Do Epidural Steroid Injections Affect the Outcome of Patients Treated for Lumbar Stenosis? A Subgroup Analysis of the SPORT
Introduction: In this subgroup analysis of the SPORT study, the researchers worked from this hypothesis: patients who received epidural steroid injections (ESI) as part of their lumbar stenosis or degenerative spondylolisthesis treatment would, when compared to those patients who did not receive ESI, have improved outcome and a lower rate of crossover to surgery.
Methods: Patients included in the subanalysis were those enrolled in SPORT who had not received ESI before enrolling. A comparison was made between patients receiving ESI during the first 3 months of the SPORT study (ESI group) and those patients who did not receive an ESI (No ESI group).
At baseline, 1 year, 2 years, 3 years, and 4 years, changes in primary and secondary outcome measures, as well as treatment effect of surgery, was assessed.
Results: The spinal stenosis cohort had 69 patients in the ESI group and 207 in the No ESI group. Between those groups, there were no significant differences in demographics, baseline clinical outcome scores, or operative details. In the ESI group, though, an increased preference for nonoperative treatment was seen (p < 0.001).
For the ESI patients who received surgery (n = 41), there was an average of a 26-minute increase in time in the OR; there was also an increased length of stay—0.9.
Using the SF-36 Physical Function (SF-36 PF), there was significantly less improvement in the ESI patients averaged over 4 years than in the No ESI patients (ESI 14.8 vs No ESI 22.5, p = 0.025). Significantly less improvement was also seen in the SF-36 Bodily Pain (SF-36 BP) (ESI 7.3 vs No ESI 16.7, p = 0.007).
For those spinal stenosis patients who were assigned to surgery at enrollment, researchers saw an increased crossover rate associated with ESI to nonsurgical treatment (ESI 32% vs No ESI 11%, p < 0.012). Also, increased crossover rates were seen in ESI patients assigned to nonsurgical treatment (ESI 58% vs No ESI 32%, p = 0.003).
The degenerative spondylolisthesis cohort had 118 No ESI patients and 45 ESI patients. At baseline, there were some significant differences between the groups in: SF-36 BP, SF-36 Mental Component Scores (SF-26 MCS), and leg pain bothersomeness index. These differences favored the No ESI group.
As with the spinal stenosis group, though, there were no significant differences in baseline demographics and the ESI group showed an increased preference for nonsurgical treatment ( p < 0.001).
In the degenerative spondylolisthesis patients, the nonsurgically-treated group had significantly less improvement (averaged over 4 years) in SF-26 BP (ESI 7.8 vs No ESI 17.8, p = 0.005), SF-36 PF (ESI 4.3 vs No ESI 13.1, p = 0.005), and sciatica bothersomeness index (ESI -2.4 vs No ESI -4.1, p = 0.041).
The ESI group showed a trend at 4 years for increased re-operation, but there was no significant difference in crossover rates with ESI in the degenerative spondylolisthesis group.
Conclusions: Even though these groups (ESI and No ESI) had equivalent baseline status, the ESI group had significantly less improvement, and they had no evidence of surgical avoidance in the long-term. While epidural steroid injections may provide short-term improvement, they are associated with poorer long-term outcomes.
This subgroup of the SPORT study provides some intriguing data against the use of epidural steroids in patients with lumbar stenosis. They cite worse outcomes in surgical and nonsurgical patients who received epidurals, even at 4-year follow-up. Epidurals were associated with a higher rate of crossover in both groups, which as mentioned previously was the major criticism of the SPORT study design as a whole.