Objective To evaluate the clinical effect of thoracic discectomy via trans-facet-joint approach in the treatment of thoracic disc herniation.Methods Thirty-three cases were included in this group from October 1994 to August 2009.There were 27 males and 6 females.The age ranged from 18 to 72 years old,with an average of 41.8 years.The course of disease ranged from 12 days to 36 months and was lesser than one month in 13 cases.The weakness and numbness of lower limbs occurred after trauma in 9 cases.Fifteen cases were diagnosed as simple thoracic disc herniation.Six cases were associated with ossification of posterior longitudinal ligament and 12 cases were associated with ossification or hypertrophy of yellow ligament.A total of 45 discs were involved,including 32 in lower thoracic segments(71.11%),8 in upper thoracic segments(17.78%)and 5 in middle thoracic segments(11.11%).All the herniated discs and the ossified OPLL were excised via the trans-facet-joint approach.For the cases with ossification or hypertrophy of yellow ligament,the laminectomy and replantation were performed.The screw-rod system was used on both sides in 14 cases,on one side in 19 cases.Results Follow-up was acquired in 27 patients,ranged from 12 to 63 months(mean,37 months).According to Epstein and Schwall grade,there were excellent in 15 cases,good in 10 cases,improved in 2 cases and poor in 2 cases.The excellent and good rate was 86.21% and total effective rate was 93.10%.Postoperative complications occurred in 3 cases,including exacerbation of preexisting deficits in 2 cases and implant failure in 1 case.The former 2 cases were treated with methylprednisolone,dehydrant,neural nutrition and hyperbaric oxygen.One patient had recovered to preoperative level,the other had not recovered to the preoperative level.The implant was removed 18 months after operation for the implant failure.The post-operative CT or MRI showed that all the replanted lamina obtained fusion,and the canal decompression was complete.Conclusion Thoracic discectomy via trans-facet-joint approach can improve the clinical result obviously.

Objective To observe the clinical advantage of treating the dens fracture with cannulated screws by microendoscopy system and navigation system through anterior approach. Methods From October 2002 to August 2004, 6 dens fracture (type Ⅱ) patients had undergone single cannulated screws fixation by the microendoscopy system and image guidance through the anterior approach, including 2 females and 4 males with an average of 53.2 years (range from 23 to 65 years). The preoperative Frankel grade was D in 1 case and E in 5 cases. 4 cases had fracture displacement. All patients accepted skull traction and external fixation before surgery. The patients accepted the MR examination and the images were reconstructed by the Vector Vision spinal navigation system in order to make the proper surgical plan, including the nails' direction and length in three-dimension. After affirming the reduction of the dens fractures according the C-arm fluoroscopy, the work channel was inserted through 2 cm long incision at the C5 level, then extended to the C2 level, to excise part of inferior C2 vertebral body. Under the navigation system guide, a guidance needle was inserted and the fracture was fixed by cannulated screw through the work channel. The patients did not need any C-arm fluoroscopy during the surgery. After the surgery the patients were with external fixation for 2 months. Results The surgery lasted 80 min (range from 50 to 150 min), blood loss was 75 ml (40 to 90 ml). There was no related complication during and after surgery including odynophagia or dysphagia. Follow-up period extended from 6 to 37 months (mean 13.5 months). All dens fractures were reduced and healed satisfactory during the follow up. All patients' postoperative Frankel grade was E in the most recent follow-up. Conclusion To fix the dens fracture through microendoscopy and navigation system guide can reduce the soft tissue injury and the incidence of odynophagia or dysphagia, make the surgery safer, avert the repeated C-arm check-up during the surgery compared with the traditional anterior fixation method. The procedure is very safe, simple and can reduce the surgical time effectively.

