Abstract
Study Design:
Retrospective.
Objectives:
To present rarely reported complex fractures of the upper cervical spine (C1-C2) and discuss the clinical results of the posterior temporary C1-2 pedicle screws fixation for C1-C2 stabilization.
Methods:
A total of 19 patients were included in the study (18 males and 1 female). Their age ranged from 23 to 66 years (mean age of 39.6 years). The patients were diagnosed with complex fractures of the atlas and the axis of the upper cervical spine and underwent posterior temporary C1-2 pedicle screws fixation. The patients underwent a serial postoperative clinical examination at approximately 3, 6, 9 months, and annually thereafter. The neck disability index (NDI) and the range of neck rotary motion were used to evaluate the postoperative clinical efficacy of the patients.
Results:
The average operation time and blood loss were 110 ± 25 min and 50 ± 12 ml, respectively. The mean follow-up was 38 ± 11 months (range 22 to 60 months). The neck rotary motion before removal, immediately after removal, and the last follow-up were 68.7 ± 7.1°, 115.1 ± 11.7°, and 149.3 ± 8.9° (P < 0.01). The NDI scores before and after the operation were 42.7 ± 4.3, 11.1 ± 4.0 (P < 0.01), and the NDI score 2 days after the internal fixation was removed was 7.3 ± 2.9, which was better than immediately after the operation (P < 0.01), and 2 years after the internal fixation was removed. The NDI score was 2.0 ± 0.8, which was significantly better than 2 days after the internal fixation was taken out (P < 0.001).
Conclusions:
Posterior temporary screw fixation is a good alternative surgical treatment for unstable C1-C2 complex fractures.
Keywords
Introduction
Anatomically, the cervical spine is divided into the upper cervical spine (UCS) (C1 and C2) and the lower cervical spine (C3-C7) which is based on morphological and functional differences. The UCS plays a crucial role by protecting the neuronal elements as well as the vertebral artery.
Upper cervical spine fractures are relatively common in practice, accounting for 68.9% of all cervical spine injuries in the elderly population and 35.8% in the younger victims. 1 Single atlas or axis fractures of the UCS account for 10.6% and 20% of cervical fractures, respectively.2,3 However, complex fractures of the UCS are relatively rare, usually caused by motor vehicle accidents, accounting for about 3% of all cervical spine injuries. 4 Five to 53 percent of odontoid fractures have concomitant atlas injuries,5,6 and 6% to 26% of traumatic spondylolisthesis. 7 This is consistent with the present study.
In this series, we define C1-C2 complex fractures as concomitant injuries involving the atlantoaxial milieu leading to instability. These complex UCS lesions are rarely reported in the literature because they are often missed due to the anatomic complexity and the occult nature of the associated injuries. As a result, there is a propensity for residual spine instability and resulting neurological deficits. Kirshenbaum et al 8 studied a series of trauma patients and found a significant number of their patients who sustained upper cervical spine fractures would have been missed on plain radiographs alone without a CT scan. Our experience in the current series is consistent with their suggestion that cervical spine injuries should be a portion of the differential diagnosis in patients with head trauma even if plain films show a normal cervical spine in situations where CT or MRI may be out of reach. The UCS complex fractures usually follow a pattern: C1 lateral mass fracture and dens fracture/ posterior arch of C1 fractures and dens fracture/ dens fracture and Jefferson fracture/ C2 articular pillar fracture and dens fracture. 9
The treatments for upper cervical spine fractures are diverse, including non-surgical and surgical treatment options. 10 This study presents the occurrence of complex fractures involving the atlas and the axis. The study conveys these purposes: 1) To evaluate the clinical efficacy of posterior temporary screw fixation for complex C1-C2 fractures without spinal cord injury; 2) To evaluate the recovery degree of cervical rotation after removal of temporary implants; 3) Discuss the advantages and some drawbacks of posterior temporary pedicle screws fixation (PTSF).
Materials and Methods
General Patient Information
The institutional review board of Qilu Hospital of Shandong University approved the study. Between January 2013 to December 2018, patients diagnosed with C1-2 complex fractures who underwent posterior temporary fixation surgery were enrolled in the study. The inclusion criteria were complex atlas and axis fractures confirmed on X-ray, CT, or MRI. Patients with a history of cervical spine surgery, severe osteoporosis, transverse ligament injury, Hangman’s fractures type II and III (Levine and Edwards modified classification), and incomplete radiological data were excluded from the study.
