The search for value in health care goes on. As we continue to grapple with what should or should not be done to treat disease, rather than what can or cannot be done, we look to rigorous prospective randomized controlled trials to lead us out of the wilderness. In the last decade, such was the case with the proposal, funding, and execution of the Spine Patient Outcomes Research Trial (SPORT). This proved to be a very valuable exercise, and yet many questions about the value of the treatment of lumbar disc herniation remain.
Radcliff et al. present a post hoc subgroup analysis from the Spine Patient Outcomes Research Trial of patients with intervertebral disc herniation and attempt to ascertain how the administration of an epidural steroid injection might affect health-related quality of life. Their hypothesis was that those who received an epidural steroid injection would be more likely to cross over from planned operative treatment to nonoperative treatment. They analyzed both observational and randomized treatment cohorts, and, in my opinion, appropriately excluded those who had received an epidural steroid injection prior to enrollment into the study. The authors of another analysis of the Spine Patient Outcomes Research Trial1 noted poor adherence to randomization in that cohort, and Radcliff et al. appropriately performed an as-treated analysis. In addition, the large proportion of crossovers in the randomized cohort essentially converted them into another observational cohort, so the authors appropriately combined both cohorts for this analysis. Obviously, given the predominantly observational nature of this study, the authors correctly point out that, in addition to baseline mental health differences between the two cohorts, there may be as yet undefined additional confounders of their results.
There were some small but interesting demographic differences in the group that elected to undergo an epidural steroid injection as compared with those who did not; these include lower percentages of patients of white race, and higher percentages of those who were employed, were depressed, or had medical comorbidity. There was also a strong and probably intuitive aversion to an epidural steroid injection exhibited by those with a baseline treatment preference for surgery. A side note of interest to practicing spine surgeons was that there was no difference in operative complications or the rate of revision surgery between the groups with and without an epidural steroid injection.
This analysis showed that patients receiving an epidural steroid injection had no significant outcome differences at one to four years; importantly, the study was adequately powered. The authors did show that those receiving an epidural steroid injection were more likely to avoid surgery, but this finding was strongly confounded by pretreatment preference for surgery. This highlights a very interesting aspect, not previously addressed well, in all studies in which avoidance of surgery is used as a treatment end point, and Radcliff et al. are to be commended for this. As Anderson et al. pointed out, “patients have opinions about treatments and act rationally on them.”2
This study does not dissuade one from opting for an epidural steroid injection in patients with symptomatic intervertebral disc herniation who are otherwise surgical candidates, particularly when the patient has a strong aversion to or contraindications for surgery. Given the explosion of billing for spinal injection procedures3 and the relative lack of treatment effect of epidural steroid injection shown in this study, we are still faced with the value question: is it worth the cost?