Comparison of Transarticular Screw Fixation and C1 Lateral Mass-C2 Pedicle Screw Fixation in Patients with Rheumatoid Arthritis with Atlantoaxial Instability

Comparison of Transarticular Screw Fixation and C1 Lateral Mass-C2 Pedicle Screw Fixation in Patients with Rheumatoid Arthritis with Atlantoaxial Instability

Original Article Comparison of Transarticular Screw Fixation and C1 Lateral Mass-C2 Pedicle Screw Fixation in Patients with Rheumatoid Arthritis with...

2MB Sizes 0 Downloads 96 Views

Original Article

Comparison of Transarticular Screw Fixation and C1 Lateral Mass-C2 Pedicle Screw Fixation in Patients with Rheumatoid Arthritis with Atlantoaxial Instability Je Il Ryu1, Koang Hum Bak2, Jae Min Kim1, Hyoung-Joon Chun2

BACKGROUND: Many surgical procedures have been introduced to manage atlantoaxial instability caused by rheumatoid arthritis (RA) to prevent complications and improve fusion rate. We report the surgical outcome between transarticular screw fixation (TAF) and C1 lateral mass-C2 pedicle screw fixation (C1LM-C2P) in patients with atlantoaxial instability from RA.

-

METHODS: Between 2002 and 2012, 58 patients were enrolled in the study. According to surgical procedures, patients were divided into 2 groups: group I who received TAF (n [ 33) and group II who received C1LM-C2P (n [ 25). Bony fusion was assessed by radiologic comparison immediately after the operation and 1 year postoperatively. In addition, complications and clinical and functional outcomes were evaluated.

postoperative period and fewer surgery-related complications in patients with RA. Because surgical complications are more likely during the learning curve (as with other surgical techniques), surgeons should carefully evaluate patients before surgery by radiologic and neurologic examinations.

-

RESULTS: Overall, bone fusion was achieved in 32 patients in group I (97%). In group II, the fusion rate was evaluated in 100% of patients. Complications (regardless of neurologic deterioration) were cable loosening and screw malposition in group I and violation into the vertebral canal and spinal canal in group II. There was no statistical significance in fusion rate, clinical outcomes, or complications. The 12-month atlantodental interval after operation for the C1LM-C2P group was significantly lower than that for the TAF group after adjusting for all variables.

-

CONCLUSIONS: Two surgical techniques showed a good fusion rate by rigid fixation in the immediate

-

Key words Atlantoaxial instability - Bone fusion - C1 lateral mass-C2 pedicle screw fixation - Complication - Rheumatoid arthritis - Transarticular screw fixation -

Abbreviations and Acronyms AAI: Atlantoaxial instability ADI: Atlantodental interval C1LM-C2P: C1 lateral mass and C2 pedicle screw C-3D CT: Cervical three-dimensional computed tomography RA: Rheumatoid arthritis TAF: Transarticular screw fixation

WORLD NEUROSURGERY 99: 179-185, MARCH 2017

INTRODUCTION

R

heumatoid arthritis (RA) is a chronic inflammatory disease that has a worldwide prevalence in the adult population of 1%e2%.1 It is characterized by symmetric progressive polyarthritis and subsequent joint destruction and is associated with high morbidity and mortality. Morbidity and mortality are particularly increased when the disease involves the upper cervical region.2-4 When this situation occurs, the patients are at risk of myelopathy, vertebrobasilar insufficiency, and sudden death as a result of brainstem compression. In particular, the atlantoaxial joint is often affected by synovitis in the upper cervical region because the transverse ligament is affected during the early stage of disease, resulting in increased laxity. Later in the disease, the cartilage and bony structures of the joint itself are eroded.5,6 This development leads to atlantoaxial instability (AAI). Several methods for overcoming AAI have been presented. They include pharmacologic intervention and surgery. Kauppi et al.7 reported that traditional disease-modifying antirheumatic drugs prevent AAI in patients with recent-onset RA. Several surgical approaches were also reported in the past, including the Gallie

VA: Vertebral artery VAS: Visual analog scale From the 1Department of Neurosurgery, Hanyang University Guri Hospital, Hanyang University College of Medicine, Guri; and 2Department of Neurosurgery, Hanyang University Seoul Hospital, Hanyang University College of Medicine, Sungdong-gu, Seoul, Republic of Korea To whom correspondence should be addressed: Hyoung-Joon Chun, M.D, Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2017) 99:179-185. http://dx.doi.org/10.1016/j.wneu.2016.12.028 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.

www.WORLDNEUROSURGERY.org

179

ORIGINAL ARTICLE JE IL RYU ET AL.

