Introduction
As the cervical spine is considered the most mobile part of the spinal column, a wide
range of normal alignment has been described.[[1]] Cervical sagittal malalignment has received increased attention in the literature
as it has found to be linked to clinical symptoms, degenerative diseases progression
as well as clinical outcomes.[[2]] Patients suffering from cervical spondylotic myelopathy (CSM) treated by posterior
cervical decompression alone without fixation are considered among the most common
causes of cervical malalignment (iatrogenic cervical kyphosis), so posterior cervical
fixation became an ideal procedure in the surgical management of the unstable spondylotic
cervical spine.[[3]]
Posterior fixation using lateral mass screw fixation has been widely accepted as a
simple, safe, and effective method with a high rate of fusion. Due to its ease of
use, as well as its good intrarater and interrater reliability, Cobb angle method
remains the clinical mainstay for the measurement of cervical lordosis.[[4]]
The aim of this study is to access the effect of the sub-axial cervical lateral mass
screw fixation on cervical sagittal alignment, and it's impact on the functional outcome
in patients with CSM.
Patients and Methods
This is a prospective observational study of pre- and post-operative clinical and
radiological data of 30 patients admitted to our hospital suffering from CSM with
instability and/or kyphosis, all patients were operated on for posterior lateral mass
fixation with fusion and laminectomy (long-segment fixation ≥3 segments).
Patients that clearly documented physical examination findings consistent with progressive
myelopathy or myeloradiculopathy who failed nonoperative measures and radiographic
(magnetic resonance image [MRI]) confirmation of cord compression at three or more
cervical levels with kyphotic curve or lordotic curve with evidence of instability
were included in this study.
Any patient underwent surgery for CSM was included, but it is not the aim of our study
to discuss these criteria, our study aims to access the relation between the changes
in the sub-axial sagittal alignment represented by Cobb angle and the functional outcome
represented in the motor power, Nurick myelopathy scale, and neck disability index
(NDI).
Patients with inability to give informed consent, stenotic single level, or patients
whose presenting complaint was axial neck pain alone, traumatic fractures, tumor,
metabolic disorders, and patients who had concurrent anterior cervical spine procedures
were excluded from this study.
Functional evaluation
Complete general and neurological examinations were performed. Evaluation of the patient
function was done using Modified Medical Research Council for motor power evaluation,
Nurick score for evaluation of the myelopathy, and the functional disability was evaluated
using NDI [[Table 1]].
Table 1: Functional evaluation
Radiographic evaluation
Preoperatively, sub-axial Cervical sagittal alignment was assessed using C2–C7 Cobb's
angle, measured by formal Cobb methods in standard lateral standing radiographs and/or
CT with sagittal reconstruction, signs of instability were detected using dynamic
X-ray films [[Figure 1]]a. MRI was done for all patients for detection of radiological signs of spondylotic
myelopathy [[Figure 1]]b. Postoperatively, the follow-up radiographs were obtained (1–2 days), and at least
6 months after surgery, the cervical sagittal alignment was evaluated.
Figure 1: Male patient aged 59 years with cervical spondylotic myelopathy. (a) preoperative
plain lateral view neutral and dynamic X-rays (b) preoperative magnetic resonance
imaging had abnormal T2-weighted signal in the spinal cord at level of C3-C4 disc.
(c) intraoperative view showing laminectomy and fixation. (d) intraoperative fluoroscopic
view and postoperative upright lateral X-ray view
Results
Demographic data
Thirty patients who underwent posterior cervical laminectomy with lateral mass screw-rod
fixation were analyzed. There were 21 male and 9 female patients [[Table 2]]. The mean age was 57.7 years ± 6.20, ranges from 45 to 72 years [[Table 3]]. A total of 226 screws were placed. All the thirty patients showed neither intraoperative
vascular injury nor neural injury or dural tears. There were 2 lateral masses of 2
separate patients were skipped because of lateral mass fracture, a C4 lateral mass
in one patient and a C5 lateral mass of the other one. One screw violated the foramen
transversarium was revealed on the follow-up, but without penetration of the vertebral
artery, the patient was clinically free, and there was no need for further investigations
or redirection otherwise there were no vertebral artery injuries. For the specific
placement of pedicle screws at the C7 level, a total of 10 screws (n = 5 patients)
were placed in the C7 pedicles. Patients were followed up for at least 6 months; all
patients had standard lateral and anteroposterior plain X-ray radiographs at the end
of the follow-up. No neurological deterioration or instances of instrumentation failure
occurred, and no lucencies observed surrounding the screws and successful fusion was
documented in all patients. The great majority of change in sagittal alignment in
this study occurred between the preoperative and the immediate postoperative imaging
with little changes in the Cobb angle observed on late follow-up imaging compared
with the immediate postoperative imaging.
