Traumatic spondylolisthesis of the axis has been

Transcription

1 Bulletin Hospital for Joint Diseases Volume 60, Number Traumatic Spondylolisthesis of the Axis 42 Cases Myung-Sang Moon MD Jeong-Lim Moon MD Young-Wan Moon MD Doo-Hoon Sun MD PhD and Won-Tai Choi MD Abstract Fourty-two patients (34 males and 8 female) with traumatic spondylolisthesis of the axis were studied in a retrospective review. There were 20 stable and 22 unstable fractures. The 22 unstable fractures were treated surgically: 16 anterior interbody fusion (10 non-plated and 6 plated), 4 pedicle screw fixation for osteosynthesis of the fractured pedicles, and 2 posterior wire fixation for flexion and axial load injury. For all non-surgical cases, head halter tractions for 1 to 8 weeks was prescribed and a cervical orthosis was worn for an additional 6 to 18 weeks. The surgical cases underwent 5 to 7 days of preoperative and 1 to 4 weeks of postoperative head halter traction. In all cases pedicle fractures united after 13 weeks on average in group treated conservatively, 12 weeks (11 to 13 weeks) in the posterior wiring group, 8 weeks (7 to 9 weeks) in the group in which pedicle screws were used, and 11 weeks (9 to 15 weeks) in the anterior fusion group (13 weeks in non-plated, and 8 weeks in plated). There were no differences in patterns of anterior fusion between those in the non-plated and plated groups. There were no non-unions of fractured pedicles and there was no late instability of the C2-C3 or neurological complications. In 2 cases in the posterior surgery group, there was mild nuchal discomfort and some rigidity for a short while postoperatively. Final outcomes were good in all cases. Myung-Sang Moon, M.D., Doo-Hoon Sun, M.D., Ph.D., and Won-Tai Choi, M.D., are from the Spine Center, Sun General Hospital, Taejon, Korea. Jeong-Lim Moon, M.D., is in the Department of Rehabilitation Medicine, Catholic University of Korea, Seoul, Korea.Young-Wan Moon, M.D., is in the Department of Orthopedic Surgery, Hallym University, Kyungki-Do, Korea. Reprint requests: Myung-Sang Moon, M.D., Ph.D., Director, Moon-Kim s Institute of Orthopedic Research, Shi-Bum Apt , Yoe-Ee-Do-Dong, Young-Dung-Po-Ku, Seoul, Korea Traumatic spondylolisthesis of the axis has been known classically as hangman s fracture because it could be produced by judicial hanging. 1-5 However, a renewed interest in this lesion has arisen from a similar lesion seen as a result of traumatic accidents, and reports of both pedicle fractures of the axis secondary to automobile accidents appeared in the literature in the 1950s and 1960s. 6,7 Schneider and colleagues were the first to report this fracture in 8 cases resulting from automobile accidents. 7 This fracture can be treated non-operatively or surgically according to its fracture stability For the unstable fracture, anterior fusion of C2-C3 is the choice of treatment regardless of the flexion and extension types. 9,10,12,14-18 There were only two previous reports on the non-judicial traumatic spondylolisthesis of the axis in Korea. 14,19 This paper reports a series of patients suffering non-judicial traumatic spondylolisthesis of axis with special attention to the indications and surgical techniques. Material and Methods Fourty-two cases of traumatic spondylolisthesis, 34 males and 8 females who were treated from 1975 through 1998 are included in this study. There were 4 flexion and 38 extension injuries. The age of the patients ranged from 26 years to 57 years. The causes of injuries were pedestrian automobile accident in 22 cases, motorcycle accident in 10 cases, fall on head in ditches in 6 cases, and diving accidents in 4 cases (Table 1). The treatment was dependent on the degree of the initial fracture stability. Stable fractures were subjected to conservative treatment while the unstable fractures underwent surgical stabilization; if there was displacement of C2 on C3 with an unusual widening or rotation of the body and neural arch and if complete disruption of the annular ligament was associated with the pedicle fracture, the lesion was judged as unstable. There were 20 stable frac-

2 62 Bulletin Hospital for Joint Diseases Volume 60, Number Table 1 Causes of Injury and Fracture Types No. Pedestrian Motorcycle Type Patients Auto Accident Accident Falling Diving I II III Total % 23.8% 14.3% 9.5% Table 2 Types and Severity of Fracture (n = 42) Fracture type Extension and axial loading 38 Flexion and axial loading 4 Stable Pedicle fractures with (n = 20) 1. Intact annular and posterior ligament complex 5 No. 2. Intact posterior ligament: 15 a. anterior ligament disruption b. avulsion fracture of anterior margin of C2 or C3 body Unstable 3. Complete disruption of entire annular ligament 8 (n = 22) with intact posterior ligament 5 [ with torn posterior ligament (vertical displacement with widened disk space) 3 4. C2 forward displacement with complete annular ligament disruption 14 with intact posterior ligament 12 [ with torn posterior ligament 2 tures and the 22 unstable fractures. When the fractures were classified by their severity, there were 19 patients with type I injury, 14 with type II, and 9 with type III (Table 2). Before 1980, 20 stable fractures were non-operatively treated initially by bed rest for 4 weeks (a halo with vest was not available at the time). This was followed by external bracing until fusion occurred (8 to 16 weeks). Twenty-two patients with unstable fractures underwent surgical correction; anterior interbody fusion was used in 16 cases (10 non-plated and 6 anteriorly plated), pedicle screw fixation was used in 4 cases, and posterior wiring of C1-C3 was used in 2 cases of flexion injury (Table 3). Ten cases of anterior fusion without plating and 2 case of posterior wire fixation were performed prior to Four weeks of bed rest was routinely prescribed before anterior interbody fusion without plating of C2-3 in order to obtain a certain degree of segmental stability through healing of the torn soft tissue. Then postoperatively an additional 4 weeks of bed rest under head halter traction was prescribed, and thereafter external support was maintained for 8 to 14 weeks until fusion had taken place. After 1980 all patients who were subjected to surgery were put under careful surveillance preoperatively for 5 to 7 days for systemic and local physical observation. In those cases with anterior interbody fusion and additional plating of C2-3, the duration of the postoperative head halter traction in bed was minimized to 7 days or less and then the halo vest was applied for 6 to 8 weeks. In four cases with unilateral facet joint luxation, osteosynthesis aided by a pedicle screw was undertaken. Postoperatively the patients were prescribed bed rest in head halter traction for 2 weeks then cervical bracing for 6 to 10 weeks. Late surgery due to the residual segmental instability was not performed in any of the non-operatively treated cases in this series. Follow-up Study Union of the fractured pedicle and segmental stability of C2-C3 were confirmed radiologically at postoperative weeks 2, 4, 6, 12, 24 and thereafter once every year; all patients were followed up postoperative for 1 to 7 years. Outcomes were graded on the basis of patients symptoms and final degree of neck motion (Table 4).