Objective To analyze the clinical results of mini-incision Schwab grade 4 osteotomy combined with percutaneous pedicle screw fixation for old thoracolumbar compression fractures combined with kyphosis deformity.Methods A retrospective case series study was made on 14 patients with old thoracolumbar compression fractures combined with thoracolumbar kyphosis undergone mini-incision Schwab grade 4 osteotomy plus percutaneous pedicle screw fixation from January 2014 to May 2015.There were 5 males and 9 females,with mean age of 64.5 years (range,58-70 years).The period between injury and surgery ranged between 6 and 14 months (mean,8.6 months).At the time of surgery,the fracture was already healed in 10 patients,while non-healing was found in other 4 patients.Preoperative visual analogue score (VAS) of back pain was (8.5 ± 1.1)points (range,6.5-10 points).Three patients were associated with neurological dysfunction [American Spinal Injury Association (ASIA) grade D].Apex of kyphosis located at T12 in 6 patients and at L1 in 8 patients.Preoperative kyphosis Cobb angle was (42.5 ± 6.0)° (range,39.5°-47.2°).Operation time,blood loss,perioperative complications,postoperative kyphosis Cobb angle,bone fusion,state of implants,neurological function and VAS were determined.Results Operation time was (280 ± 50) min,and blood loss was (110 ±70)ml.No segmental vessels injury,neurological deficit or dural disruption occurred during the surgery and after surgery.Compared to the preoperative detection,kyphosis Cobb angle was improved to (9.3 ±1.7) ° (range,6.2°-12.1 °) after operation (P ＜ 0.05),with the correction rate of 78.1％.Postoperative CT showed 7 screws (6.3％,7/112) were grade 1 screws.The follow-up was lasted for mean 25.2 months (range,14-28.9 months).At the latest follow-up,the kyphosis Cobb angle was (9.6 ±4.1)°(range,6.0°-13.1°),revealing no correction loss.Interbody bone fusion was good,with no instrumentation-related complications observed.Three patients with neurological dysfunction (ASIA grade D) were recovered to ASIA grade E.Compared to the preoperative detection,back pain was improved with the VAS of (2.6 ± 1.0) points at the latest follow-up (P ＜ 0.05).Conclusion Mini-incision Schwab grade 4 osteotomy combined with percutaneous pedicle screw fixation of old thoracolumbar compression fractures with kyphosis deformity can attain satisfactory and reliable efficacy and bone fusion,with low incidence of complications.

Objective To compare the clinical outcome of minimally invasive transforaminal lumbar interbody fusion (M-TLIF) and Wiltse-approach TLIF (W-TLIF) in treating single-level degenerative lumbar disease.Methods A retrospective review was performed on the 57 patients with single-level degenerative lumbar disorder managed via M-TLIF (n =27) and W-TLIF (n =30) from December 2009 to December 2010.In M-TLIF group degeneration at the L4-5 disc were noted in 11 cases and at the L5-S1 disc in 16 cases.And 19 cases were diagnosed with lumbar isthmus spondylolisthesis (17 with Grade Ⅰ spondylolisthesis and 2 with Grade Ⅱ spondylolisthesis),4 lumbar spinal stenosis and instability,2 lumbar disc herniation combined with huge posterior osteophytes,1 recurrent lumbar disc herniation after lumbar fenestration,and 1 recurrent lumbar spinal stenosis after decompression.In W-TLIF group degeneration at L4～5 disc were noted in 12 cases and at the L5-S1 disc in 18 cases.There were 19 cases diagnosed with lumbar isthmus spondylolisthesis (18 with Grade Ⅰ spondylolisthesis and 1 with Grade Ⅱ spondylolisthesis),3 with lumbar disc herniation,and 8 with lumbar spinal stenosis.Japanese Orthopedic Association (JOA) score and Visual Analogue Scale (VAS) were used to measure low back and leg pain.Modified Brantigan score was used to assess lumbar interbody fusion.Results Operative time was not significantly different between the two groups (P ＞ 0.05).Incision length and mean blood loss were (5.1 ± 0.7) cm and (90.1 ± 10.5) ml in M-TLIF group,but were (6.9 ± 1.0)cm and (155.3 ±21.2)ml in W-TLIF group (P＜0.05).At postoperative 1 and 3 days VAS in M-TLIF group was (2.1 ± 0.5) points and (1.0 ± 0.1) points respectively,but in W-TLIF group was (3.6 ± 0.1) points and (2.4 ± 1.0) points respectively (P ＜ 0.05).Intraoperative X-ray fluoroscopy frequencies were (46 ± 9) times in M-TLIF group and (7 ± 2) times in W-TLIF group (P ＜ 0.05).Mean period of follow-up was 26.7 months (range,24-36 months).At final follow-up,JOA score,VAS for leg pain and lumbar interbody fusion rate revealed no significant differences between the two groups (P ＞ 0.05),but VAS for lumbar pain was (1.0 ± 0.2) points in M-TLIF group versus (1.9 ± 0.3) points in W-TLIF group (P ＜0.05).Twenty-four cases (89％) had Brantigan score of 3 or over in M-TLIF group and 27 cases (90％) in W-TLIF group,indicating a similar interbody fusion rate (P ＞ 0.05).Conclusions Both lumbar fusion methods are effective in treatment of lumbar degenerative disease.M-TLIF lumbar fusion results in small amount of bleeding,small incision and significantly improved lower back pain as compared with W-TLIF,but W-TLIF involves less exposure to the X-rays.

