Preoperative epidural injections are associated with increased risk of infection after single-level lumbar decompression

Background Context Lumbar epidural steroid injections (LESIs) are often performed as a treatment option for lumbar stenosis and radiculopathy before lumbar decompression surgery. Several case series have reported spinal infections after LESIs. There is lack of literature on the rate of postoperative infections after lumbar decompression in patients who had prior LESIs. Purpose The goal of the present study is to employ a large national database to determine if there is an association between preoperative LESIs at various time intervals before lumbar decompression and the incidence of postoperative infection. Study Design/Setting Retrospective case control database study, Level III was used in this study. Patient Sample This study comprised Medicare patients over age 65 years who had a LESI within 1 year of single-level lumbar decompression surgery. Outcome Measures International Classification of Diseases, 9th Revision diagnosis codes for postoperative infection and Current Procedural Terminology procedure codes for treatment of postoperative infection were the outcome measures for this study. Methods The PearlDiver Patient Records Database, an insurance-based database of patient records, was used for this study. The database was queried for LESI and single-level lumbar decompression procedures using Current Procedural Terminology codes. These study patients were then divided into four separate cohorts: (1) lumbar decompression within 1 month following LESI, (2) lumbar decompression between 1 and 3 months following LESI, (3) lumbar decompression between 3 and 6 months following LESI, and (4) lumbar decompression between 6 and 12 months following LESI. Unique control groups for each study cohort were created with patients who underwent single-level lumbar decompression without previous LESI and matched for major risk factors for infection, including age, gender, smoking status, diabetes, and obesity. Results Overall, the rate of postoperative infection after single-level lumbar decompression after LESI remained relatively low, ranging between 0.8% and 1.7%. The incidence of 90-day postoperative infection after lumbar decompression was significantly higher than matched controls in groups with LESI within 1 month (OR=3.2, p<.0001) and 1-3 months before surgery (OR=1.8, p<.0001). The incidence of 90-day postoperative infection was not significantly different from matched controls in groups with LESI between 3-6 months (OR=1.3, p=.15) and 6-12 months before decompression surgery (OR=1.3, p=.18) Conclusions Single-level lumbar decompression within 3 months after LESI may be associated with an increased rate of postoperative infection. Increasing the time interval between LESI and single-level lumbar decompression surgery to at least 3 months may decrease postoperative infection rates.