The nerve damage caused by free lumbar disc herniation is more serious, with poorer prognosis and diversified clinical manifestations, which is a more serious type of lumbar disc herniation and requires surgical treatment, but it is difficult to operate and prone to leakage of nucleus pulposus and nerve damage. The author retrospectively analyzed 26 cases of patients with free lumbar disc herniation surgically treated in our hospital from July 2007 to April 2010, and achieved good clinical results, reported as follows. Clinical data The 26 cases in this group included 22 males and 4 females, aged 24-81 years old, with an average of 49 years old. The average age of the patients was 49 years. The duration of the disease ranged from 6d to 3 years, with an average of 5.7 months. There were 5 cases of L3/4, 9 cases of L4/5, 7 cases of L5/S1, 2 cases of L3/4 and L4/5, and 3 cases of L4/5 and L5/S1. 23 cases had recurrent lumbar and leg pain, which was aggravated suddenly, with decreased sensation and muscle strength in corresponding innervated areas, and 1 of which was accompanied by numbness of the perineum and difficulty in urination and defecation; 4 cases showed obvious decrease in the muscle strength of the lower limbs, decreased sensation, and inability to walk; 3 cases had only lower limb pain; and 3 cases had only lower limb pain, with decreased sensation, and inability to walk. Among them, one case was accompanied by perineal numbness and difficulty in urination and defecation; four cases showed obvious weakening of muscle strength of both lower limbs, sensory loss and inability to walk; three cases only had radiating pain in the lower limbs, and were unable to stand or walk for a long time, accompanied by sensory and muscle weakness of the lower limbs; 18 cases had positive straight leg raising test. After admission, lumbar spine front and side view films, power view films, CT and MRI were taken. Imaging examination showed that there were 9 cases of giant central lumbar disc herniation and 8 cases of extra-radicular lumbar disc herniation; 5 cases had posterior displacement of the upper vertebral body and 21 cases had posterior displacement of the lower vertebral body; 6 cases had a free distance of half of the vertebral body, 1 case had a free nucleus pulposus that broke through the dural sac and reached the surface of the arachnoid membrane, and no free nucleus pulposus could be seen in the subarachnoid space. Six cases were associated with spinal stenosis, and five cases were associated with non-free disc herniation in the adjacent upper or lower intervertebral space. Treatment Surgery was performed under general anesthesia or combined subarachnoid and epidural anesthesia. The preoperative design was based on the clinical characteristics and imaging manifestations of the patients, and the surgical principles of three-dimensional localization, layered exploration, adequate exposure, fine separation, and reduction of injury were followed during the operation. Unilateral openings were used in 5 cases, unilateral enlarged openings were used in 6 cases, hemilaminectomy was used in 11 cases, and total laminectomy was used in 4 cases, of which 13 cases were fixed with pedicle screws after decompression, including 7 cases of lumbar disc herniation with extra-root canal, and 6 cases of giant disc prolapse with more than half of the vertebral body, of which 12 cases had intervertebral fusion, and one case of posterior posterolateral fusion. After surgery, all cases were given hormones and antibiotics for 3d, and after 3d of bed rest, X-ray films were reviewed, and they wore a waist cuff to get out of bed and avoided physical activities for 3 months. The VAS score and JOA lumbar pain scale were used to score the pain and neurological function at preoperative, 2 weeks, 6 months and the last follow-up. Results The patients were followed up for 6-36 months, with an average of 14 months and 2 cases were lost. None of the patients had complications such as cerebrospinal fluid leakage, infection, neurological decompensation, etc. The surgical incisions healed in one stage. The results of VAS and JOA scores at different times before and after surgery are shown in Table 1. The difference between VAS and JOA scores at 2 weeks after surgery and before surgery was significant (P<0.05); the difference between VAS and JOA scores at 6 months after surgery and at the last follow-up was significant (P<0.05); and the difference between VAS and JOA scores at the last follow-up and at 6 months after surgery was insignificant (P>0.05), but the difference between JOA scores was significant (P>0.05). At the last follow-up compared with 6 months after surgery, the difference in VAS score was not significant (P>0.05), but the difference in JOA score was significant (P<0.05).