A total of 19 patients (18 males and 1 female) with C1-C2 complex fractures were selected according to the inclusion and exclusion criteria (Figure 1). Their age ranged from 23 to 66 years (mean age of 39.4 years). The mechanisms of injury were motor vehicle accidents in 17 patients and fall accidents (from a height) in 2 patients.
Study flow diagram showing the inclusion and exclusion of patients.
On physical examination, the patients presented with neck pain, cervical muscle spasms, and restricted neck movement; consistent with the classic clinical sign of upper cervical spine injury (pain, cervical muscle spasm), and limitation of neck movement.11,12 None of the patients included in the study had other potential contraindications such as irreducible fractures, poor bone mineral density, chronic steroid use, and chronic renal disease. After pre-operative plain X-ray films, CT scan, and MRI assessments, 3 patients (15.8%) had a concurrent C1 Jefferson fracture and a dens fracture (Figure 2) Fourteen patients (73.7%), including-thirteen males and 1 female were diagnosed with 3-part fractures of the axis (odontoid (dens) plus hangman’s fracture) (Figure 3 and Figure 4). Two patients (10.5%) had combined Jefferson fracture and Hangman’s fracture, of which one patient also suffered C6/C7 dislocation with minor neurological deficit (Figure 5). The demographics and clinical details of the patients are shown in Table 1.
A 23-year-old male sustained atlantoaxial instability due to a motor vehicle accident. He had a concomitant Jefferson fracture and dens fracture with neck pain. (A): A pre-operative sagittal CT scan showing odontoid fracture measuring 32 degrees angle with 4 mm displacement (solid white arrow); (B and C): pre-operative axial CT scans of the atlas depicting Jefferson fracture (dotted white arrows); (D): postoperative sagittal CT scan at 9 months showing a solid bony union with trabeculation across the fracture site (black solid arrow).
A 36-year-old male presented with neck pain and restricted neck movement after a motor vehicle accident. He sustained odontoid (dens) fracture and Hangman’s fracture. (A): Pre-operative sagittal CT scan showing a complex type IIC odontoid fracture (dotted white arrow); (B): a sagittal CT scan showing fracture of the axis (solid white arrow); (C and D): post-operative coronal and sagittal CT scans at 9 months indicating a solid bony fusion.
A 51-year-old male sustained a UCS complex fracture involving odontoid (dens) fracture and Hangman’s fracture. (A): Pre-operative sagittal CT scan showing dens fracture; (B and C): postoperative sagittal and coronal CT scans at 6 months showing a solid bony union with trabeculation across the fracture site.
A 29-year-old male involved in a motor vehicle accident with severe neck pain was evaluated and diagnosed. He sustained Jefferson fracture combined with Hangman’s fracture. He also suffered a lower cervical spine injury, C6/C7 dislocation. (A): Pre-operative axial computed tomography reconstruction indicating Jefferson fracture (white dotted arrows); (B): axial CT scan of the axis showing a Hangman’s fracture (solid white arrows); (C): post-operative lateral X-ray at 6 months; (D): post-operative lateral view radiology at 9 months follow-up after posterior instruments removal; showing a normal cervical spine alignment and bone union.
Demographics and Characteristics of the Patients (n = 19).
Abbreviations: SD = standard deviation, LCS = lower cervical spine, SCI = spinal cord injury, VA = vertebral artery.
Due to the unstable nature of the injuries, operative treatment was indicated and all patients underwent a posterior temporary C1-2 pedicle screw fixation (PTSF). The time range between injury and operation was 1 to 10 days. All patients underwent pre-operative plain X-rays, MRI, and CT scans assessments to ensure that no bony and vessel anomaly or destruction exits to precludes pedicle or arch screws placement.
Surgical Procedures
All surgeries were performed by the same surgeon. Patients were kept in a prone position under general anesthesia, with continued skull traction and neuromonitoring in place. After a posterior midline incision, the paravertebral muscles from the spinous processes and the lamina were detached to expose the surgical level using monopolar and bipolar cautery.
The C2 screw entry point is ascertained by palpating the medial wall of the C2 pedicle with a Penfield elevator. The screws’ path was directly in line with the visualized C2 dorsal pedicle. A series of hand drills with different lengths were used to create the C2 pedicle trajectory under fluoroscopic guidance. Then bilateral screws with accurate length were then properly placed.
To maintain a secure distance to the vertebral artery, the C1 pedicle screw starting point was placed on the lower half of the posterior ring. A 2-mm burr was used to create the starting point. A ball-tipped feeler was then used to palpate the 4 walls of the trajectory, making sure that there were no breaches. The screw was placed after its length was measured and confirmation of inserting point matched with the trajectory angle by fluoroscope. Finally, appropriate rods size and length were placed without arthrodesis at C1-C2.
Postoperative Clinical Assessment and Follow-Up Course
All patients underwent a series of postoperative clinical evaluations at 3 months, 6 months, 9 months, and annually thereafter. The patients were scheduled and reminded via phone calls to visit the hospital in person for examinations and fractures were assessed with X-ray films and CT scan at each visit. The examinations were executed by 2 experienced doctors who were not involved in the primary operations. The neck disability index (NDI) and visual analog scale (VAS) score were used to evaluate neck pain and discomfort following surgery. At each visit, the neck rotary motion was assessed by calculating the average degree of right and left neck cervical range of motion. This was performed with the patient sitting upright on a chair. Two measuring indicators were applied, one fixed on the head and another taped over the head. Keeping the shoulders still and the eyes horizontally positioned, the patient rotated the neck from a neutral position to the maximum comfortable limit (Figure 6). The starting and ending points are noted and the range of neck rotary motion of both sides calculated. Screws position, fracture alignment, and fracture union status were evaluated. A successful fracture union was considered after both X-ray and reconstruction of cervical CT scans indicated trabeculation across the fracture line (Figures 2D, 3C, 3D, 4B, 4C). The implants were removed between 6 and 12 months after a solid bony fusion of the fracture was confirmed and flexion and extension radiographs indicated stability of the atlantoaxial ligamentous complex. All patients were followed up for at least a year following the second surgery. The neck rotary motion, NDI, and VAS were examined at subsequent visits.
Measurement of the neck rotary motion. The patient sits on a chair with the head in a neutral position, upright with the eyes straight ahead. As he turns the head to the utmost left, the rotation angle is measured. Two indicating devices are used, one fixed on the head (white arrow) and the other fixed over the head (black arrow). The patient then turns the head to the utmost right and the rotation angle is measured. The range of neck rotary motion of both sides is calculated.
Efficacy Evaluation Index
Pre- and post-operative follow-up self-reported clinical outcomes were evaluated. The neck disability index (NDI) was used to measure functional disability resulting from neck pain. Visual analog scale (VAS) was also used to assess neck pain (a score of 0 indicated no pain and 10 signified the worst pain). The range of neck rotary motion (left rotation + right rotation) was evaluated for limitation. Pre-operative neck mobility was severely limited due to the nature of the injuries and to avoid further fracture disruption and potential deterioration of symptoms, neck rotary motion was only measured and evaluated postoperatively (Table 2).
Summary of the Clinical Efficacy of the Patients After PTSF.
Note: * Compared with the previous measured value, P < 0.05.
Statistical Analysis
Statistical analyses were performed using the SPSS 22.0 (SPSS, USA) computer software package. Continues variables were presented as mean and standard errors. Paired t-test was used to evaluate the difference in the intergroup comparison. The α value of the inspection level is 0.05 on both sides.
Results
The average operation time and blood loss were 110 ± 25 minutes and 50 ± 12 ml, respectively. There were no vascular injuries, infections, or neurological lesions observed in any of the patients treated.
Philadelphia collar was used post-operatively for a period of 6 to 8 weeks. Postoperative computed tomography scans were obtained in series at 3, 6, or 9 months (average 7.2 ± 2.0 months). The temporary internal fixations were removed between 6 and 12 months after a solid bony fusion of the fracture was confirmed by CT scans (Figures 2D, 3C, 3D, 4B, 4C). All patients in this series achieved fracture union. Following the implant removal, the atlantoaxial ligamentous complex stability was evaluated with flexion and extension radiographs to ascertain stability. The mean follow-up period was 38 ± 11 months (range 22∼60 months). The pre-and post-operative NDI were 42.7 ± 4.3, 11.1 ± 4.0, respectively (P < 0.01). The average NDI was 7.3 ± 2.9 2 days after instrumentation removal, which was better than that of before instrumentation removal (P < 0.01). At 1-year follow-up after instrumentation removal, the NDI was 2.0 ± 0.8, which was much better than that of 2 days after instrumentation removal (P < 0.01). (Figure 7). The neck rotary motion before and 2-day after instrumentation removal were 68.7 ± 7.1°, 115.1 ± 11.7°, respectively (P < 0.01). At a 2-year follow-up after instrumentation removal, the average neck rotary motion was 149.3 ± 8.9°, which had a significant difference from rotary motion 2-day after instrumentation removal (P < 0.01). (Figure 8).