Objective To review and analyze the clinical effect of combined posterior mini-invasive fixation with anterior debridement via small incision for the treatment of single segment lumbar vertebral tuberculosis. Methods Totally 31 cases with single segment lumbar tuberculosis (both borderline tuberculosis) without attachment involvement underwent one-stage anterior debridement, interbody fusion and posterior mini-invasive fixation from July 2010 to July 2015. Among these patients, 19 were male and 12 were female. The average age was (36.1±17.8) years old (ranged 21-61 years old). The average course of disease was 11(9, 12) months (ranged from 2 to 16 months). All were single segment involvement, and the involved segment was L2-3 in 7 cases, L3-4 in 10 cases, L4-5 in 6 cases, and L5-S1 in 8 cases. The clinical manifestations included lumbar back pain in 31 cases with an average pain visual analog score (VAS) of 7(6, 8) points. ASIA grade of spinal cord injury was E in 25 cases and D in 6 cases. Paravertebral abscess occurred in 22 cases and iliac fossa gravity abscess appeared in 9 cases. Kyphosis was observed in 12 cases and the average Cobb angle was 21° ± 6° . Quadruple anti-tuberculosis chemotherapy was used for at least 2 weeks preoperatively. Posterior mini-invasive fixation was fulfilled on prone position, including mini-invasive percutaneous screws in 18 cases and pedicle screw fixation via Wiltse approach in 13 cases. Posterior distraction and deformity correction were performed simultaneously for patients with kyphosis. Then the patients were changed to lateral position for anterior debridement, bone grafting and/or titanium mesh fusion. Results The average operation time was (204±54) min (ranged 160-240 min) in 31 patients, and the mean blood loss was (168±73) mL (ranged 100-300 mL). Delayed healing of anterior incision occurred in 1 case and the incision healed after two-week dressing of wound. The incision healed well in the rest 30 cases. No complications such as nerve function, blood vessel injury were found in patients. The VAS scores of the 3 days after operation were 1.3 ± 0.3 and 2.1 ± 1.4 in percutaneous group and Wiltse approach group, respectively, and the difference between them was statistically significant ( P<0.05). The VAS score of low back pain was 2(1, 3) points in all the 31 patients three months after operation, which was significantly lower than that before surgery (P<0.05). The six patients with neurological symptoms recovered to E grade after operation. The average Cobb angle correction was 15°±5° in 12 patients with kyphosis (P<0.05), which was significantly decreased compared with that before surgery (P<0.05). All patients were followed up for an average of (36.8 ± 9.3) months (ranged from 12 to 72 months). The clinical healing of tuberculosis was achieved at the final follow-up in all the 31 patients. No complications were observed, such as lumbar kyphosis, internal fixation loosening and breakage, dislocation and titanium mesh subsidence. Conclusion Mini-invasive posterior internal fixation and anterior debridement via small incision is effective for the treatment of single segment lumbar vertebral tuberculosis in lesion debridement and spine stability reconstruction by short segment fixation. This technique can reduce fused segments, surgical trauma of anterior approach and related complications.