Table 2: Sex distribution
Table 3: Age distribution
Functional outcome
Motor power
Motor weakness was present in 26 patients of which 23 patients (88.5%) improved and
3 patients (11.5%) remained same, with no case of worsening. Weakness was absent in
4 patients (13.3%) of 30 patients whose power remained the same after surgery. Statistical
analysis proved that there is significant improvement of the motor power postoperatively
[[Table 4]].
Table 4: Comparison between pre- and post-operative motor power
Nurick's score scale
Twenty-seven patients (90%) showed improvement at least one grade up, while three
patients (10%) remained the same with no cases of worsening in their Nurick score.
Statistical analysis proved that there was significant improvement of the Nurick score
postoperatively [[Table 5]].
Table 5: Pre- and post-operative Nurick score
Neck disability index scale
According to the NDI outcome score, 27 (90%) patients showed improvement at least
one grade up. Preoperatively, there was one (3%) patient with complete disability,
eight (27.7%) patients with severe disability, 16 (53.3%) patients with moderate disability,
and five (16.7%) patients mild disability [[Table 6]].
Table 6: Mean for pre- and post-operative neck disability index-score
Changed postoperatively to be as follow, 9 patients (30%) had no disability, 13 patients
(43.3%) had mild disability, 6 patients (20%) with moderate disability, 8 patients
(27.7%) and 2 patients (6.7%) had severe disability. Statistical analysis proved that
there was significant improvement of the NDI postoperatively [[Table 7]].
Table 7: Pre- and post-operative neck disability index-score
Radiographic outcome
There were no incidences of instrumentation lucencies or hardware failures, with good
fusion seen in all patients during follow-up radiographs.
Cervical alignment
We chose negative angulation as indicative of lordosis and positive angulation to
represent kyphosis. For all patients, the mean preoperative Cobb angle (inferior border
of C2 to the inferior border of C7) was −8.51° ± 14.07°, ranged from −30° to + 17°
while the mean immediate postoperative Cobb angle was −10.49 ± 10.53 (ranging from
−30 to + 15) and the mean Cobb angle at the end of our follow-up was −10.29 ± 12.43.
There was no significant difference between the preoperative and the postoperative
mean Cobb angles for all patients [[Table 8]].
Table 8: Mean Cobb angle for 30 patients, preoperative, immediate postoperative, and follow-up
Patients were divided into 2 study groups according to their cervical sagittal alignment
as follows: the first group (Group 1) included patients that showed lordotic sagittal
alignment preoperatively the mean Cobb angle for this group was −16.55° ± 7.81° (ranged
from −30° to −5°). All 21 patients (100%) retained lordotic curve postoperatively
with the mean Cobb angle equals − 15.89 ± 6.02 immediately postoperative and −15.61
± 5.32 at the end of our follow-up. There were no significant difference between the
preoperative, the immediate postoperative and the late follow-up of this group.
The second group (Group 2) included nine (20%) patients that showed loss of the lordotic
sagittal alignment (kyphotic or straight curve) preoperatively with the mean Cobb
angle equals 9.89° ± 4.59° (ranged from −25 to −5). The mean postoperative curve of
this group was 2.67 ± 8.08 at the immediate postoperative and remained unchanged till
the end of the follow-up. There was a significant difference in the mean Cobb angle
for this group between preoperative and immediate postoperative and between the preoperative
and at the end of the follow-up. There was significant shift of this group toward
lordosis.
Postoperatively, 2 subgroups were reported:First, subgroup (subgroup A included 5
patients (13.66%) who had a preoperative kyphotic curve who remained kyphotic postoperatively
and retained kyphotic to the end of our follow-up. The mean preoperative Cobb angle
was 12.20 ± 3.96 changed to 9.22 ± 7.33 at the immediate postoperative and 9.0 ± 4.06
at the end of the follow-up. There was no significant difference in the mean Cobb
angle for this group between preoperative and immediate postoperative and between
the preoperative and at the end of the follow-up. This group showed a slight improvement
in the kyphosis or at least no deterioration of the cervical curve postlaminectomy.
The second subgroup (subgroup B) includes 4 patients (13.33%) who had a preoperative
kyphotic curve, and they shifted to have a lordotic curve postoperatively. The preoperative
mean Cobb angle for them was 7.0 ± 3.92, changed to −5.25 ± 1.25 at the immediate
postoperative and remained unchanged at the end of the follow-up. There was a significant
difference in the mean Cobb angle for this group between preoperative and immediate
postoperative and between the preoperative and at the end of the follow-up. There
was significant shift of this group toward lordosis. The total number of patients
with lordotic group postoperatively was 25 patients (86.66%) with lordotic sagittal
alignment; the mean Cobb angle for this group was −14.23° ± 6.02° in the immediate
postoperative period and −13.91 ± 7.46 in the end of our follow-up. The aim of our
study was not to access the amount of correction of the kyphotic deformity, but it
aimed for the assessment of the change in the Cobb angle pre- and post-operatively.
Group A (lordotic) indicating negative angle while the kyphotic group represent the
positive angle, and we studied the impact of the changes in the Cobb angle and the
functional outcome.