TAF AND C1LM-C2P FIXATION IN RA

fusion, the Brooks fusion, and interlaminar clamping. However, more recently, 2 different screw systems that aim to fuse C1-2 have become widely used: transarticular screw fixation (TAF) and C1 lateral mass-C2 pedicle screw fixation (C1LM-C2P).8-15 These screw systems are both associated with favorable fusion rates,8,9,15-19 although the atlantoaxial segment generally has lower fusion rates than the subaxial cervical spine because of its high mobility (50% of total neck rotation and 12% of total neck flexion-extension occurs between C1 and C2). Although many studies have compared the outcomes of surgical procedures in AAI, the patients were heterogeneous in terms of the cause of AAI and the studies reported only the rate of biomechanical instability correction. However, studies of the success of these surgical procedures in patients with RA only are warranted because it is more difficult to achieve C1-2 bone fusion in these patients. This situation is because their disease causes additional problems, including severe laxity of the ligament, poor vascularity, osteoporosis, and demineralization as a result of longterm use of steroid and cytotoxic agents.20,21 The aim of the present retrospective cohort study was to compare TAF and C1LM-C2P for AAI in terms of both clinical and surgical outcomes in patients with RA only. METHODS This retrospective cohort study was approved by the institutional review board of Hanyang University Medical Center in Seoul, Republic of Korea (approval no. HYUH 2016-06-032-001) and was conducted according to the Declaration of Helsinki and its revisions. Study Design All consecutive patients with RA and AAI who underwent TAF or C1LM-C2P between January 2002 and December 2012 at Hanyang University Medical Center were identified by retrospective review of a prospectively collected medical database. Surgery was performed by 3 surgeons. All patients underwent only C1-2 level fusion. Patients with a traumatic lesion, os odontoideum, basilar invagination, occipitocervical fusion, and less than 1 year of follow-up were excluded. All patients were referred to our department when AAI developed during follow-up for RA in the rheumatology department of our hospital. All patients were followed up by scheduled visits to our department 1, 6, 12, and, in some patients, 24 months after surgery. Preoperative Evaluation All patients underwent the following radiologic evaluations before surgery. Cervical radiography with flexion/extension views was performed to assess the cervical alignment and the atlantodental interval (ADI). Fine-cut cervical three-dimensional computed tomography (C-3D CT) scans (1-mm slices) were used to predict the screw pathway and assess the course of the vertebral artery (VA). Cervical magnetic resonance imaging was performed to assess the status of the spinal cord and the atlantoaxial ligaments. Management in Perioperative Period When AAI was diagnosed in patients with RA, they were treated with several drugs to control the RA, including a cytotoxic agent

180

www.SCIENCEDIRECT.com

(methotrexate), steroids, and antiinflammatory agents. One week before surgery, the patients stopped taking the agents to avoid perioperative complications such as hematoma and infection. The RA medication was restarted 2 weeks after the operation. All patients wore a neck brace for 3 months after atlantoaxial fusion. Surgical Procedures TAF was used for AAI in our institution until 2007, at which time we started to use C1LM-C2P for AAI. Thus, during the 10-year study period, patients were treated with TAF from 2002 to 2007, after which C1LM-C2P was used in 2008e2012. Interlaminar Wiring. Both TAF and C1LM-C2P were combined with interlaminar wiring according to the Brooks method using an interlaminar spacer such as an allograft or titanium mesh cage. For the TAF cases, the titanium mesh cage was harvested with either bovine xenograft (Lubboc [Transphyto S.A., ClermontFerrand, France]) or demineralized bone matrices (DBX [Synthes, Paoli, Pennsylvania, USA]). Thus, under general anesthesia, the patient was placed in a prone position in which the neck was slightly flexed to achieve cervical alignment. After the operative site was prepared and draped in the usual aseptic fashion, the atlas and axis were exposed sufficiently by midline suboccipital incision. To place the interlaminar spacer, a titanium cable (Atlas cable [Medtronic Sofamor Danek, Memphis, Tennessee, USA]) was passed under the posterior arch of the atlas and coiled up the spinous process of the axis. The fusion bed was prepared by drilling on the inferior aspect of the C1 posterior arch, superior margin of the C2 lamina, and spinous process of the axis. The interlaminar spacer (a titanium mesh cage or an allograft) was anchored tightly between the C1-C2 space, and the titanium cable was tightened over it. Permanent tightening of the interlaminar wire and the inserted spacer was performed before TAF and after C1LM-C2P. TAF. After reducing the C1-2 alignment by fastening the titanium cable, guide pins were inserted under C-arm guidance along their pathway, namely, through the dorsal elements of C2 and across the C1-2 articulation into the lateral mass of C1. The entry point was on the posterior margin of the inferior articular process of the axis and 3e4 mm lateral to the medial border of the facet. When the guide pins were located on the pathway, cannulated transarticular screws with a diameter of 3.5e4.0 mm were inserted. C1LM-C2P. After the C1-C2 posterior structures were widely exposed with meticulous coagulation of the venous plexus in the C1-C2 articulation, the starting point for the C1 lateral mass was established 2 mm lateral to the connection point of the posterior arch and the lateral mass. If the C1 posterior arch was not thick enough, the C2 nerve was retracted caudally and the entry point was made directly on the posterior portion of the C1 lateral mass. A 2.7-mm drill was then passed gently through the C1 lateral mass with a straight or slightly convergent trajectory. After tapping the drilled pathway, screws that had a diameter of 3.5e4.0 mm and a length of 28e32 mm were inserted into the lateral mass. C2 pedicle screw fixation was performed as described by Chun et al.22 Before marking the entry point, the medial border of the C2 pars interarticularis was identified to prevent screw malposition into the spinal canal. After identifying the transitional corner, which