Neck disability index at different times of follow-up.
Neck rotation measurements at different times of follow-up.
Discussion
A discrete atlas or axial fractures of the UCS do occur and are often reported. However, C1-C2 complex fractures are often occult, and are rarely reported in the literature. The primary aim of this series is to highlight these life-threatening but often missed complex injuries of the UCS. The injury mechanism of UCS complex injuries is an essential component for accurate diagnoses and treatment.
Suchomel et al 13 reported that oblique compression load is the main mechanism of type II odontoid fracture, over-extension is the main mechanism of the anterior and posterior arch of atlas fractures, axial compression load is the main mechanism of atlas burst fracture, and compression combined with lateral flexion is the main mechanism of unilateral atlas lateral mass. Therefore, it is crucial for the physician to select the appropriate treatment for UCS complex fractures, taking into consideration the patient’s characteristics and injury pattern. There is currently no concrete consensus for the treatment of cervical instability at the C1-2 level. However, it is suggested that the treatment goal should focus on reducing the fracture, maintaining stability with the adequate alignment of the spine, and minimizing neurological injuries. 14
Atlantoaxial instability can be fatal due to its anatomic complexity. Hence, the treatment techniques in this region can be an onerous task. 15 Mountains of effort have been made over years to surgically treat injury-related C1-C2 instabilities. It dates back to 1910, when Mixter and Osgood 16 used heavy silk thread wiring techniques to stabilize the atlantoaxial joint. Since then, cumulative efforts have been made to improve the drawbacks of the initial wiring techniques.
Gallie’s approach in 1939, promised good stability on extension and flexion ROM, but rotation was impaired. 17 Brooks and Jenkins improved on that in 1978 by placing 2 iliac crest grafts bilaterally between C1-C2 arches with wiring for stabilization. 18 However, the trajectories of the wires under the C2 laminar posed neurological injury risks. In 1991, Dickman and Sonntag 4 upgraded on the aforementioned drawbacks. In their modification, a sublaminar wire was passed from inferior to superior around the C1 arch, posteriorly. The wire was wedged underneath the C1 arch, different from Gallie’s approach.
The transarticular screw (TAS) technique introduced by Magerl, in 1979 was considered revolutionary as it eliminated the rotational movement drawback. However, the technique endangers the vertebral artery and challenging for less experience surgeons. The application of C1 lateral mass and C2 pedicle screws was noted to be a less difficult procedure and posed lesser vertebral artery risks. Although the C1-C2 cervical spine stabilization techniques have improved significantly over the years, some challenges may still perdue because of the arduous steep learning curve, the anatomic complexity of the UCS, the occult nature of complex C1-2 injuries, and patient characteristics.
All patients in this study were treated with posterior temporary screw fixation and the aforementioned ROM limitations were preserved with this technique. Some of the advantages offered by this surgical option are as follows:
First, posterior internal screw fixation provides a close-up fixation of the fracture and offers a high fusion rate without requiring prolonged halo vest immobilization. Atlantoaxial fixation and stability can be achieved by applying different techniques, such as anterior transarticular screw fixation, posterior transarticular screw fixation, transoral atlantoaxial reduction plate. Harms and Melcher used C1 lateral mass/pedicle, and C2 pedicle screws with polyaxial screw-rod system to achieve atlantoaxial stability. 19 The biomechanical rigidity and firmness achieved with PTSF technique is similar to that of transarticular screws application. We adopted pedicle screws instead of lateral mass screws for C1 temporary fixation because of rigorous dissection to expose the C1 for lateral mass screw fixation which decreases occipital neuralgia and blood loss 20 ; pedicle screws have a higher resistance to pullout than lateral mass screws. 21 In this series, all patients achieved a bone fusion between 6 and 9 months after surgeries.
Second, posterior temporary C1-2 pedicle screw fixation preserves C1-2 rotary function. Sacrificing the atlantoaxial range of motion entirely, especially in younger patients, can have a devastating consequence on the patient. As a result, traditional atlantoaxial bony fusion is not thought to be a preferred treatment in order to preserve the atlantoaxial rotary function.
