Objectives: To verify the feasibility and safety of the anterior transpedicular screw(ATPS) fixation of the upper thoracic spine (T1-T4) through the radiological anatomy study on the cadaveric specimens. Methods: The upper thoracic spine thin-section CT data of 40 cases were collected from the radiology de-partment′s database(20 males and 20 females, aged from 18 to 68 years, the mean age was 39.7 years). The data of OPW(outer pedicle width), OPH(outer pedicle height), PAL(pedicle axis length), TPA(transverse section angle), SPA(sagittal section angle), DTIP(distance of transverse intersection point) and DSIP(distance of sagittal intersection point) of each pedicle were measured on the transverse and sagittal sections through the axis of each pedicle. The data were recorded and statistically analyzed. 10 upper thoracic spine(C7-T6) specimens of adults(5 males and 5 females, with unknown ages), with no damage to their appearance, the costovertebral joints and paravertebral soft tissue were completely retained. Then simulate surgical operations were done on the cadaveric specimens based on the obtained data. Screws were implanted anteriorly by free hand. After that, the specimens accepted X-ray fluoroscopy and CT scan. At last, the screws were removed, the speci-mens were sawed along the transaction and sagittal section of the screw channel. Then the success rate of the screw placement was evaluated according to Rao′s worn out classification standard of pedicle screws. Results: From T1 to T4, the OPW decreased from 8.14mm to 3.47mm; the OPH increased from 6.89mm to 10.29mm; the TPA decreased from 32.96° to 11.64°; the DTIP increased from 1.80mm to 5.50mm; the SPA increased from 104.95° to 115.74°; the DSIP increased from 5.95 to 8.76mm; the PAL changed irregularly, from 32.95 to 35.96mm. The pedicle diameters of T3 and T4 were too small to implant ATPS, but the ARTPS can be implanted successfully. The diameter of ATPS was about 4.0mm; the length of ATPS was about 35mm. The diameter of ARTPS was about 5.0mm; the length of ARTPS was about 35mm. 80 pedicle screws were implanted anteriorly, according to Rao′s worn out classification standard of pedicle screws, the fine rate was 90%. The internal walls of 7 pedicles were broken by screws of less than 2mm and no compression to the spinal cord. The internal walls of 5 pedicles were broken of 2 to 4mm, 1 at T1, 1 at T3 and 3 at T4, with varying degrees of spinal cord compression. The internal walls of 2 pedicles were broken of greater than 4mm, 1 at T2 and 1 at T4, with serious spinal cord compression. The external wall of 1 pedicle was broken at T2. Conclusions: The ATPS techniques at T1, T2 and the ARTPS techniques at T3, T4 are feasible, but the safety and clinical practice and further research is needed.

Objectives: To explore the association between the polymorphism in -857 site of tumor necrosis factor (TNF)-α promoter region and the susceptibility to ankylosing spondylitis (AS). Methods: Case-control studies Pubmed, Cochrane Library, Ovid, Chinese Biomedical Database(CBM), Chinese National Knowledge In-frastructure(CNKI), Wanfang and Weipu data bases from inception to October 2013 for the association between TNF-α-857 C/T polymorphism and the susceptibility to AS were collected. Meta-analysis was performed by Revman 5.2 and Stata 12.0 software. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were de-rived from random-effects or fixed-effects models to assess the strength of the association. Results: Nine case-control studies were included in the final meta-analysis, including a total of 933 AS patients and 1094 controls. Statistically significant differences between AS and control groups were observed in the susceptibility to AS and TNF-α-857 genotype CC [OR=0.46, 95%CI (0.26, 0.81), P=0.007], allele C [OR=0.61, 95%CI (0.41, 0.91), P=0.02] and T [OR=1.64, 95%CI (1.10, 2.43), P=0.02]. But, no statistical difference in the fre-quency of genotype TT [OR=1.49, 95%CI (0.95, 2.34), P=0.08] was observed between AS and control groups. There were obvious heterogeneities among the studies of genotype CC(P<0.00001, I2=87%), allele C(P<0.00001, I2=84%) and allele T(P<0.00001, I2 =84%), except genotype TT(P=0.09, I2=42%). Sensitivity analysis was per-formed by leaving out one study at a time, but the heterogeneity remained obvious. It was symmetric of the funnel plots of genotype CC, allele C and T with AS, but genotype TT. There was no statistical significance in genotype CC[Begg′s test(z=0.52, P=0.602), Egger′s test(t=0.23, P=0.825)], genotype TT[Begg′s test(z=0.94, P=0.348), Egger′s test(t=1.26, P=0.248)], allele C[Begg′s test(z=0.31, P=0.754), Egger′s test(t=0.72, P=0.494)] or allele T[Begg′s test(z=0.31, P=0.754), Egger′s test(t=-0.72, P=0.494)]. Conclusions: Genotype CC, allele C and T of TNF-α-857 are associated with the susceptibility to AS, and T-allele carriers have higher risk of AS.