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.12.028

ORIGINAL ARTICLE JE IL RYU ET AL.

was the more cephalad portion of the lamina and the C2 isthmus, the entry point was made 4 mm lateral and 4 mm caudal to the transitional corner using the drill. Under C-arm guidance, the direction was 20 e30 in a medial and cephalad direction with palpation of the medial border of the C2 pars interarticularis. After drilling and tapering, a polyaxial screw with the appropriate length and diameter was inserted. The interlaminar spacer and wiring between C1 and C2 were tightly fastened, and the rod was applied between C1 and C2 with screw compression. Postoperative Evaluation Follow-Up Imaging. Plain radiographs were obtained on the operative day and 1, 6, 12, and 24 months after atlantoaxial fusion surgery. Anteroposterior and lateral views (neutral, flexion, and extension) were taken. Cervical radiographs were assessed for bone union, cervical alignment, and instrumentation stability. C3D CT was performed 1 month after surgery to assess the screw placement. In addition, if there were any concerns about screw placement during the operation, CT was also performed during surgery. Additional CT scans were performed 6 and 12 months after surgery to evaluate the bony fusion process. Assessment of Bony Fusion. Bone fusion was considered to have occurred when there was no significant difference between the ADIs in the plain radiographs that were taken immediately after surgery and 12 months later. Bony union was also considered to have occurred when there was no abnormal motion on C1-2 in the flexion and extension view and when instrumentation failure was not observed. If the cervical CT scans 12 months after surgery indicated bridging trabeculation on the C1-C2 interlaminar space, we considered that the fusion had been successful. Measurement of Clinical Outcomes. At each follow-up visit, clinical and functional outcomes were assessed using the visual analog scale (VAS), Japanese Orthopedic Association score, and the Ranawat classification. In addition, complications were evaluated, including infection, instrumentation failure, spinal canal encroachment, screw malposition, and VA injury. Statistical Analysis The Mann-Whitney U test or c2 test was used as appropriate to compare data from the TAF and C1LM-C2P groups (demographic variables, operative times, complication rates, follow-up duration, and overall successful fusion rates). These nonparametric tests were selected because of the low sample numbers. The Wilcoxon signed rank test was performed to assess whether ADI changed significantly between 1 and 12 months after surgery. A Fischer exact test was used to compare the 2 groups in terms of the frequency of bilateral fixation. Univariable and multivariable linear regression analyses were performed to identify factors predictive of 12-month ADI after surgery. The following variables were included: sex, age, treatment type (TAF and C1LM-C2P), preoperative VAS score, preoperative ADI, preoperative myelopathy, preoperative Ranawat classification, preoperative Japanese Orthopedic Association score, and screw placement. Statistical analyses were performed using R version 3.1.2 (R Foundation for Statistical Computing, Vienna, Austria) and SPSS version 21.0 (IBM Corp., Armonk, New York, USA).

