During lumbar spinal fusion, spacer cages are implanted to provide vertebral stability, restore sagittal alignment, and maintain disc and foraminal height. Polyetheretherketone (PEEK) is commonly used by most spine surgeons. Silicon nitride (Si3N4) is a less well-known alternative although it was first used as a spacer in lumbar fusion over 30 years ago. The present study was designed to see if Si3N4 cages would perform similarly to PEEK in a randomized controlled trial. A non-inferiority multicenter 100-patient study was designed where both the observer and patient were blinded. Single- or double-level transforaminal lumbar interbody fusion with pedicle screw fixation using an oblique PEEK or Si3N4 cage was performed. The primary non-inferiority outcome was the Roland-Morris Disability Questionnaire (RMDQ). Secondary measures included the Oswestry Disability Questionnaire, Visual Analogue Scales (VAS) for back and leg pain, SF-36 Physical and Mental Function indices, patient and surgeon Likert scores on perceived recovery, and X-ray and CT radiological evaluations for subsidence, segmental motion, and fusion. Follow-up evaluations occurred at 3, 6, 12, and 24 months. After exclusions for protocol violations and canceled surgeries, 92 patients were randomized (i.e., 48 for PEEK and 44 for Si3N4). There were no differences in baseline demographics, pre-operative disabilities, or pain scores between the groups. Both treatment arms showed significant improvements in disability, pain, and recovery scores. No significant differences were observed for subsidence, segmental motion, or fusion. For the primary outcome (i.e., RMDQ scores), the non-inferiority of Si3N4 compared to PEEK could not be established using the original protocol criteria. However, the comparison was undermined by larger than anticipated patient fallout coupled with higher than expected RMDQ score standard deviations. A post hoc analysis coupled with a more extensive review of the literature was conducted which resulted in the selection of a revised clinically justified non-inferiority margin; and using this method, the non-inferiority of Si3N4 was affirmed. This study demonstrated that the use of either PEEK or Si3N4 cages is safe and effective for patients undergoing lumbar spine fusion for chronic degenerative disc disease.