Objective To report the clinical results of surgical treatment of fracture of the anteromedial facet of the coronoid process in terrible triad injury.Methods Of the 59 patients with terrible triad injury of the elbow,17 were surgically treated for fracture of the anteromedial facet of the coronoid process from July 2010 to July 2014.They were 12 men and 5 women,from 29 to 70 years of age (average,50 years).By the Mason classification for the radial head fractures,2 cases were type Ⅰ,13 type Ⅱ and 2 type Ⅲ;by the O'Driscoll classification for coronoid process fractures,one was type Ⅰ,14 were type Ⅱ and 2 type Ⅲ.All patients were treated through combined approaches.The lateral Kocher approach was used to fixate or replace the radial head and to repair the lateral ligament complex;the anteromedial approach was used to fixate the coronoid process fracture and to explore and repair the medial collateral ligament if necessary.Fracture union,implant loosening,ectopic ossification,regression and articular alignment were assessed on the postoperative X-rays.At final follow-ups,the elbow function was evaluated using Mayo elbow performance score (MEPS) and Broberg & Morrey grading system for traumatic arthritis.Results The average follow-up duration for the 17 patients was 32 months (range,from 24 to 60 months).Final follow-ups showed that the mean arc of flexion-extension was 97° (from 70° to 120°),the mean extension limitation 23° (from 0 to 40°),the mean arc of forearm rotation 139° (from 90° to 145°),the mean pronation 71° (from 60° to 90°)and the mean supination 67° (from 60° to 85°).The MEPT scores averaged 87 points (from 80 to 100 points),yielding 7 excellent,8 good and 2 fair cases.By the Broberg & Morrey grading,there were 4 cases of grade 1 and one of grade 2.No evidence of elbow instability,nonunion or implant failure was found.Ectopic ossification around the elbow happened in 4 cases,transient ulnar nerve palsy was found in one,and the Kirschner wire was removed operatively in 2 because it had loosened after fracture union.Conclusion The treatment of anteromedial coronoid fractures in terrible triad injury should be anatomically reduced through direct exposure and fixated rigidly.

Objective To explore the classification system and outcomes of surgical treatment for terrible triad of the elbow.Methods Data of 42 patients with terrible triad elbow injuries who were surgically treated between 2009 and 2015 were retrospectively analyzed.There were 29 males and 13 females with a mean age of 47 years at the time of injury.42 patients of terrible triad injuries were classified into four types:type ⅠA,ⅠB,Ⅱ,Ⅲ,and Ⅳ injuries.Type ⅠA and Ⅱ injuries were treated through an isolated lateral approach,while type ⅠB,Ⅲ and Ⅳ injuries were treated by a combined lateral and anteromedial approach.Operative treatment consisted of repair or replacement of the radial head,repair of the lateral collateral ligament (LCL) and coronoid fracture fixation.Type ⅠA injuries were treated with radial head and LCL repair without coronoid fixation.Type Ⅳ elbow injuries were treated with medial collateral ligament (MCL) repair.Elbow functional status was evaluated using the Mayo elbow performance score (MEPS).Results There were three patients with type ⅠA injuries,7 patients type ⅠB injuries,15 type Ⅱ injuries,10 type Ⅲ injuries,and 7 type Ⅳ injuries.The average follow-up period was 30 months (range,24-56 months).All fractures of coronoid got union at average 11.5 months except for type ⅠA injuries.40 patients with fractures of radial head got union at average 12.4 months and two patients underwent radial head replacement without loosening.The mean flexion-extension arc was 107°±22°,the mean flexion contracture was 20°±10° and the mean flexion was 127°±14°.The average forearm rotation arc was 145°±14°,which included an average pronation of 73°±8° and an average supination of 71°±9°.The mean MEPS was 89±9 points (range,55-100 points),with excellent results in 24 elbows,good result in 17 and poor result in one;the excellence rate was 97％ (41/42).Thirteen patients had radiographic signs of arthrosis according to the Broberg-Morrey system (9 elbows were grade 1 and 4 were grade 2).5 patients had evidence of heterotopic ossification,of which four had minimal periarticular ossification and did not require additional surgery.The remaining patient showed significant heterotopic ossification and required an elbow release.1 patient with type Ⅲ injury developed transient median nerve paralysis and got full recovery after conservative treatment for 8 weeks.1 patient with type Ⅲ injury developed ulnar neuropathy and required an anterior ulnar nerve transposition.2 patients,who had shifting hardware but still achieved union,required a second surgery to remove the implant:one patient had a Kirschner wire shift from the radial head at 6 months after surgery,and the other had a loose screw in the coronoid process at one year after surgery.Conclusion Our classification system of terrible triad of the elbow may provide a guide for the selection of an ideal surgical approach and treatment modality.

