the groups actually were. The authors attempted to control for gender in the statistical analysis, though this was probably not completely effective given the lack of females in the discography group. Additionally, the discography group had more advanced disc degeneration at baseline, with 82% of disks graded as Pfirrmann III or IV. Pfirrmann V disks are relatively uncommon in this age group (average age 44 to start the study), so this likely created a ceiling effect in that most of the Grade III and IV patients were unlikely to progress to Grade V, discogram or not. In the Carragee paper, nearly 60% of the discography and control disks were Pfirrmann Grade I or II at baseline, so this group had a much greater risk of progression. Creating separate Grade I, II, III, and IV groups rather than combining them may have allowed for greater precision in evaluating progression of disc degeneration, however, this would have likely resulted in an underpowered analysis. Additionally, the authors did not evaluate or control for smoking status, a known risk factor for disc degeneration. While this study revisits the question of whether or not discography is harmful, the bigger question is whether discography is ever indicated. Given its poor prognostic ability, discography should probably not have a major role in the work-up of degenerative disc disease. Knowing this, the authors suggest that disc puncture for procedures developed in the future (i.e. stem cell injection, etc.) are likely not harmful. Given the limitations of the current study and the contradictory findings of the Carragee study, that conclusion seems to be a stretch.
Please read Dr. McCormick's article on this topic in the October 1 issue. Does this article change your viewpoint on the safety of discography? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor