2009 ISSLS Prize Winner: Does Discography Cause Accelerated Progression of Degeneration Changes in the Lumbar Disc: A Ten-year Matched Cohort Study

Eugene J. Carragee, MD; Angus S. Don, FRACS; Eric L. Hurwitz, DC, PhD; Jason M. Cuellar, MD, PhD; John Carrino, MD; Richard Herzog, MD


Spine. 2009;34(21):2338-2345. 

In This Article


Of 75 discography subjects, 71 completed the baseline evaluation: 4 subjects in the somatization group refused the disc injection and 1 also could not complete the MRI. Of the 75 control subjects, 73 completed the entire baseline evaluation; 2 subjects in the somatization group could not tolerate the MRI scanner. At 10 years after the baseline assessment, 57 of 71 discography subjects (80.3%) and 54 of 73 control subjects (74.0%) were successfully contacted. After the scripted medical interview, 5 discography subjects were excluded (all had undergone new lumbar surgery before 7 years follow-up) and 4 control subjects were excluded (1 had died, 1 had new lumbar surgery before 7 years follow-up, 1 had sustained a high-energy L2 burst fracture, and 1 had new onset psoriatic spondyloarthropathy), leaving 52 discography subjects and 50 control subjects eligible for the follow-up MRI study.

During the 3 years before final follow-up, acceptable new lumbar MR images had been performed and were available for 13 discography subjects and 8 control subjects. New MR imaging was performed on the remaining subjects who had not received a recent clinical scan; that is, all 52 discography subjects and 50 control subjects had MRI available for comparison (100% enrollment of remaining eligible subjects) (Figure 1).

Figure 1.

Subject flow chart.

Baseline Clinical and Demographic Characteristics of the Study Population

The baseline demographic as well as clinical characteristics of the discography and controls subjects are given in Table 1. These are well matched without important differences between groups. The subjects lost to follow-up did not appear to differ in baseline characteristics from the subjects completing the study.

Reader Agreement

There was agreement in qualitative findings scores between the initial 2 reviewers in 95 of 102 subjects. These remaining seven scans were then evaluated by 2 additional readers and reviewed for analysis of 18 discrepant findings in these seven scans. In all but 3 findings, these 2 additional reviewers agreed with one of the original readers. The remaining 3 items were scored by a consensus reading involving 3 graders (EJC, JC, RH) by conference.

Progression of Disc Degeneration at L3-S1 in the Discography and Control Subjects

Table 2 shows the interval findings in the discography group compared to the control group for the L3-S1 discs. In all graded or measured parameters, discs which had been exposed to puncture and injection had greater progression of new degenerative findings.

There was progression of disc degeneration from normal (Grade 1 and 2) or from moderate (Grade 3 and 4) degeneration in 54 discs (35%) in the discography group compared to 21 (14%) in the control group (P = 0.03). There were 55 new disc herniations of all types in the discography group compared to 22 in the control group (P = 0.0003). Notably, foramenal and far-lateral disc herniations, at the site of anular puncture in the discography group, appear to show the greatest difference between groups.

Similarly there were more new endplate signal changes (P = 0.04) and anular fissures with bright signal (P = 0.1) in the discography group compared to the controls. The quantitative measures of disc height and disc signal also showed significantly greater loss of disc height (P = 0.05) and signal intensity (P = 0.001) in the discography disc compared to the control disc.

Progression of Disc Degeneration at L1-L3 (Noninjected Levels) in Discography and Control Subjects

Comparing the upper lumbar discs (L1/2 and L2/3) in both groups (none of which had discography punctures or injection), there was no apparent difference in qualitative progression of disc degeneration grades, the incidence of new disc herniation, endplate changes or anular fissures. The loss of disc height and disc signal appeared similar in both groups at these noninjected levels (Table 3).

Impact of Needle Size

There were 134 needle injections with a 25-gauge needle and 21 injections with a 22-gauge needle. No statistical differences were found in the progression of degeneration findings between these 2 groups, although there was a trend to greater degeneration in the discs punctured with a 22-gauge needle. This analysis is limited by small numbers of injections using the larger bore needle.

Side of Injections

There were 34 new disc herniation on the ipsilateral side of the disc injection, 14 on the contralateral side and 4 central disc herniations, reflecting a strong statistical association of new pathology to the side of disc injection (χ2 P = 0.0006). Furthermore, the new pathology seemed to be greatest in the foramen and far lateral regions, ipsilateral to the injection (i.e., at the expected side of needle entry). In the control group there did not appear to be any difference (in right or left sided occurrence of disc herniation (10 on the right and 11 on the left) (Table 4).

Painful Disc Injection

There were 32 painful disc injections and 273 negative disc injections during the baseline discography evaluations. There did not appear to be a greater frequency of disc degeneration progression, disc herniation, or new anular fissure findings in injections that were painful with injection when compared with those discs without significant pain on injection. (Table 5) In fact, there was a trend toward more frequent new foramenal or far-lateral protrusions in the discs that were not significantly painful on the baseline injection.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.