Guo et al 22 compared the clinical outcomes of posterior C1-2 temporary fixation with posterior C1-C2 fixation and fusion surgical techniques for odontoid fractures. In their study, 22 patients underwent posterior C1-2 temporary fixation and 21 patients underwent posterior C1-C2 fixation with fusion. The authors observed significantly better outcomes in the temporary-fixation group for visual analog scale score for neck pain, NDI, and neck stiffness. The clinical outcomes in the temporary fixation group were superior to those in the fusion group in all dimensions of the 36-Item Short-Form Health Survey. However, they did not identify any significant differences in fracture healing rate and time of healing between the 2 techniques. In our study, the neck disability caused by posterior fixation was obvious. The average NDI before and 2 days after instrumentation removal were 11.1 ± 4.0 and 7.3 ± 2.9, respectively (P < 0.01). The neck rotary motion before and 2-day after instrumentation removal were 68.7 ± 7.1°, 115.1 ± 11.7°, respectively (P < 0.01). At 1-year follow-up, the NDI and neck rotary function were significantly improved. The outcome suggests that temporary fixation can effectively reduce the rate of neck stiffness and disability and maintain the neck rotary motion.
Third, posterior temporary C1-C2 screw fixation can be applicable in injuries that are not suitable for anterior cannulated screw fixation (ACSF). For C2 injuries involving type II odontoid fractures, ACSF is considered the treatment of choice when the transverse ligament is intact since it preserves the atlantoaxial joint rotary function. If successful, ACSF for odontoid type II fractures has a high fusion rate as well as maintaining the rotary motion of the atlantoaxial joint. The reported union rates of surgical treatment ranged from 88% to 100%.23-26 However, it is contraindicated when the fracture line is oblique and anterior. 27 A type IIC fracture with an oblique fracture line from posterior to anterior is considered contraindicated to anterior fixation due to the difficulty associated with interfragmentary compression by the odontoid screw occurring. 28 This suggests the feasibility of posterior temporary screw fixation for C1-C2 complex fractures. The indications include odontoid fracture (Grauer IIC fracture, poor reduction of Grauer II A or II B type fracture) which is not suitable for the ACSF, atlas fracture(simple anterior arch fracture, hemicyclic fracture, anteroposterior arch fracture), and axial fracture (Hangman fracture Levine type II, IIa, Ⅲ and not associated with intervertebral disc and spinal cord injury).29-32
All patients in our group were treated with PTSF, achieved complete fracture healing, and maintained a significant cervical ROM after removal of the instrumentations.
A major drawback of PTSF technique is that a second surgery is required for instrumentation removal which is a concern for both the patient and the surgical team. However, the need to preserve the atlantoaxial ROM, which is responsible for approximately 54% of cervical rotary function, especially in young patients, outweighs the concerns of a second surgery. Though the cervical ROM, for the greater part, is maintained after implant removal, the rotary function of the neck may be affected even after instrumentation removal. The patient may need a prolonged time to recover to the normal range of motions.
In this series, the neck rotary motion at POD2 after instrumentation removal was only 115.1 ± 11.7°. At 1-year follow-up, the average neck rotary motion was improved to 149.3 ± 8.9°(P < 0.01). All the patients in this study were treated successfully with PTSF without any significant complications. However, this technique is not recommended for the treatment of a patient with associated C2-C3 disc disruption. For patients with anterior longitudinal ligament, intervertebral disc injury, or spinal cord compression, a combined anterior and posterior approach should be considered.
Some of the potential complications often encountered using this surgical method include blood loss, vertebral artery injury, and infection. 33
The limitations of the current study include its retrospective nature, small sample size, lack of control group, and potential for selection bias. A prospective randomized control trial with a larger sample size may be necessary for an absolute clinical evaluation of this technique.
Conclusion
Our results suggest that a posterior temporary screw fixation is good alternative management for complex C1-C2 fractures. This approach has a high fusion rate, low complications, maintains the primary atlantoaxial stability, and achieves optimal clinical results.
Footnotes
Authors’ Note
Xinyu Liu, Yakubu Ibrahim, Lianlei Wang, Geng Zhao, Hao Li, and Suomao Yuan participated in the design of the study. Yakubu Ibrahim, Lianlei Wang, Geng Zhao, Hao Li, and Suomao Yuan collected data. Yakubu Ibrahim Lianlei Wang, Hao Li, Geng Zhao, Yiwei Zhao, and Wubo Liu performed the statistical analysis. Yakubu Ibrahim, Hao Li, Geng Zhao, Suomao Yuan, Wubo Liu, and Yonghao Tian conceived of the study, participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript. This study was a retrospective study. The institution review board and ethics committee of Qilu Hospital of Shandong University approved the study.
Acknowledgments
The authors are grateful to the reviewers for their valuable comments on improving this paper.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. It was a retrospective study, so formal patient informed consent was not needed.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Nature Science Foundation (81 874 022) to Xinyu Liu.
