Purpose: To retrospectively evaluate flow patterns of interlaminar lumbar epidural steroid injections and compare these patterns to needle tip position. Materials and Method: This study was approved by the institutional review board and complied with HIPAA guidelines. The requirement for informed consent was waived. Fluoroscopic images from 701 consecutive lumbar epidural steroid injections in 485 patients were reviewed. Vertebral level, needle placement, and injectate flow patterns were determined from procedural images. History of previous spinal operations was determined from procedural records. A subset of patients with three or more injections at the same site was also examined. Patterns in the data were evaluated with x2 tests and t tests. Results: Images from 406 lumbar epidural steroid injections, one per patient (214 women, 192 men; mean age, 59.8 years), were used for final analysis. Midline injections were less likely to result in unilateral flow than a more lateral approach (P > .001). Flow within only the posterior epidural space was associated with unilateral flow (P = .001). Patients with previous spinal operations were more likely to have cephalad or caudad flow of less than one vertebral level than patients without a history of spinal operations (P < .001). Flow varied considerably within a series of injections in a single patient. Conclusion: Epidural injectate flow is highly variable, both among patients and between injections in a single patient. Fluoroscopic monitoring and administration of contrast material mixed with medication, allowing visualization of the full extent of medication flow, are essential to ensure adequate coverage of the target area.