3 Bulletin Hospital for Joint Diseases Volume 60, Number Table 3 Types of Treatment Bed Rest with Cervical No. Traction (weeks) Brace Union Site of Type of Treatment cases Preop Postop (weeks) Time Union Nonoperative treatment (11-17) Surgical treatment: 22 Posterior wiring only 2 < (11-13) Pedicle screwing only 4 < (8-12) Anterior interbody fusion: (9-15) Anterior fusion a. without anterior plating (10-15) Anterior fusion b. with anterior plating 6 <1 Up to (7-10) Anterior fusion Table 4 Final Neck Motion and Outcome Assessment Criteria Pain No Score Outcome (nuchal) Neck motion cases 0 Excellent None Full range of motion 40 (flexion-extension, rotation) 1 Good Mild One side rotation less than 20 2 (Occassional) 2 Fair Moderate Both side rotation less than 20 0 (Intermittent) 3 Poor Severe Restriction of all motion 0 (Persistent) Table 5 Union Time According to Type of Fracture and Treatment Type of fracture I II III Union Time Site of Type of Treatment (n = 19) (n = 14) (n = 9) (weeks) Union Nonoperative treatment (n = 20) (11-17) Pedicle fusion Surgical treatment (n = 22) Posterior wiring only (n = 2) (11-13) Pedicle fusion Pedicle screwing only (n = 4) (7-9) Pedicle fusion Anterior interbody fusion: (n = 16) (9-15) Anterior fusion a. without anterior plating (n = 10) (10-15) Anterior fusion b. with anterior plating (n = 6) (7-10) Anterior fusion Results In all cases pedicle fracture union was obtained at an average of 13 weeks (range: 11 to 17 weeks). In the nonoperatively treated cases union was obtained at an average of 12 weeks (range: 11 to 13 weeks). In the two cases in which posterior wiring was used fracture union was obtained at 8 weeks (range: 7 to 9 weeks). In the 4 cases in which pedicle screws were used union occurred at 11 weeks on average (range: 7 to 15 weeks). Union occurred at 13 weeks (range: 10 to 15 weeks) in the non-plated cases and 8 weeks on average (range: 7 to 10 weeks) in the plated cases (Table 5). There were no significant differences in fusion pattern between the anteriorly nonplated and plated patients. In none of the non-operatively treated patients did spontaneous fusion of C1-C2 develop. There were no residual symptoms related with the fractures and in the surgical cases there were no instances of painful residual instability. Most of the patients had excellent (30 cases) to good (11 cases) results clinically, although there was mild nuchal rigidity and axial pain for a few weeks postoperatively, which was attributed to the nuchal muscle injury in the cases of flexion type injuries and in the cases of posterior surgery (Table 6). However, there were no limitation of flexion and extension of the cervical spine in the non-operatively and anteriorly fused cases, and no adjacent joint problems at the final follow-up, which ranged from 1 to 7 years. In the two cases of posterior wire fixation there was slight limitation of forward flexion and rotation, but the final outcomes were fairly good in these patients (Table 6). Three illustrative cases are shown in Figures 1 through 3.

4 64 Bulletin Hospital for Joint Diseases Volume 60, Number Table 6 Outcome Type of Treament Excellent Good Fair Poor Remarks Nonoperative treatment (n = 20) 18 2 No spontaneous C2-C3 fusion Surgical treatment (n = 22) Posterior wiring only (n = 2) 1 1 Pedicle screwing only (n = 4) 1 3 Anterior interbody fusion: (n = 16) 11 5 a. without anterior plating (n = 10) 7 3 b. with anterior plating (n = 6) 4 2 Total Discussion A fracture by judicial hanging was never experienced by any of us in Korea. 12,14,19 It is a known fact that the mechanism of the classical hangman s fracture differs from the fracture induced by other trauma; hyperextension and distraction in the former, 1,2,13,20 and hyperextension and axial loading or occasionally flexion and axial loading in the latter. 3,6,7,10,21 With a flexion moment there are usually associated compression fractures to support this mechanism, generally involving C3 with anterior subluxation of C2. However, physicians, even in modern times, still call this fracture the hangman s fracture even though it is not produced by the judicial hanging. Therefore, to eliminate possible misconceptions we believe that it is important for surgeons to be aware of the true nature and etiology of this fracture and that it is important to provide a more accurate method of assessment of fracture stability and establish a proper treatment protocol. The stability of the fractured segment is influenced by the severity of the disk disruption of C2-C3, which is well illustrated by Francis and Fielding. 