Objectives: To compare the efficacy of patient-controlled epidural analgesia and patient-controlled intravenous analgesia in postoperative analgesia after spinal fusion. Methods: The CNKI, Wanfang, CBM, Pubmed, Embase, Ovid, Cochrane library databases were searched by computer. The randomized controlled trails with patient-controlled epidural analgesia or patient-controlled intravenous analgesia after spinal fusion published from May 1985 to July 2013 were selected. All of them were assessed by the standard of Cochrane systematic review. Data of postoperative VAS and the rate of side effects after using the analgesic drugs were extracted into an electrical sheet, which was synthesized by a Meta-analysis with RevMan 5.2 software. Re-sult: Eight randomized controlled trails involving 482 patients met the inclusion criteria. And the quality grade of 3 literatures was A, 5 literatures was B. The result of meta-analysis showed that: ①Visual analogue score(VAS). The postoperative first day′s VAS[WMD=-0.47, 95%CI(-0.74, -0.20)], second day′s VAS[WMD=-0.66, 95%CI(-1.14, -0.19)] showed that the patient-controlled epidural analgesia had a good analgesic effect compared with the patient-controlled intravenous analgesia, and both had statistic differences(P<0.05). No dif-ference in the postoperative third day′s VAS[WMD=-0.58, 95%CI(-1.38, 0.21)] was observed between the two methods, and no statistic difference(P>0.05); ②The side effects. The patient-controlled epidural analgesia had a higher incidence of skin itch[RR=1.53, 95%CI(1.08, 2.16)], paresthesia[RR=3.34, 95%CI(1.12, 9.98)] after surgery than the patient-controlled intravenous analgesia, and both had statistic differences(P<0.05). While two groups had no a significant deviation about nausea [RR=1.05, 95%CI (0.79, 1.40)], vomiting [RR=0.80, 95%CI (0.48, 1.31)], and no statistic difference(P>0.05). Conclusions: The patient-controlled epidural analgesia has better analgesic effects on the postoperative first and second day after spinal fusion, but obviously has a high-er incidence of skin itch and paresthesia than the patient-controlled intravenous analgesia.

Objectives: To reveal the possible role and mechanism of disc-facet angle in the occurrence and progress of Hirayama disease by analyzing the correlation of clinical appearance and the variation of disc-facet angle in Hirayama disease. Methods: 45 patients with Hirayama disease treated in the orthopedic de-partment of Peking University Third Hospital from October 2006 to January 2012 were reviewed. There were 44 males(97.8%) and 1 female(2.2%) with an average age of 19.33±3.89(range 13-37) years old. The onset age was 16.33±2.73 (range 10-27) years old and the duration of history was 35.64±23.24 (range 1 to 120) months. The disc-facet angle of C3-T1 was measured on PACS system. An extension line was made along the upper edge of objective vertebral body on the CT scan sagittal reconstruction image. An extension line was made along the joint surface of upper articular process of objective vertebra on the sagittal plane which crossed the midpoint of the facets. The blunt angle between two lines was the disc-facet angle. The cases were divided into unilateral affect and bilateral affected group according to history, spinal cord function and spinal cord lesion level. The correlation of above data and the variation of disc-facet angle were analyzed. Results: The duration of history in unilateral affected patients was significantly shorter than that of bilateral affected patients (P<0.05). The duration of history in sequential onset patients was significantly shorter than that of simultaneous onset in bilateral affected patients (P<0.05). The JOA score 17 and Chinese score 40 in unilateral affected patients were higher than those of bilateral affected patients. The C3-C7 disc-facet angle of affected side was bigger than that of normal side in unilateral affected patients, and C5, C6 had statistically significant difference (P<0.05). The spinal cord lesion level was mainly at C5 and C6 in unilateral affected group. The disc-facet angle had no difference on both sides in bilateral affected patients but the figures at C4 and C5 level were bigger than that of normal side in unilateral affected patients. The spinal cord lesion level was mainly at C4 and C5 in bilateral affected group. The stability of upper cervical spine segments was even worse in unilateral affected group and resulting in higher level of spinal cord lesion comparing with bi-lateral affected group. Conclusions: The disc-facet angle of C4, C5 is bigger in bilateral affected patients and may be correlated with higher level spinal cord lesion and more severe clinical symptoms. The C5, C6 disc-facet angle of affected side is significantly bigger than that of normal side in unilateral affected patients. This may be the key factor of spinal cord lesion mainly at C5 and C6 level on the same side.