Objective:To determine the biomechanical stability of atlantoaxial and occipitocervical fixations with unilateral spinous process screw plus contralateral pedicle screw.Methods:After nonlinear finite element models of the intact upper cervical spine (C0-3) were constructed and validated, they were processed into 3 kinds with normal axis anatomy, a high-riding vertebral artery and a thin axis lamina, re-spectively.In the models, the odontoid fracture stabilized by atlantoaxial fixation and the atlas fracture sta-bilized by occipitocervical fixation were simulated.In the atlantoaxial fixation, the group of unilateral spinous process screw+contralateral pedicle screw+bilateral atlas lateral mass screws (spinous process screw group) was compared with that of bilateral axis pedicle screws+bilateral atlas lateral mass screws (pedicle screw group); in the occipitocervical fixation, the group of unilateral spinous process screw+contralateral pedicle screw+occipital screws (spinous process screw group) was compared with that of bilateral axis pedicle screws+occipital screws (pedicle screw group).Three different techniques in insertion of spinous process screws (horizontal, oblique and vertical) were applied.The range of motion (ROM) was measured in extension-flexion, lateral flexion and rotation of occipitocervical bodies (C0-C3) after internal fixation.Results:Compared with the intact models, the ROMs of extension-flexion, lateral flexion and rotation of C1-C2 were obviously reduced in the spinous process screw group and the pedicle screw group in the models with atlantoaxial and occipitocervical fixations.In the atlantoaxial fixation, the ROMs of extension-flexion, lateral flexion and rotation of C1-C2 were greater in the spinous process screw group than in the pedicle screw group; in the occipitocervical fixation, the spinous process screw group had greater ROM of lateral flexion of C1-C2 and greater ROM of rotation of C0-C2 than the pedicle screw group.There were unobvious differences in insertion of spinous process screws between the horizontal, oblique and vertical techniques.Conclusions:In the atlantoaxial and occipitocervical fixations, the unilateral spinous process screw plus contralateral pedicle screw can provide as fine stability as the bilateral axis pedicle screws.In the atlantoaxial fixation, the bilateral axis pedicle screws may lead to better stability of C1-C2 than the unilateral spinous process screw plus contralateral pedicle screw.In the occipitocervical fixation, the bilateral axis pedicle screws may lead to better stability of C1-C2 in lateral flexion and rotation than the unilateral spinous process screw plus contralateral pedicle screw. The 3 different techniques in insertion of spinous process screws lead to an unobvious difference in the stability of C1-C2.

Objective:To compare the effects of 3 spatial locations of the screw at the injured vertebra on the vertebral height in AO type A thoracolumbar fracture.Methods:A retrospective analysis was performed of the 156 patients with type A thoracolumbar fracture who had been hospitalized at Department of Spine Surgery, The Sixth Hospital of Ningbo from January 2016 to June 2019. They were divided into 3 groups according to the spatial location of the screw at the injured vertebra. In group A of 55 cases, the screws were located in the vertebral body between the longitudinal axis bisector of the vertebral pedicle and the upper endplate; in group B of 52 cases, the screws were located in the vertebral body between the vertical axis bisector of the vertebral pedicle and the horizontal line of the apex of the inferior pedicle notch; in group C of 49 cases, the screws were located in the vertebral body between the horizontal line at the apex of the inferior pedicle notch and the inferior endplate. The anterior, middle and posterior heights of the injured vertebra, Beck index and angulation of the injured vertebra at preoperation, one week postoperation and the last follow-up were compared between the 3 groups and within the same group.Results:There was no significant difference in preoperative general data between the 3 groups, showing comparability ( P>0.05). In all the 3 groups, the anterior and middle heights of the injured vertebra and Beck indexes at one week postoperation and at the last follow-up were significantly larger than those before operation while the angulations of the injured vertebra at one week postoperation and at the last follow-up were significantly smaller than the preoperative values (all P<0.05), but there was no significant difference between one week postoperation and the last follow-up in any of the above indexes ( P>0.05). In all the patients, the posterior height of the injured vertebra at one week postoperation was significantly larger than those before operation and at the last follow-up ( P<0.05), but there was no such a significant difference in comparison between preoperation and the last follow-up ( P>0.05). At the last follow-up, groups A and B had significantly larger anterior and middle heights of the injured vertebra and Beck indexes but significantly smaller angulations of the injured vertebra than group C, but such significant differences did not exist when the above indexes were compared between groups A and B ( P>0.05). Conclusions:In insertion into an injured vertebra, the screw should be parallel and close to the upper endplate, and located in the middle and upper part of the vertebra corresponding to the longitudinal axis of the vertebral pedicle, because this spatial position is conducive to intraoperative reduction, maintaining the postoperative height of the injured vertebra, and decreasing loss of the vertebral height.