9,10 They classified the severity of injury into 5 grades: A, B, C, D and E. In this study the fracture stability was defined based on the degree of annular and posterior ligament disruption around the vertebral bodies of C2-C3 and the for- ward and/or rotatory vertebral body slip of C2. When the main anterior annular ligament alone in the presence of fractures of both pedicles was disrupted, it was defined as stable. When there was complete annular and/ or posterior ligament disruption with or without C2 body displacement, and with or without compression fracture of anterior upper margin of C3, the fractured segment of C2-C3 was defined as unstable ,22 Before 1980, in 10 cases of the non-plated anterior interbody fusion, 4 weeks of preoperative head halter traction was applied. Prolonged preoperative cervical traction in bed was applied first in order to obtain the local stability through soft tissue healing around the injured level, because anterior surgery destroyed the anterior stabilizers, and at that time the anterior cervical plate and halo skeletal fixator were not available in Korea. 12,14,22 Therefore for the cases of the non-plated anterior fusion, the preoperative and postoperative 4 weeks of head halter traction and the subsequent bracing were thought to be the key in bringing about the successful interbody fusion, although this type of treatment in the era of anterior plating is not welcome. 16,19 After 1980, anterior plate fixation was added to the anteriorly fused segment in 6 cases to provide the stability of the fused segment, to avoid postoperative pro- Figure 1 Case 1 (Conservative). A, A 28-yearold male sustained type I fracture by falling, and was treated nonoperatively; head halter traction for 4 weeks, and cervical brace for 8 weeks. Radiographs of 10 (B) and 12 weeks (C) are shown. The fracture united in 10 weeks.

5 Bulletin Hospital for Joint Diseases Volume 60, Number Figure 2 Case 2 (AIF with plate). A, Lateral view of a type II traumatic spondylolisthesis of axis in a 37-year-old man. B, Radiograph taken one year after anterior intracorporal fusion and plate fixation shows the united pedicle fractures together with solid anterior fusion of C2-C3. Figure 3 Case 3 (Pedicle screw). A, A type II traumatic spondylolisthesis of axis in a 32-yearold man by car accident. Bilateral pedicle screw fixation after reduction of the unilaterally luxated facet joint was performed three days after injury. Postoperatively head halter traction for 4 weeks and then a cervical brace was worn for an additional 6 weeks. Fracture healed in 10 weeks. B, Follow-up radiographs at one year shows solid union of the fractured pedicles. There was no residual segmental instability. longed bed rest, and to mobilize the patients earlier. The benefits of the plate in regard to graft collapse and kyphosis appear to lie in the increased union rate that protects against deformity. The additional benefit of the anterior fusion was that it saved rotatory motion, and left minimum sequela in spite of the sacrifice of the C2- C3 segment. In this study there were no complications associated with the use of the cervical plate. After 1987, when it was introduced in Korea, pedicle screw fixation was performed in four cases with unilateral facet luxation with rotatory displacement. However, after 1991 pedicle screw fixation alone was not performed when there was complete disruption of the entire annular and/or posterior ligaments (as confirmed by magnetic resonance imaging) and the stabilization of the involved segment through spontaneous intracorporal fusion could not be expected even after pedicle screw fixation, and might lead to late instability and surgery-related nuchal discomfort and/ or axial symptoms. Fortunately none of the four pedicle screwed cases in our series developed late segmental instability or spontaneous fusion of C2-C3. Screw fixation of the fractured pedicles could provide accurate reduction and early union of the fractured pedicles. However, it is not indicated for unstable flexion-type injuries associated with a compression fracture of the anterior upper margin of C3. Also, it is thought that the procedure has narrow indications and no advantages over other procedures. Posterior C1 sublaminar to C3 spinous process wire fixation in 2 cases of concomitant supra-spinous and interspinous ligament rupture was also found helpful in regaining the stability of the fractured segment and in healing of the torn posterior ligament. It is thought that posterior wire stabilization is indicated in the presence of the associated C2-C3 supra-spinous and inter-spinous ligament rupture. In this type of injury, pedicle screwing alone is not thought to provide flexion stability of the unstable C2-C3 segment, while the combined anterior fusion and plating of C2-C3 can avoid possible late instability of C2-C3.