Objective To describe the technique and therapeutic effect of modified unilateral laminotomy for bilateral decompression (M-ULBD) for lumbar spinal stenosis (LSS).Methods A total of 56 patients with LSS were randomly divided into group A and B.The 27 patients in group A (15 males and 12 females,with an average age of 59.4 years) underwent M-ULBD.The other 29 patients in group B (18 males and 11 females,with an average age of 61.6 years) received conventional laminectomy.JOA score of low back pain,VAS,CPK three days after operation,pre-and post-operative cross-sectional areas of multifidus were used to evaluate the clinical results.Results A total of 45 patients (21 in group A and 24 in group B) completed 2 years of follow-up.The preoperative VAS of low back pain,leg pain,numbness,JOA score and cross-sectional areas of multifidus were 5.6±1.7,7.1±0.4,11.6±2.6,5.8±1.8 cm2 in group A and 6.2±1.2,7.9±1.3,10.9±1.0,6.1±2.0 cm2 in group B.There was no significant difference in preoperative data between both groups.The union of split spinous process was observed in all cases 6months later according to computed tomography.The postoperative CPK was lower in group A.The postoperative JOA and VAS scores in both groups were improved significantly compared with the corresponding preoperative ones.The VAS of leg pain,numbness,JOA score,and JOA recover rate in latest follow-up were 1.3±0.2,1.5±0.7,26.7±2.1,86.1％±3.1％ in group A,and 1.7±0.3,2.0±1.3,24.3±2.5,83.6％±6.4％ in group B,respectively.All these data have no difference between group A and B.The VAS of low back pain and atrophy rate of multifidus were 1.0±0.5,6.4％±1.2％ in group A,and 2.6±0.7,15.7％±3.0％ in group B respectively.All these data are lower in group A.Conclusion Our two years follow-up shows that this method is efficient for lumbar spinal stenosis treatment,however,it still need long term follow-up and to compare with other modified methods.

Objective:To investigate and compare the biomechanical strength of paravertebral foramen screws (PVFS), lateral mass screws (LMS) and pedicle screws (PS).Methods:A total of 30 human cervical spine vertebrae (C 3-C 6) were harvested from 8 fresh-frozen cadaver specimens whose mean age was 45.3±11.2 years at death. The vertebrae were randomly divided into three groups for specific screws. For each vertebra, one side was randomly chosen for direct pullout strength test (speed 5 mm/s), and the other side for fatigue test (displacement ±1.0 mm, frequency 1 Hz, 500 cycles) and residual pullout strength test. 4.5 mm × 12 mm screws were used for PVFS, 3.5 mm × 14 mm screws for LMS, and 3.5 mm × 24 mm screws for PS. Results:The direct pullout strength was 327.10±17.07 N for PVFS, 305.71 ± 11.63 N for LMS, and 635.67 ± 22.82 N for PS. The residual pullout strength was 265.62 ±18.19 N for PVFS, 192.80 ±17.10 N for LMS, and 494.89 ±41.79 N for PS. The residual pullout strength of PVFS, LMS and PS respectively, compared with the direct pullout strength, decreased by 18.8%, 36.93% and 22.15% ( tPVFS=7.795 , tLMS=17.267 , tPS=9.349 , P<0.001). The direct pullout strength of PS was higher than that of PVFS and LMS( t=34.245, t=40.741, P< 0.001), as well as PVFS was slightly higher than LMS ( t=3.275, P=0.004). The residual pullout strength of PS was the highest, PVFS was the second, and LMS was the smallest ( F=314.619, P<0.001). For the fatigue test, the load at the first cycle and the first time when the set position was reached of PVFS were higher than those of LMS ( t=3.625, P=0.002; t=5.388, P<0.001) and PS ( t=2.575, P=0.019; t=2.680, P=0.015), but there was no difference between those of LMS and PS ( t=0.609 , P=0.550; t=1.953 , P=0.067). The load at the last cycle of PVFS and PS was higher than that of LMS ( t=5.341 , P<0.001 ; t=3.439 , P=0.003), while there was no difference between PVFS and PS ( t=1.606, P=0.126). Conclusion:The direct pullout strength of PVFS was slightly higher than that of LMS, and the residual pullout strength was significantly higher than LMS. The property of fatigue resistance of PVFS was similar to PS and obviously better than LMS. In summary, PVFS can be used as an effective substitute for LMS and PS.

Objective To analysis causes of surgical failure of percutaneous lumbar endoscopic discectomy (PLED) for lumbar degenerative diseases.Methods Forty-six patients (31males,15 females),who underwent unsuccessful PLED (including percutaneous transforaminal endoscopy discectomy,PTED;percutaneous interlamina endoscopy discectomy,PIED) or percutaneous endoscopic lumbar decompression,were included in this study.Unsuccessful surgeries included no relieve of lumbar and limb pain and numbness right after surgeries;aggravated after surgeries that need revision 1～3 moths after surgeries;new symptoms appeared after surgeries;still had severe low back pain (VAS ＞5 points) 3 months after surgeries;had recurrence of lumbar disc herniation at the same level.The average age was 46±11 years old (20-81 years old).The primary diagnosis was lumbar disc herniation in 43 cases,and lumbar spinal canal stenosis in 3 cases.Forty-two cases accepted single level surgeries,others accepted twolevel surgeries.One case underwent PLED twice,others underwent one-time surgery.Results The causes of surgical failure included misdiagnosis in 10 cases,inappropriate surgical indication in 10 cases,inappropriate surgical technique in 12 cases,recurrent disc herniation in 9 cases,and persistent low back pain in 6 cases.Misdiagnosis cases included avascular necrosis of femoral head in 2 cases,missed diagnosis of cervical myelopathy in 1 case,mental disorder in 1 case,severe central spinal canal stenosis in 3 cases,and unidentified diagnosis in 3 cases.Inappropriate surgical indication cases included performing PLED for severe central spinal canal stenosis in 3 cases,PLED for only low back pain in 6 cases,untreated responsible disc herniation at adjacent level in 1 cases.Inappropriate surgical technique cases included incomplete removal of protruded disc in 11 cases,nerve root injuryin 1 case.Conclusion The causes of surgical failure of PLED mainly included misdiagnosis,inappropriate surgical indication,incomplete removal of protruded disc,and recurrent disc herniation.Improving diagnosis and indication selecting ability may help to avoid surgical failure.

Objective: To compare the incidence of adjacent segment degeneration (ASDeg) and clinical outcomes of minimally invasive versus traditional transforaminal lumbar interbody fusion (TLIF) in the treatment of L _4,5 single-segment lumbar spinal stenosis (LSS) and explore the risk factors of ASDeg. Methods: All of 115 patients with LSS who were treated by the same group of doctors from 2009 to 2013, with a minimum follow-up of 5 years. Thirty-eight patients underwent minimally invasive trans-foraminal lumbar interbody fusion (MIS-TLIF) and 77 patients underwent traditional TLIF. Standing radiographs at the preopera-tive period and the final follow-up were assessed. Radiological parameters included lumbar lordosis (LL), fused segment angle (FSA), disc height (DH) and range of motion (ROM). Babu classification was used to identify facet joint violation (FJV) in patients at 5-year follow-up. Clinical outcomes were assessed according to visual analog scale (VAS) score, Japanese Orthopaedic Associa-tion (JOA) score and Oswestry Disability Index (ODI). Student's t-test, Chi-square test, and non-parametric test were used as the main statistical methods. Results: The mean age of MIS-TLIF group was 58.2±8.8 years, and that of TLIF group was 54.7±11.2 years, and there was no significant difference between the two groups. The mean follow-up time was 64.5±3.8 months in the MIS-TLIF group and 63.9±3.3 months in the TLIF group, and there was no significant difference between the two groups. There were 17 cases of degenerative spondylolisthesis in MIS-TLIF group (44.7%) and 35 cases of degenerative spondylolisthesis in TLIF group (45.5%), and there was no significant difference between the two groups. There was no significant difference in DH and ROM of L_3,4, L_4,5, L₅S₁ between the two groups before operation. There was no significant difference in VAS, JOA and ODI scores between the two groups before operation. The VAS, JOA and ODI scores were significantly improved at the last follow-up compared with those before operation. After 5-year follow-up, 56 cases (48.7%) had ASDeg. The incidence of ASDeg was 31.6% in MIS-TLIF group and 57.1% in TLIF group, and there was statistical differences between the two groups (χ²=6.656, P <0.01). Among them, 32 cases only had upper segment ASDeg (6 cases in MIS-TLIF group, 26 cases in TLIF group), 19 cases only had lower segment ASDeg (6 cases in MIS-TLIF group, 13 cases in TLIF group), and 5 cases had both upper and lower ASDeg (5 cases in the TLIF group). The DH of adjacent segments decreased after operation, but the loss of DH in MIS-TLIF group was smaller than that in TLIF group, including L_3,4 segments (-4.9%±6.4% vs-8.7%±7.2%, t=-2.761, P <0.01), L5S1 segment (-4.7%±9.8% vs-10.5%±11.7%, t=-2.623, P <0.01). The ROM of adjacent segments increased in both groups, but the increase of ROM in MIS-TLIF group was smaller than that in TLIF group, including L _3,4 segments (1.1°±1.8° vs 2.3°±2.5°, t=-3.122, P <0.01), L₅S₁ segment (0.9°± 1.9 ° vs 1.8°±1.9 °, t=-2.353, P <0.01). The incidence of FJV was 54.2% in patients with ASDeg in MIS-TLIF group and 47.7% in patients with ASDeg in TLIF group. Chi-square analysis showed that FJV was related to ASDeg in both groups (χ²=3.869, P < 0.05). Conclusion Both of the two surgical methods have good clinical effects on L _4,5 single-segment LSS. The incidence of AS-Deg after MIS-TLIF is lower than that of TLIF. FJV is a risk factor for ASDeg.

Objective: To analyze the necessity of routinely performing foraminoplasty during percutaneous transforaminal endoscopic discectomy (PETD). Methods: A total of 412 patients including 231 males and 181 females with an average age of 39.1±13 (20-80) years were enrolled in the present study. All patients were preoperatively diagnosed with single-segment lumbar disc herniation and underwent PETD by the same surgical group. The affected segments were at L_3-4 in 32 cases, L_4-5 in 289 cases, and L₅S₁ in 91 cases. Among them, 306 cases had no prolapse, 89 had mild up/down prolapse, and 17 had severe prolapse. MRI sagittal imaging was used to measure the height and width of the intervertebral foramen of L_{3, 4}, L_{4, 5} and L₅S₁ segments, the distance between the lower edge of vertebral pedicle and the upper edge of the lower vertebral pedicle and the distance between the point 3 mm to the ventral side of the intervertebral space to the superior articular process. The necessity of performing foraminoplasty was evaluated by measuring the change of intervertebral foramen width using dynamic X-ray and verified during operation. Results: The height of the intervertebral foramen of L_3,4, L_4,5 and L₅S₁ segments were 1.99±0.25, 1.89±0.15 and 1.52±0.26 cm, respectively. The width of the intervertebral foramen was 0.78±0.14, 0.75±0.13 and 0.64±0.13 cm, respectively. The distance between the lower edge of vertebral pedicle and the upper edge of the lower vertebral pedicle were 1.14±0.17, 1.05±0.16, and 0.98±0.19 cm, respectively. The distance between the point 3 mm to the ventral side of the intervertebral space to the superior articular process were 1.11±0.31, 1.17±0.20, and 0.95±0.14 cm, respectively. The width of the intervertebral foramen of the L_{3, 4} and L_4,5 segments was significantly greater at the over-flexion position than at the over-extension position (P<0.05). Intraoperative verification showed that 347 cases (group A) did not need foraminoplasty. However, the other 65 patients (group B) needed foraminoplasty, including 31 at L_{4, 5} segment and 34 at L₅S₁ segment. One patient in group A and one in group B underwent revision operation due to residual intervertebral disc. At 2 years of follow-up, recurrence occurred in 4 patients in group A and 2 patients in group B. The ODI score and JOA score in group A and B were 18%±9%, 24.2±1.3 and 16%±7%, 23.9±1.3, respectively. There were not significantly different between patients in group A and B (t=1.70, P=0.090; t=1.71, P=0.088). The VAS score of lumbar pain of patients in group A was better than that of patients in group B (P<0.05). Conclusion Most of PETD of L₃-S₁ segments can reach the therapeutic target without performing foraminoplasty with half-half technique combined with far lateral access technique. Due to the special anatomical position of L₅-S₁ segment, the probability of performing foraminoplasty during operation is much higher. Performing foraminoplasty or not depends on the preoperative measurement of foramina and verification during the operation.