Objectives: To study the morphological difference of luscka joints between Hirayama disease patients and non-Hirayama disease patients on CT scan and to provide a new possible mechanism of Hirayama disease. Methods: 32 patients(all males) with a mean age of 19.4±4.1(range 16-37 years) and with Hirayama disease were treated in our hospital from October 2006 to January 2012, the mean course of disease was 31.7±23.7 months(range, 1-120 months). 32 patients(all males) with a mean of age was 19.1±4.3 (range, 12-26 years) suffering from acute neck pain and having no Hirayama disease were reviewed as control. Both groups showed no age related difference. From the cervical CT coronal plane reconstruction images which passing through the transverse foramen center of C3-C7 in GE-PACS system, the following data were measured in both sides: ①The width of the uncinate process base: the distance between inner and outer margin of the uncinate process at the upper edge of the vertebral body. ②The height of the uncinate process:the vertical distance from the top of the uncinate process to the upper edge of the vertebral body. ③The distance between two uncinate processes: the distance between the tips of the bilateral uncinate processes. ④The inclination angle of the uncinate process: the angle between the uncinate process and the upper edge of the vertebral body. ⑤The inclination angle of the inferior endplate: the angle between the uncinate process:the vertical distance from the top of the uncinate process to the upper edge of the vertebral body. Results:There were no significant side-related differences on the width of the uncinate process base, the height of the uncinate process, the distance of the uncinate process, the inclination angle of the uncinate process and the inclination angle of the inferior endplate at the same segment from C3 to C7 in Hirayama disease patients(P>0.05). However, differences were found on the height of the uncinate process and the distance between two uncinate processes of C3-C7 (P<0.05), C6 and C4 had the highest and lowest height of the uncinate process as C6>C5>C7>C3>C4. The distance of the uncinate process gradually increased from C3 to C7. There were no significant differences on the width of the uncinate process base, the inclination angle of the uncinate pro-cess and the inclination angle of inferior endplate(except for C3) of C3-C7(P>0.05). Then, using the mean value of the left and right sides as the width of the base of uncinate process, the height of uncinate process, and calculating the sum of inclination angle of the uncinate process, the sum of inclination angle of inferior endplate of the upper vertebra and the difference between the sums (the sum of inclination angle of uncinate process - the sum of inclination angle of inferior endplate of the upper vertebra), then calculating the mean value and standard deviation. Compared with the non-Hirayama disease patients, there were no significant dif-ferences on the uncinate process base, the distance of the uncinate process and the sum of inclination angle of the uncinate process at the same segment of Hirayama disease patients (P>0.05), while the height of the uncinate process and the sum of inclination angle of uncinate process of Hirayama disease patients were sig-nificantly smaller than those of the control group, respectively(P<0.05), and the differences between the sums were larger than those of the control group (P<0.05). Conclusions: Hirayama disease patients may possess a dysplasia in the luscka joint, manifesting the nonuniform development of the uncinate indicators. Lower unci-nate process and smaller inclination angle of inferior endplate of the upper vertebra are common, The conse-quential cervical instability may play a significantly important role in the pathogenesis and progress of Hi-rayama disease.