WORLD NEUROSURGERY 99: 179-185, MARCH 2017

TAF AND C1LM-C2P FIXATION IN RA

Table 1. Demographic and Surgical Characteristics of the Patients Group I: Transarticular Fixation

Group II: C1 Lateral Mass-C2 Pedicle Screw Fixation

Surgery period

2002e2007

2008e2012

Total patients

33

25

46.61  14.19

54.68  11.40

0.023 0.523

Mean age standard deviation Female (%)

30 (90.9)

22 (88.0)

12.49 (12e36)

18.72 (12e24)

Both

14 (42.4)

14 (56.0)

Right

11 (33.3)

3 (12.0)

Left

8 (24.2)

8 (32.0)

9 (27.3)

9 (36.0)

Mean follow-up period, months (range)

P Value

Screw placement (%)

<0.001

Interlaminar spacer (%) Mesh þ demineralized bone matrix Mesh þ Lubboc Allobone

0.171

24 (72.7)

0

0

16 (64.0)

RESULTS Demographic and Surgical Characteristics of the Patients In total, 85 patients underwent TAF or C1LM-C2P during the study period (Table 1). Of these patients, 27 were excluded because they had a traumatic lesion, os odontoideum, basilar invagination, or occipitocervical fusion, or were followed up for less than 1 year. Thus, 58 patients with AAI caused by RA were enrolled in the present study. Of these patients, 33 underwent TAF in 2002e2007 and 25 underwent C1LM-C2P in 2008e2012. In the TAF and C1LM-C2P groups, the average age at surgery was 46.61 years (range, 13e70 years) and 54.68 years (range, 30e74 years), respectively (P ¼ 0.023), and the male/ female ratio was 1:10 and 1:7.33, respectively (P ¼ 0.523). All patients were followed for at least 12 months, and the mean follow-up periods were 12.49 and 18.72 months. TAF was performed bilaterally and unilaterally in 14 (42.4%) and 19 (57.6%) patients, respectively. In all cases, a titanium mesh cage with bovine xenograft or demineralized bone matrices served as the interlaminar spacer. C1LM-C2P was performed bilaterally and unilaterally in 14 (56.0%) and 11 (44.0%) patients, respectively. Thus, C1LM-C2P was performed bilaterally more frequently than TAF, although this difference did not achieve statistical significance (P ¼ 0.427). The interlaminar spacer in C1LM-C2P was either allobone (n ¼ 16) or mesh cage (n ¼ 9). This situation reflects the fact that C1LM-C2P was performed by 2 neurosurgeons with different preferences in terms of the interlaminar spacer.

www.WORLDNEUROSURGERY.org

181

ORIGINAL ARTICLE JE IL RYU ET AL.

TAF AND C1LM-C2P FIXATION IN RA

all remaining 53 patients indicated bony union with trabecular formation on the C1-2 interlaminar space.

Table 2. Clinical and Surgical Outcome Group II: C1 Group I: Lateral Mass-C2 Transarticular Pedicle Screw P Fixation Fixation Value Atlantodental interval 1 month, postoperative, mean  SD

2.66  0.44

1.97  0.68

<0.001

12 months, postoperative, mean  SD

2.60  0.41

1.91  0.45

<0.001

Preoperative (S3a), n (%)

3 (9.1)

3 (12.0)

0.523

Postoperative (S3a), n (%)

1 (3.0)

2 (8.0)

0.395

Ranawat classification

Visual analog scale score Preoperative, mean  SD

8.18  0.95

8.24  0.93

0.816

1 month, postoperative, mean  SD

2.55  0.71

3.12  0.88

0.008

12 months, postoperative, mean  SD

1.67  0.54

1.60  0.58

0.653

Japanese Orthopedic Association score Preoperative, median (interquartile range)

15.0 (14.0e16.0) 15.0 (14.5e16.0)

0.822

Postoperative, median (interquartile range)

16.0 (16.0e17.0) 16.0 (15.0e17.0)

0.512

Fusion, n (%)

32 (97)

25 (100)

SD, standard deviation.

Clinical Outcomes Before surgery, all patients reported neck pain that was aggravated by neck rotation (Table 2). In the TAF and C1LM-C2P groups, the mean preoperative VAS scores were 8.18 and 8.24, respectively (P ¼ 0.816). These scores were lower 1 month after surgery (2.55 and 3.12, respectively; P ¼ 0.008) and much lower 12 months after surgery (1.67 and 1.60, respectively; P ¼ 0.653). After adjusting for all variables, the 12-month postoperative ADI for the C1LM-C2P group was significantly lower than that for the TAF group (B, 0.654; 95% confidence interval, 0.924 to 0.384; P < 0.001) (Table 3).

Bone Fusion Assessment In the TAF and C1LM-C2P groups, overall successful fusion was noted in 97% (32/33) and 100% (25/25), respectively. Analysis of the mean change in ADI at 12 months relative to 1 month after surgery showed that the change was not significant for the TAF or C1LM-C2P groups (Figure 1). This finding was taken to indicate bone fusion at C1-2. Radiologic follow-up with lateral radiographs taken in flexion and extension indicated that all but 1 patient did not show abnormal motion 12 months after surgery. Five patients could not undergo C-3D CT scans at the 12-month follow-up visit because they could not afford it. The CT scans in

182

www.SCIENCEDIRECT.com

Surgery-Related Complications There were 4 complications, 2 in each group. In the TAF group, 1 patient showed fusion failure as a result of screw malposition (the patient mentioned earlier) and 1 patient showed cable loosening 1 month after surgery. The patient who had the screw malposition underwent revision surgery with C1LM-C2P because of the development of increased neck pain and atlantoaxial motion. None of the TAF patients showed neurologic deficits, pain aggravation relative to before surgery, or migration of the interlaminar spacer during surveillance. In the C1LM-C2P group, 2 patients showed screw malposition and violation of the vertebral foramen. In 1 of these patients, the screws bilaterally violated the medial cortical bone of the C1 lateral mass. At the 1-month postoperative visit, the patient reported slight neck pain and reported a VAS score of 4. During subsequent follow-up, we confirmed that the screw was stable because there was no hollow formation or loosening. Bony bridging through the interlaminar spacer was noted at the 12-month postoperative visit. The patient also no longer reported neck pain. In the second patient, the surgeon noted massive bleeding that was suspicious of VA injury during tapering of the right-side C2 pedicle. Consequently, the surgeon had to place the screw into the left side of the C2. Immediate postoperative CT scans showed that the right vertebral foramen was violated by about 50% by the C2 screw. Conventional angiography showed that the VA had not been directly torn, although the screw had compressed the VA such that it had 50% stenosis. Although the patient did not report any neurologic symptoms or deficits, follow-up CT angiography showed right-side VA shrinkage. None of the C1LM-C2P patients showed neurologic deficits, pain aggravation relative to before surgery, or migration of the interlaminar spacer during surveillance. DISCUSSION RA is a chronic systemic autoimmune disorder of unknown cause that is characterized by symmetric synovial inflammation and tenderness in multiple joints, particularly the minor joints of the upper and lower extremities.2,10,21,23,24 RA affects an estimated 1%e2% of the adult population worldwide. In the United States alone, more than 2 million people are affected. The most important concept in contemporary RA treatment is the administration of disease-modifying antirheumatic drugs as soon as the disease is diagnosed.7 Spine involvement in RA is almost completely limited to the upper cervical spine, particularly the atlantoaxial joint. The synovial joints around the atlantoaxial segment are frequently affected. This situation is associated with chronic instability of the atlantoaxial joint and is affected by repetitive motions. This synovitis results in clinical and radiologic AAI.2,3,6,11 Obvious risk factors for cervical spine involvement in patients with RA include erosive peripheral joint disease and the diagnosis of RA at an early age. Before the development of screw systems for the C1-2 lesion, various cable systems were used to control the AAI. However, these approaches were associated with low fusion rates. This

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.12.028

ORIGINAL ARTICLE JE IL RYU ET AL.

TAF AND C1LM-C2P FIXATION IN RA

Table 3. Predictive Factors for 12-Month Atlantodental Interval After Surgery Using Univariable and Multivariable Linear Regression Univariable Linear Regression B (95% Confidence Interval)

Multivariable Linear Regression P

B (95% Confidence Interval)

P

Sex

0.023 (0.454 to 0.499)

0.925

0.027 (0.403 to 0.457)

0.899

Age

0.016 (0.026 to 0.006)

0.002

0.008 (0.018 to 0.002)

0.116

<0.001

C1 lateral mass and C2 pedicle screw (vs. transarticular screw fixation)

0.695 (0.922 to 0.469)

Preoperative visual analog scale score

0.039 (0.196 to 0.118)

0.622

0.008 (0.149 to 0.133)

0.654 (0.924 to 0.384)

<0.001 0.911

Preoperative atlantodental interval

0.005 (0.081 to 0.072)

0.901

0.002 (0.072 to 0.069)

0.960

Preoperative myelopathy

0.067 (0.354 to 0.488)

0.751

0.060 (0.856 to 0.737)

0.881

S3a (vs. <3a)

0.225 (0.248 to 0.698)

0.346

0.234 (0.516 to 0.984)

0.534

Preoperative Japanese Orthopedic Association score

0.058 (0.146 to 0.030)

0.190

0.017 (0.156 to 0.122)

0.809

Left (vs. right)

0.018 (0.425 to 0.389)

0.930

0.125 (0.231 to 0.481)

0.484

Both (vs. right)

0.095 (0.459 to 0.270)

0.605

0.114 (0.214 to 0.443)

0.487

situation encouraged the development of surgical techniques involving screws. Thus, in 1987, Magerl et al. were the first to show the TAF method for stabilizing AAI.25 However, although this method is associated with high fusion rates, it has drawbacks, including the possibility that the screw may violate the vertebral foramen and spinal canal, which could lead to VA injury: the overall risk of VA injury for patients undergoing TAF is reported to be 4.1%. Moreover, the risk of VA injury per screw inserted is 2.2%.26-28 To prevent VA injury in TAF, modifications of the original method have been presented, including use of a stereotactic system and an aiming device and modification of the entry point. We suggested previously that TAF in patients with AAI should be limited to unilateral screw insertion when bilateral TAF is contraindicated because of complications such as a high-riding VA.26 However, recently, surgeons started replacing TAF with C1LM-C2P fixation because it associates with a lower risk of VA injury.13,16,20 The other advantages of C1LM-C2P fixation over TAF include the ability to perform intraoperative reduction and fixation of the atlantoaxial segment, the biomechanical stability of the flexion-extension, lateral bending forces, and resistance against translational force, and a lower risk of injury to the spinal cord.15,28 We found no statistically significant difference in ADI between the TAF group and the C1LM C2P group (Table 2), suggesting that there is no difference in the rate of fusion between the 2 surgical methods, as reported by previous studies.8,14,16,17 In addition, we found that the 12-month postsurgical ADI in the C1LM-C2P group decreased by about 0.7 mm compared with that in the TAF group (after adjusting for all variables) (Table 3). However, these values were not clinically significant. Therefore, the surgeon may either decide that they are familiar with the surgical method or they may favor the patient’s anatomic and clinical situation. Also, there were no marked differences between the 2 surgical techniques except that C1LM-C2P tended to be associated with a higher rate of bilateral screw fixation (14/25, 56.0%) than did TAF (14/33, 42.4%). We believe that this trend reflects the fact that the

WORLD NEUROSURGERY 99: 179-185, MARCH 2017

three-dimensional pathway of the screw is more difficult and complicated in the TAF technique. There are 2 factors that can inhibit the use of bilateral screw fixation, namely, the risk of spinal canal encroachment and the presence of an anomalous VA course. However, because the ratio of the cord to the spinal canal is small at the C1-2 level, it is less likely that violation of the canal will lead to neurologic deficits. Thus, the main concern associated with screw fixation is an anomalous VA course such as a highriding course or anomalous traversing of the vessel into bony structures. In TAF, the screw passes from the lamina of the axis to

Figure 1. Average diameter of the atlantodental interval (ADI) (with standard deviation) before surgery and at 1 month and 12 months after surgery (based on the transarticular fixation [TAF] and C1 lateral mass-C2 pedicle screw fixation [C1LM]).

www.WORLDNEUROSURGERY.org

183

ORIGINAL ARTICLE JE IL RYU ET AL.

TAF AND C1LM-C2P FIXATION IN RA

the lateral mass of the atlas, thus crossing the facet joint of the C12. Because it passes through these different vertebrae, it is difficult to ensure good screw location and avoid VA injury. By contrast, in C1LM-C2P, the screw is inserted only into individual vertebrae (e.g., the lumbar pedicle screw) and the VA course is visualized by careful dissection. If there is sufficient thickness for screw placement, the screw can be inserted without any concern. The fact that many patients who underwent C1LM-C2P in this study could receive bilateral screw fixation is a good result, considering that patients with RA have poor bone condition and a higher risk of VA anomalies than do patients with AAI caused by traumatic lesion.6,8,14 However, there were some mild differences between the TAF and C1LM-C2P groups in terms of age and gender (the TAF patients tended to be younger and more likely to be female), which suggests that the groups may not be entirely comparable because of patient-related bias. Thus, ever since Magerl and Seemann first described bilateral C1-2 TAF with interspinous wiring (i.e., TAF), it has been in general the best treatment for AAI. However, although it is associated with an excellent fusion rate because it involves the union of 4 cortical surfaces, it is a technically challenging procedure that requires strong visual-spatial perception in the operators. In addition, the advantages of an interlaminar spacer include a lack of donor-site morbidity and the attainment of immediate, rigid mechanical stability via 3-point fixation. As a result, satisfactory placement of bilateral screws is not always possible. This technique is also not sensitive to anatomic variation in VA, and thus is associated with a considerable risk of VA injury. Consequently, this method was rapidly abandoned when Harms and Melcher13 described the technique of connecting C1 lateral mass screws to C2 pars screws using rods. This method is associated with a smaller risk of VA injury and is not markedly affected by anatomic variation. The technique can be performed by beginners without any difficulty or risk. Patients with RA with upper cervical spine disorders are at particular risk of AAI. Until this study, it was not known how

REFERENCES 1. Linos A, Worthington JW, O’Fallon WM, Kurland LT. The epidemiology of rheumatoid arthritis in Rochester, Minnesota: a study of incidence, prevalence, and mortality. Am J Epidemiol. 1980;111:87-98. 2. Halla JT, Hardin JG, Vitek J, Alarcon GS. Involvement of the cervical spine in rheumatoid arthritis. Arthritis Rheum. 1989;32:652-659. 3. Heywood AW, Learmonth ID, Thomas M. Cervical spine instability in rheumatoid arthritis. J Bone Joint Surg. 1988;70:702-707. 4. Moncur C, Williams HJ. Cervical spine management in patients with rheumatoid arthritis. Review of the literature. Phys Ther. 1988;68:509-515. 5. Lehto EM, Heikkila S, Kautiainen H, Kauppi MJ. The clinical picture of atlantoaxial subluxation changes when atlantoaxial impaction develops. Joint Bone Spine. 2010;77:159-164.

184

www.SCIENCEDIRECT.com

relatively effective TAF and C1LM-C2 are in this population because previous studies comparing these techniques used heterogeneous patient populations. This is an important question because patients with AAI caused by RA differ from other patients with AAI in several unique ways that can interfere with bone fusion. First, patients with RA have increased atlantoaxial motion because of ligament laxity and the pathology. Second, the medications against RA, including steroids, nonsteroidal antiinflammatory drugs, and cytotoxic agents, can impair bone fusion and healing. Third, most patients have osteoporosis and poor vascularity, both of which significantly hamper fusion. Thus, it is more difficult to induce immobility and stabilization in patients with AAI caused by RA. This study has several limitations. First, the size of the patient group was relatively small. Second, the retrospective nature of the analysis in this study means that there may be some patient selection bias. Third, the only symptom that was investigated before and after surgery was neck pain, as measured by VAS: the effect of AAI fixation with screws on other symptoms was not assessed. These limitations indicate the need for further research and clinical studies in this field. CONCLUSIONS TAF and C1LM-C2P in patients with RA yielded good fusion rates in the immediate postoperative period, although bilateral screw placement tended to be more common in C1LM-C2P than in TAF (56.0% vs. 42.4%). Both techniques were associated with few surgery-related complications. Nevertheless, given that surgical complications are more likely during the learning curve (as with other surgical techniques), surgeons should carefully evaluate patients before surgery by radiologic and neurologic examinations. ACKNOWLEDGMENTS English grammar was revised by Bioedit Ltd.

6. Riise T, Jacobsen BK, Gran JT. High mortality in patients with rheumatoid arthritis and atlantoaxial subluxation. J Rheumatol. 2001;28:2425-2429. 7. Kauppi MJ, Neva MH, Laiho K, Kautiainen H, Luukkainen R, Karjalainen A. Rheumatoid atlantoaxial subluxation can be prevented by intensive use of traditional disease modifying antirheumatic drugs. J Rheumatol. 2009;36:273-278. 8. Elliott RE, Tanweer O, Boah A, Morsi A, Ma T, Frempong-Boadu A. Outcome comparison of atlantoaxial fusion with transarticular screws and screw-rod constructs: meta-analysis and review of literature. J Spinal Disord Tech. 2014;27:11-28. 9. Goel A, Laheri V. Plate and screw fixation for atlanto-axial subluxation. Acta Neurochir (Wien). 1994;129:47-53. 10. Goel A, Dange N. Immediate postoperative regression of retroodontoid pannus after lateral mass reconstruction in a patient with rheumatoid disease of the craniovertebral junction. Case report. J Neurosurg Spine. 2008;9:273-276.

