INTRODUCTION: The use of discography has been questioned based on the concern of causing harm in the form of accelerated degenerative changes, possibly leading to future surgery. A recent study with minimum 10-year follow-up found no clinical evidence to support this concern, reporting an 8.1% rate of surgery among discs not undergoing discography and a 6.3% rate among discographically normal discs; however, a significantly greater surgery rate (18.4%) was seen among discs that were already abnormal on discography (abnormal image and/or provoking any pain response).[1] This last group has particular implications in addressing the traditional question of including an abnormal appearing disc in the index surgery: is it unnecessary surgery or beneficial to in favor of potentially avoiding additional surgery in the relatively near future. The purpose of this study was to investigate the fate of discs that were not completely normal at the time of discography, performed when evaluating patients for TDR surgery at least 10 years earlier.
METHODS: This study was based on 196 patients who underwent TDR or hybrid surgery (combined TDR/fusion) at least 10 years earlier, and there were details available on the radiographic and pain provocation findings for each injected level. Data were analyzed for each included disc level (n=237). A disc level was included in the analysis if it underwent discography and was not included in the index TDR or hybrid surgery. The rate of re-operation for disc related pain (symptomatic disc degeneration or disc herniation) was calculated and compared based on the CT/discographic images (classified as normal, degenerated (changes throughout the disc), or had a bulge, protrusion, or fissure (termed fissure group).
RESULTS: Based on the radiographic findings, the rate of subsequent surgery was statistically significantly greater in the fissure group 21.7% compared with the degeneration group 12.0% or the normal group 7.2% (p<0.03). With respect to the pain provocation portion of the discogram, the rate of subsequent surgery was significantly greater (p<0.03) among discs with any level of concordant pain (31.6%) than in the discordant pain group (6.9%) or in the no pain group (9.0%). When the images and pain responses were combined, there were statistically significant differences in the subsequent surgery rates (p<0.05), the greatest rate of surgery being discs with both fissure and concordant pain (36.4%).
DISCUSSION: This study found that during minimum 10-year follow-up, the greatest rate of subsequent surgery involved discs that demonstrated fissures/bulges/protrusions combined with some level of concordant pain provocation. The re-operation rates were 12.0% or less among disc with more widespread degeneration, normal images, those with discordant pain, or no pain provocation during discographic injection. Of note, not all combinations of imaging and pain responses could be analyzed in a meaningful manner due to small sample sizes in these subgroups. These results may facilitate decision making in which discs to include in the index surgery and discussion with patients on the rationale for operating, or not, on abnormal appearing discs and the risk of subsequent surgery.