Objective:To investigate the surgical outcome of one-stage combined anterior and posterior ap-proach for severe thoracolumbar and lumbar spine fracture.Method:A total of 62 cases suffered from severe thoracolumbar and lumbar spine fracture undergoing surgery from Jan 2003 to Jan 2008 were reviewed retro-spectively.Of these,there were T11 involved in 2 cases,T12 in 13 cases,L1 in 28 cases,L2 in 10 cases,L3 in 6 cases and L4 in 3 cases.There were 58 fresh fractures and 4 old fractures.Based on Dennis classifica-tion,12 were compression fracture,33 were burst fracture and 17 fracture dislocation.All cases had spine load score≥7 and TLICS score≥5.Of 19 cases with neurological deficit according to Frankel grade,there were 7 A,5 B and 7 C.Combined anterior and posterior approach was performed in all cases,anterior bony graft plus posterior pedicle instrumentation were performed either,of these,52 cases had additional anterior decompres-sion.Result:All operations were performed successfully,with the mean surgical time of 170min (range, 150-210min) ,the average blood loss was 819ml(range,400-2900ml).No iatrogenic neuroinjury,skin infection, dural matter tearing and graft displacement were noted.The preoperative Cobb's angle was 8°-40°(mean, 23.9°), while the postoperative counterpart returning to normal with 5 cases having 2°-10° kyphosis.The preoperative compression rate was 20%-95%(mean,54.5%),while the postoperative counterpart returning to normal in 47 cases,with 15 cases having 2%-30%.The preoperative canal stenosis rate was 5%-90%(mean,51.1%) while the postoperative counterpart was 0-30%(mean,4.7%),which showed significant difference with regarding to these 3 parameters (P<0.05).All cases were followed up for an average of 31 months (range,12-72 months). Bony fusion was evidenced in cases undergoing anterior bony graft.At 10-12 months, the Cobb's angle was 0°-15°(mean,0.62°) ,the vertebral compression rate was 0-30%(mean,4.6%),no significant difference were noted between them and their postoperative counterparts(P>0.05).At final foUow-up,15 of 19 cases with neu-rological deficit had neurofunction improved,while 4 remained unchanged.According to our hospital criteria,of 43 cases with no neurological deficit,there were 30 excellent,9 good,3 fair and 1 bad with the total excel-lent to good rate of 90.6%.Cage subsidence and pedicle screw breaking was noted in 1 case,who developed severe kyphosis presenting with irreducible back pain.Conclusion:One-stage combined anterior and posterior approach for severe thoracolumbar and lumbar spine fracture can ensure three column stability as well as complete decompression,which has good early outcome.

Objective:To evaluate the feasibility and clinical efficacy of the treatment of lumbar spine steno-sis by modified unilateral approach for bilateral decompression under microendoscope.Method:217 cases with degenerative lumbar spine stenosis from September 2003 to December 2008 were treated by modified unilater-al approach for bilateral decompression of central spinal canal and nerve root canal under microendoscope.A-mong them laminectomy of one segment was done in 163 cases,two segments were done in 54 cases.After surgery,the routine radiograph were carried out and the Nakai criterion was used for assessment.Result:The mean operative time was 48±13min (range,25 to 95min),the average blood loss was 37±9ml (range,25 to 180ml),the average skin incision length was 2.2±0.2cm(1.8 to 2.4cm).2 cases had dural matter tearing,which healed after filling with gelatin foam and 2 weeks bed rest,after that no cerebrospinal fluid leakage was noted after surgery.Mislocation was noted in 1 case.No nerve inury and postoperative infection oceurred.Postoperative CT scan demonstrated complete decompression of the central spinal canal and nerve root canal.All patients were followed-up for an average of 14 months (range,3 months to 24 months).The Nakai criterion at final follow-up showed clinical excellent in 134 cases,good in 63 cases,fair in 16 cases and worse in 4 cases, with the excellent and good rate of 90.8%.No lumbar spine instability was noted.Conclusion:Treatment of de-generative lumbar spine stenosis by modified unilateral approach for bilateral decompression under mieroendo-scope has the merit of minimal invasive,less complications and reliable therapeutical effect.

Objective:To investigate the results of surgical treatment for multi-level spinal tuberculosis with focal debridement ,bone grafting and primary internal fixation.Method:27 cases of multi-level spinal tuberculosis treated by operation from January 1999 to January 2007 were reviewed retrospectively.3 vertebral bodies involved in 15 cases,4 vertebral bodies involved in 7 cases,5 vertebral bodies involved at least in 5 cases.Of these,there were 5 cases in thoracic vertebra,7 cases in lumbar vertebra,11 cases in thoracolumbar vertebra,4 cases in lumbosacral vertebra.11 cases showed neurological deficit before surgery,According to the Frankel classification,Frankel B in 1 case,Frankel C in 2,Frankel D in 8.The preoperative average Cobb angle of kyphosis was 35°±7° (range,26°-43°).The focal debridement,bone grafting and primary internal fixation were performed for all patients.Among these protocols, 16 cases underwent one staged posterior spinal internal fixation and anterior focal debridement,bone grafting.11 cases had anterior spinal internal fixation with focal debridement and bone grafting.All patients received anti-TB chemotherapy before and after operation.Anti-TB chemotherapy protocols consisting of combinations of rifampin,isoniazid,ethambutol and streptomycin were administered for 3 months which including 3 weeks before the operation and 3 months after the operation,followed by rifampin,isoniazid and ethambutol for a total of 9 months.The clinical outcomes were evaluated according to the criteria designed by WU Qiqiu et al.Result:The average surgical time was 240 minutes (range,150-300min),with the average intraoperative blood loss of 1000ml (range,600-1400ml),There was no complication such as nerve injury and cerebrospinal fluid leakage.The incision in 1 case presented sinus which had a secondary healing later on,others obtained primary healing.The average postoperative Cobb angle of kyphosis was 15°±5°(range,12°-20°).All cases were followed up for an average of 18 months(range, 14-48 months).The patients with paraplegia recovered to the normal in 8 months.Bony fusion was achieved in all cases with mean fusion period of 7 months(range,5-14 months) with no incidence of hardware failure.At the final follow-up,all patients had evidence of clinical heal and the solid bony fusion.Conclusion:Focal debridement,bone grafting and primary internal fixation is reliable in dealing with multi -level spinal tuberculosis.

