Objective To explore the safety, efficacy and outcomes of posterior transforaminal spinal cord reduction in the treatment of severe thoracic ossification of the posterior longitudinal ligament. Methods Since May 2010 to May 2016, 21 patients with thoracic spinal stenosis with ossification of the posterior longitudinal ligament and embedding rate of more than 50% underwent posterior circular decompression, 12 males and 9 females; age 42-71 years old, with an average of 63.3±1.6 years old. There were severe spinal cord compression symptoms before surgery. The operation was from the posterior median approach. First, the pedicle screw tract was preset in the upper and lower segments where decompression was required, and then the posterior wall of the spinal canal was removed by the uncovering method to complete the posterior decompression. Through the superior and inferior articular joints, the joints were removed, the intercostal nerves were pushed outward, the extracorporeal intervertebral space and intervertebral disc tissue were revealed, and the vertebral cancellous bone of the lower vertebral body was removed. Separate the adhesion of the dura mater, use the special tool to cut the upper and lower edges of the posterior longitudinal ligament, collapse, remove the posterior longitudinal ligament ossification block, and complete the decompression of the front of the spinal cord through one or both sides. Finally, the pedicle screw fixation and the intervertebral bone graft were completed. Postoperative follow-up (mean 38.5 months) CT and MRI scans were performed to observe the patency of the spinal canal and the state of the internal fixation. The Frankel graded evaluation function was restored, the JOA score was used to evaluate the neurological function, the Otani evaluation system was used to evaluate the surgical outcome, the clinical excellent rate was calculated, and the treatment effect was evaluated. Results Nineteen cases recovered after operation, and 1 case showed a transient decrease in unilateral lower extremity motor function on the second day after surgery. It returned to normal after dehydration and neurotrophic support treatment, and 1 case had bilateral lower limbs on the day after surgery. Muscle strength was transiently lost, feeling recovered after 6 hours, and exercise resumed after 48 hours. The operation time was 180-300 min, with an average of 240±23 min. The intraoperative blood loss was 168±12 ml. Postoperative spinal cord function recovered significantly. At the last follow-up, Frankel graded F in 8 cases and grade E in 13 cases. Preoperative average JOA score was 3.5, and the JOA score of the follow-up was increased from 3.5±0.98 points to 9.0±0.9 points. The improvement rate of JOA score was excellent in 14 cases, good in 5 cases, and fair in 2 cases. According to Otani system there was excellent in 14 cases, good in 6 cases, and fair in 1 case. The excellent and good rate was 95.24%. Cerebrospinal fluid leakage occurred in 3 cases and healed after symptomatic treatment. No chest occurred during the follow-up period. The vertebral instability, internal fixation loose, broken. Conclusion Posterior transforaminal spinal decompression of posterior longitudinal ligament ossification of thoracic spine can be finished using self-developed special surgical tools, which is safe, effective and can achieve full decompression of spinal cord. Key words: Thoracic vertebrae; Spinal stenosis; Ossification of posterior longitudinal ligament; Decompression, surgical; Spinal fusion