To evaluate the feasibility of MRI-guided discography with optical tracking. 12 consecutive patients who had a clinical suspicion of lumbar discogenic pain and/or suggestive finding of disc degeneration in imaging studies (MRI, CT, plain radiography) underwent MRI-guided discography in order to determine possible pain provocation during puncture and contrast injection. An 0.23 T open configuration MRI device with interventional tools (Outlook Proview, Philips Medical Systems, MR Technologies, Finland) was used in procedural imaging and instrument guidance. An optical guidance tool was attached to the MRI compatible needle (Chiba-type MReye, Cook, Bloomington, IN). After initial disc puncture, 1-2 mL of gadolinium contrast (Magnevist, 469 mg/mL, Schering AG, Germany) saline mixture (1:8) was injected into the disc. Immediately after injection, sagittal FE T1 weighted images were obtained to verify the final position of the needle and formation of the MRI discogram. On nine patients, additional noncontrast sagittal fast spin echo (FSE) T1, FSE T2, and axial 3D T1 gradient echo imaging was performed before and after contrast media injection to obtain MRI discograms. Overall, 35 disc punctures were initialized and 34 MRI discograms were obtained. In all punctures, a positive or negative pain response was obtained. The average time for performing a procedure for three discs was 1 hour 25 minutes (minimum 45 minutes, maximum 2 hours, 15 minutes), and the average number of imaging sequences used for a puncturing one disc was 12. On one disc, the puncture failed and a discogram was not acquired. There was one complication (disc collapse) reported during follow-up. Our results show that MRI guidance in performing discography is accurate and relatively safe. It is a technically comparable method to CT-guidance or fluoroscopy.