11. Goel A, Sharma P. Craniovertebral realignment for basilar invagination and atlantoaxial dislocation secondary to rheumatoid arthritis. Neurol India. 2004;52:338-341. 12. Goel A, Pareikh S, Sharma P. Atlantoaxial joint distraction for treatment of basilar invagination secondary to rheumatoid arthritis. Neurol India. 2005;53:238-240. 13. Harms J, Melcher RP. Posterior C1-C2 fusion with polyaxial screw and rod fixation. Spine. 2001;26: 2467-2471. 14. Hartl R, Chamberlain RH, Fifield MS, Chou D, Sonntag VK, Crawford NR. Biomechanical comparison of two new atlantoaxial fixation techniques with C1-2 transarticular screw-graft fixation. J Neurosurg Spine. 2006;5:336-342. 15. Sim HB, Lee JW, Park JT, Mindea SA, Lim J, Park J. Biomechanical evaluations of various c1-c2 posterior fixation techniques. Spine. 2011;36: 401-407.

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.12.028

ORIGINAL ARTICLE JE IL RYU ET AL.

16. Claybrooks R, Kayanja M, Milks R, Benzel E. Atlantoaxial fusion: a biomechanical analysis of two C1-C2 fusion techniques. Spine J. 2007;7: 682-688. 17. Du JY, Aichmair A, Kueper J, Wright T, Lebi DR. Biomechanical analysis of screw constructs for atlantoaxial fixation in cadavers: a systematic review and meta-analysis. J Neurosurg Spine. 2015;22: 151-161. 18. Lapsiwala SB, Anderson PA, Oza A, Resnick DK. Biomechanical comparison of four C1 to C2 rigid fixative techniques: anterior transarticular, posterior transarticular, C1 to C2 pedicle, and C1 to C2 intralaminar screws. Neurosurgery. 2006;58:516-521. 19. Lee SH, Kim ES, Sung JK, Park YM, Eoh W. Clinical and radiological comparison of treatment of atlantoaxial instability by posterior C1-C2 transarticular screw fixation or C1 lateral mass-C2 pedicle screw fixation. J Clin Neurosci. 2010;17: 886-892. 20. Chun HJ, Oh SH, Yi HJ, Ko Y. Efficacy and durability of the titanium mesh cage spacer combined with transarticular screw fixation for atlantoaxial instability in rheumatoid arthritis patients. Spine. 2009;34:2384-2388.

TAF AND C1LM-C2P FIXATION IN RA

21. Yurube T, Sumi M, Nishida K, Takabatake M, Kohyama K, Matsubara T. Progression of cervical spine instabilities in rheumatoid arthritis: a prospective cohort study of outpatients over 5 years. Spine. 2011;36:647-653. 22. Chun HJ, Bak KH. Targeting a safe entry point for c2 pedicle screw fixation in patients with atlantoaxial instability. J Korean Neurosurg Soc. 2011;49: 351-354. 23. Ahn JK, Hwang JW, Oh JM, Lee J, Lee YS, Jeon CH. Risk factors for development and progression of atlantoaxial subluxation in Korean patients with rheumatoid arthritis. Rheumatol Int. 2011;31:1363-1368.

27. Wright NM, Lauryssen C. Vertebral artery injury in C1-2 transarticular screw fixation: results of a survey of the AANS/CNS section on disorders of the spine and peripheral nerves. American Association of Neurological Surgeons/Congress of Neurological Surgeons. J Neurosurg. 1998;88: 634-640. 28. Yoshida M, Neo M, Fujibayashi S, Nakamura T. Comparison of the anatomical risk for vertebral artery injury associated with the C2-pedicle screw and atlantoaxial transarticular screw. Spine. 2006; 31:513-517.

24. Imagama S, Oishi Y, Miura Y, Kanayama Y, Ito Z, Wakao N. Predictors of aggravation of cervical spine instability in rheumatoid arthritis patients: the large joint index. J Orthop Sci. 2010;15:540-546.

Conflict of interest statement: This work was supported by the research fund of Hanyang University (HY-2013).

25. Grob D, Magerl F. Surgical stabilization of C1 and C2 fractures. Orthopade. 1987;16:46-54.

Citation: World Neurosurg. (2017) 99:179-185. http://dx.doi.org/10.1016/j.wneu.2016.12.028

26. Hue YH, Chun HJ, Yi HJ, Oh SH, Oh SJ, Ko Y. Unilateral posterior atlantoaxial transarticular screw fixation in patients with atlantoaxial instability: comparison with bilateral method. J Korean Neurosurg Soc. 2009;45:164-168.

Journal homepage: www.WORLDNEUROSURGERY.org

WORLD NEUROSURGERY 99: 179-185, MARCH 2017

Received 2 October 2016; accepted 8 December 2016

Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.

www.WORLDNEUROSURGERY.org

185