The effect of preoperative lumbar epidural corticosteroid injection on postoperative infection rate in patients undergoing single-level lumbar decompression

Abstract Background Context Lumbar epidural corticosteroid injections (LECIs) are frequently used in the treatment of lumbar intervertebral disc herniation with radiculopathy and lumbar spinal stenosis. Although widely used, their effect on the outcomes and complications of subsequent surgery is unclear. Postoperative infection can be a morbid complication following spine surgery, and recent literature has suggested that the risk may be increased in patients undergoing lumbar spinal surgery who had previously received LECIs. Purpose The purpose of this study is to define the overall postoperative infection rate in patients undergoing lumbar spine decompression surgery in the Military Health System (MHS) patient population and examine the effects of LECIs on postoperative infection rates. Study Design/Setting This is a retrospective case control database study (Level III study). Patient Sample The sample comprised all patients in the MHS who had a LECI before single-level lumbar decompression surgery from 2009 to 2014. Outcome Measures Postoperative infection within 90 days of surgery was used as the primary outcome measure for this study. Postoperative infection was identified using the International Classification of Diseases, 9th revision (ICD-9) diagnosis codes for postoperative infection. Methods The Military Health System Data Repository (MDR) database was searched for all patients who underwent single-level lumbar spine decompression surgery from 2009 to 2014 using Current Procedural Terminology (CPT) codes. Current Procedural Terminology codes were used to identify the subset of patients who received preoperative LECIs. For patients receiving an injection, cohorts were established based on the timing of the preoperative injection: 365 days. An age-based cohort, composed of patients 65 years of age and older, was also analyzed. A subgroup analysis of patients receiving more than one preoperative injection was performed. Postoperative infection within 90 days of surgery was identified using ICD-9 codes, and infection rates for all groups were calculated and compared with the control group who did not receive preoperative LECIs. No external funding was received for this study. Results We identified 6,535 patients (847 preoperative LECI and 5,688 control) for analysis. The overall infection rate for patients undergoing single-level lumbar decompression surgery in the MHS was 0.81%. The rate ranged fr