Objective:To evaluate the neurologic complications associated with posterior lumbar interbody fusion (PLIF).Method:From March 2003 to May 2008,a total of 178 patients experienced PLIF procedure due to lumbar disease.Of these,forty two had lumbar disc herniation,39 had lumbar spine stenosis,61 had degenerative spondylolisthesis, 22 had isthmic spondylolisthesis and 14 had post-laminectomy/discectomy syndrome. One hundred and thirty patients had single level fused,of these,8 at L3/4.64 at L4/5 and 58 at L5/S1.Forty six cases had two levels fused,of these, 19 at L3/4 and L4/5,27 at L4/5 and L5/Sl.Two patients had three levels fused,all at the L3/4.L4/5 and L5/S1.A11 causes possibly related to the neurologic complications were documented.Result: 13 cases(7.3%) were noted to have neurologic complications,of these,one at L4,7 at L5 and 5 at S1.6 cases had nerve root stretch injury,3 cases had nerve root injury from instrument maneuver, 1 case had nerve root deformity, 1 case had developed epidural hematoma, 1 case had nerve root injury from pedicle fracture and 1 case had secondary lateral recess stenosis.2 cases had complete neurological injury,and 11 cases had incomplete injury.According to American Spinal Injury Association (ASIA) criteria,there was complete recovery of neurologic function in 7 cases, incomplete recovery in 4 cases and no change in 2 cas-es.Conclusion: Many causes may be associated with the neurologic complications due to posterior lumbar interbody fusion,nerve root stretch injury is the most common.

Objective:To determine the most common patterns in symptomatic lumbar intervertibral disc degeneration at various age period.Method:The lumbar MRI imaging data of 529 patients treated for low back pain with or without skelagia in Santa Monica hospital in USA from 2004 to 2006 were reviewed retrospectively in this study .There were 191 females, 338 males.Degenerative grading was made for all disc levels in the lumbar spine for each patient. The patients were also classified into five age different groups starting from below age 30,with the intervals of a single decade,until over age 60 (≤29 years,30-39 years,40-49 years, 50-59 years and≥60 years).The degenerative grade of all disc levels (L1/2 to L5/S1) in each patient was classified based on Pfirrmann's gading system,and locations and the progressive manner of the degeneration were determined for age groups.SPSS software version 13.0 (SPSS Inc.,Chicago,IL) on a personal computer was used for statistical analyses.Result:The most common pattern in all ages is normal in all the discs and it is described as 1-1-1-1-1 and this normal pattern was found in 71 patients (13.4%).There was rare change to develop severe degeneration (grade 4) for low age groups.The upper levels of the lumbar spine had more commonly lower degenerative grades,and the lower levels had relatively higher grades of advanced disc de-generation.In the age groups of below 29 years and 30-39 years,the most common pattern was normal in all the discs of 5 levels as 1-1-1-1-1.And this pattern was more prominent in the age group below 29 years.In the age group of 40-49 years.it was changed to single level L5-S1 degeneration with 1-1-1-1-3(7.6%) butthe normal discs in all levels had also a similar portion(6.8%).In the age groups of 50-59 years and over 60 years,it was changed to the pattern with all discs abnormal as 3-3-3-3-3.Single level involvement was negatively correlated with age (P<0.01 ).On the contrary,whole symmetrical degenerative involvement of all levels was positively correlated with age (P<0.01).Single L5/S1 involvement was most common in the age group of below 29 years,but in the age group of 30-39 years,the most common pattern was changed to L4/5 and L5/ S1 double involvement(P< 0.05).A relative small portion appeared as the most common pattern with less than two-level involvement in the age groups over 50 years.Conclusion:The rate of unilevel degenerative pattern of lumbar spine in overall population decreases with the progression of age, while that of all-level degenerative pattern increases with the progression of age.