6 66 Bulletin Hospital for Joint Diseases Volume 60, Number Posterior wire stabilization and fusion for the traumatic spondylolisthesis of axis would necessitate fusion from C1 to C3 involving some loss of rotatory motion. In our series, however, we did not do posterior fusion after posterior C1 to C3 wire stabilization in order to save the motion of the C1-C3 segment. Posterior wire fixation without fusion in cases of flexion injuries without a C3 compression fracture was found simple and effective in producing union of the pedicle fractures and regaining segmental stability. Conclusion In summary, union of the axis pedicle fractures in this series occurred within 3 months after injury with minimum residual symptoms, regardless of the types of treatment. Nonoperative treatment of the stable fracture can provide successful results. The unstable fractures could be successfully treated by a number of surgical procedures. Pedicle screw fixation brought the earliest fusion of the fractured pedicles, although whether it provides anterior stability can be questioned. Anterior interbody fusion with additional plating was found most effective in stabilization of the unstable fractures with early mobilization and with a minimum of residual symptoms. Therefore, it is thought that anterior interbody fusion and plating is the choice of treatment for the unstable traumatic spondylolisthesis of the axis. References 1. Marshall JJ: Judicial hanging. Brit Med J 2: , Marshall JJ: Judicial hanging. Lancet 2:639, Paterson AM: Fracture of cervical vertebrae. J Anat Lond 24:ix, Williams TG: Hangman s fracture. J Bone Joint Surg 57B:82-88, Wood JF: The ideal lesion produced by judicial hanging. Lancet 1:53, Grogono BJS: Injuries of the atlas and axis. J Bone Joint Surg 36B: , Schneider RC, Livinston KE, Cave AJE, Hamilton G: Hangman s fracture of the cervical spine. J Neurosug 22: , Borne GM, Bedon GL, Penaudeau M: Treatment of pedicular fracture of the axis: A clinical study and screw fixation technique. J Neurosurg 60:88-93, Fielding JW, Francis WR, Hawkins RJ, Pepin J, Hensinger R: Traumatic spondylolisthesis of the axis. Clin Orthop 239:47-52, Francis WR, Fielding JW: Traumatic spondylolisthesis of the axis. Orthop Clin North Am 9(4): , Levine A, Edward C: The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg 67A: , Moon MS, Kim I, Kim OB: Anterior interbody fusion for the traumatized cervical spine [in Korean]. J Korea Orthop Assoc 16(3): , Moon MS, Ok IY, Song SW: Anterior interbody fusion of the cervical spine: Study of 56 cases [in Korean]. J Korean Orthop Assoc 20(5): , Moon MS, Ok IY, Song SW, Kim HK: Treatment of traumatic spondylolisthesis [in Korean]. J Korea Orthop Assoc 21(6): , Roy-Camille R, Saillant G, Mazel C: Internal fixation of unstable cervical spine by a posterior osteosynthesis with plates and screw. In: Sherk HH (ed): The Cervical Spine (2nd ed). Philadelphia: J.B. Lippincott Co., 1989, pp Tuite GF, Papadopoulos MD, Sonntag VKH: Caspar plate fixation for the treatment of complex hangman s fracture. Neurosurg 30: , Segal L, Grimm J, Stanffer E: Non-union of fractures of the atlas. J Bone Joint Surg 69A: , Sherk HH, Howard T: Clinical and pathologic correlations in traumatic spondylolisthesis of the axis. Clin Orthop 174: , Chung JY, Song JY, Choi BH: The results of the operative treatment for the traumatic spondylolisthesis of axis: Anterior plate fixation and transpedicular screw fixation [in Korean]. J Korea Orthop Assoc 29(3): , Garber JN: Abnormalities of the atlas and axis vertebra: Congenital and traumatic. J Bone Joint Surg 46A:1782, Levine A, Edward C: Fracture of the atlas. J Bone Joint Surg 73A: , Moon MS, Kim I, Woo YK, Lee JJ: Anterior interbody fusion in fractures and fracture-dislocations of the spine. Internat Orthop (SICOT) 5: , 1981.