Objective:To compare the therapeutic results between axis pedicle screwing assisted by intraoperative 3-D navigation and freehand axis pedicle screwing in the treatment of Hangman fracture.Methods:A retrospective analysis was performed of the 64 patients with Hangman fracture who had received posterior axis pedicle screwing at Department of Spinal Surgery, The Sixth Hospital of Ningbo from May 2014 to December 2019. According to the placement methods of axis pedicle screws, they were divided into a navigation group ( n=34, subjected to axis pedicle screwing assisted by intraoperative 3-D navigation) and a freehand group ( n=30, subjected to freehand axis pedicle screwing). Pedicle screw placement time, operation time, intraoperative bleeding, fluoroscopy time, hospital stay, total hospitalization cost and complications were recorded and compared between the 2 groups. The accuracy of axis pedicle screw placement was evaluated according to the postoperative cervical CT and screw grading criteria proposed by Park et al. At admission, 3 months postoperation, and the last follow-up, neurological function of the patients was evaluated by modified Japanese Orthopedic Association (mJOA) score, neck pain was evaluated by visual analogue scale (VAS), and C2/3 vertebral body angulation and C2 forward displacement were measured. The clinical efficacy was evaluated by Moon grading at the last follow-up. Results:The navigation group and the freehand group were comparable due to insignificant differences between them in the preoperative general data ( P>0.05). The accuracy of screw placement in the navigation group (98.2%, 54/55) was significantly higher than that in the freehand group (85.2%, 46/54) ( P<0.05). The screw placement time, operation time, fluoroscopy time and total hospitalization cost in the navigation group were significantly more than those in the freehand group ( P<0.05). Vertebral artery injury occurred in 3 cases in the freehand group. Screw loosening, screw breakage or rod breakage occurred in none of the patients after operation. There was no significant difference between the 2 groups in the intraoperative bleeding, hospital stay or follow-up time ( P>0.05). In both groups, the VAS score, mJOA score, C2/3 vertebral body angulation and C2 forward displacement were significantly improved at 3 months postoperation and the last follow-up compared with those at admission ( P<0.05), but there was no significant difference between the 2 groups in the contemporary comparisons ( P>0.05). At the last follow-up, Moon grading in the navigation group was significantly better than that in the freehand group ( P<0.05). Conclusion:In the treatment of Hangman fracture, compared with freehand screw placement, axis pedicle screwing assisted by intraoperative 3-D navigation can improve accuracy and safety of screw placement and reduce postoperative complications, leading to better clinical efficacy.

Objective To analyze the causes of revision surgery after percutaneous transforaminal endoscopic discectomy (PTED) for lumbar spinal stenosis,and to provide references for indications and operative methods.Methods From January 2015 to October 2017,206,491 and 60 patients of lumbar spinal stenosis were treated with PTED in Tianjin Hospital,Shanxi People's Hospital,Ningbo Sixth Hospital,respectively;among them,4,10 and 4 cases received revision surgery.Another 13 patients of lumbar spinal stenosis were treated with revision surgery due to poor results after PTED in other hospitals.Among 31 cases of reoperation,there were 16 males and 15 females,aged 27-82 years (average,66.2±12.7 years).The lesion segments included 1 case of L3,4,23 cases of L4,5,5 cases of L5S1,1 cases of L3-L5,and 1 cases of L4-S1.Patients were followed up after reoperation from 3 to 24 months (average,12.1 months).The causes of poor result and revision surgery were analyzed according to preoperative,intraoperative and postoperative data.Results All of 757 cases of lumbar spinal stenosis were treated with PTED in three hospitals,of which 18 cases (2.4％) were re-operated.The causes of reoperation included:bone slice displacement in 1 case;nerve injury in 4 cases;lumbar instability in 4 cases;disc protrusion in 10 cases (residual or recurrence);insufficient decompression in 21 cases;planed staging operation in 4 cases with bilateral or two-level stenosis.32 revision surgeries were performed for 31 patients,including PTED in 15 cases,microendoscopic discectomy (MED) in 1 case,mobile MED (MMED) in 5 cases,MMED assisted fusion in 2 cases,transforaminal lumbar interbody fusion (TLIF) in 4 cases,Minimally invasive TLIF (Mis-TLIF) in 2 cases,and open decompression and fusion in 3 cases.All patients experienced relieve of symptoms after revision surgery.At final follow-up,VAS leg pain deceased form 7.1±3.9 before revision surgeries to 1.9±1.2,VAS low back pain decreased form 6.3±3.2 to 1.8±1.3,ODI score decreased from 35％± 14％ to 7.6％±5％.According to the MacNab score,the result was excellent in 11 cases,good in 16 cases,and fair in 4 cases.Conclusion The treatment of lumbar stenosis with PTED has high technical requirements,the indications of PTED for lumbar stenosis should be strictly controlled according to technical conditions,and appropriate operative methods should be chosen according to the specific conditions of the lesions.Insufficient decompression,disc protrusion,lumbar instability and nerve injury are the common causes of reoperation.Suitable indications and proper operation should be selected.