Correct placement of epidural steroid injections: fluoroscopic guidance and contrast administration.

We prospectively evaluated 316 caudal-approach epidural steroid injections given by staff radiologists and residents in our department over a 1-year period. Needle placement was checked with fluoroscopy and corrected if necessary. When the needle tip was within the sacral canal, nonionic contrast material was injected. If epidural contrast was not observed, the needle tip was repositioned. Of 111 procedures performed by physicians who had given fewer than 10 epidural steroid injections, 53 (47.7%) resulted in correct nonfluoroscopically directed placement of the needle. For physicians who had performed between 10 and 50 such procedures, 62 (53.4%) of 116 had correct nonfluoroscopically directed placement. For staff physicians, 55 (61.7%) of 89 placements were correct. Even when the sacral hiatus was easily palpated and a staff physician was confident that he or she was within the epidural space, fluoroscopy revealed incorrect placement 14.2% of the time (seven of 49 procedures). In addition, when the needle was positioned within the sacral canal and no blood was evident on Valsalva maneuver or aspiration, the injection was venous in 29 of 316 procedures (9.2%). The presence of blood on the needle stylus was not a reliable indicator of venous placement of the needle. Our findings indicate that fluoroscopy is essential for correct placement of epidural steroid injection. Contrast administration is necessary to avoid venous injection of steroids.