The relation between the preoperative duration of symptoms and the functional outcome
The duration of symptoms ranged from 4 to 30 months with a mean duration of 14.6 months.
To study the relation between the preoperative duration of symptoms and the functional
outcome, we divided the patients into 5 groups with a 6-month interval between each
group and the next one. We found that there was a significant correlation between
the preoperative duration of symptoms and the postoperative motor power improvement,
Nurick score, and the NDI score [[Table 9]], [[Table 10]], [[Table 11]].
Table 9: The relation between the duration of symptoms and the motor power
Table 10: The relation between the duration of symptoms and the neck disability index (NDI)
Table 11: The Relation between the duration of symptoms and the Nurick score
Relation between age and functional outcome
We divided the patients according to their ages into younger group <60 and older group
≥60 years, and we found that there was no correlation between the patient age and
their functional outcome [[Table 12]].
Table 12: Relation between age and functional outcome
The relation between the change in the cervical alignment and the functional outcome
Overall, we found that patients with preoperative cervical lordosis have a better
functional outcome than those who have kyphotic one [[Table 13]].
Table 13: The relation between the change in the cervical alignment and the functional outcome
for Group 1 and Group 2
In patients with maintained cervical lordosis (Group 1), there was significant improvement
in the functional outcome. In patients with preoperative kyphotic curve who remained
kyphotic postoperatively (Subgroup A), there was no significant difference between
the pre- and post-operative functional condition.
In patients with preoperative kyphotic curve who shifted toward lordosis postoperatively,
Subgroup B showed significant improvement in the functional outcome [[Table 14]].
Table 14: The relation between the change in the cervical alignment and the functional outcome
for subgroup A and subgroup B
Discussion
In our series, all patients presented with CSM were operated on for posterior cervical
laminectomy and lateral mass screw fixation and fusion [[Figure 1]]c, [[Figure 1]]d. We combined the lateral mass fixation to the laminectomy to improve outcome by
prevention of postlaminectomy kyphotic deformity. In a study conducted by Suk et al.[[5]] on 85 patients underwent laminoplasty and they found a significant loss of cervical
lordosis with 10% of previously lordotic patients converting to kyphosis.
However, in a recent study by Nurboja et al.,[[6]] 268 patients undergoing cervical laminectomy and laminoplasty were followed up
for an average of 6.7 years, and no significant change in sagittal alignment was noted.
Determination of the presence and extent of instability depends on meticulous assessment
of not only static and dynamic radiological imaging but also on the detection of clinical
signs of instability as Olson and Joder,[[7]] considered the presence of paraspinal muscle spasm, decreased cervical lordosis,
and pain with sustained posture are signs of instability.
McAllister et al.[[8]] concluded that posterior cervical fusion usually appropriate if patients showed
one or more of the following: the presence of preoperative instability, significant
axial neck pain, younger age, minimal lordosis or straightening of the cervical spine,
or if the patient will need a postoperative radiation. Epstein [[9]] demonstrated on his study that cervical instability was evaluated before surgery
using dynamic films and all patients with even loss of cervical lordosis operated
for posterior decompression and fixation as these patients had a great risk for postlaminectomy
kyphosis. As regard to complications, two lateral mass fractures occurred intraoperatively
in two cases in our series, none of them were at the extremity of the fixation construct,
and the fractured lateral masses were bypassed without screw insertion as shown in
[[Figure 2]].
Figure 2: Intraoperative views of laminectomy and lateral mass fixation showing fracture of
the left lateral mass of the 4th cervical vertebra and Figure 3: Showing violation
of the foramen transversarium without that level was skipped on the left side penetration
of the vertebral artery
Inoue et al.[[10]] reported 18 intraoperative lateral mass fractures of the 471 lateral masses; they
reinserted the screw with a different trajectory angle, it succeeded in four lateral
masses but reinsertion was impossible in the remaining cases, and the levels.
In our series, there were four facet violations. Heller et al.[[11]] reported that the risk of FV was higher in Roy–Camille technique (22.5%) than Magerl
technique (2.4%). While Barrey et al.[[12]] reported that facet violation occurred in 4 of 80 lateral mass screws (5.0%) with
the use of Magerl technique.
We did not observe any manifestations of nerve root injury or other neural injury
related to screws insertion. Postoperatively, there was no clinical evidence of vertebral
artery injury as all patients were observed for local neck hematomas, vertebrobasilar
stroke, and for any further neurological deterioration.
Al-Barbarawi et al.[[13]] reported that there was no active bleeding occurred neither intraoperative as a
result of vertebral artery injury nor postoperatively. Only one case had CSF leak
from the wound that treated successfully with reinforcement sutures and lumbar drain
for 3 days.
One screw showed violation of the cortex of the vertebral artery foramen opposite
C5 lateral mass without penetration of the vertebral artery [[Figure 3]]. Kim et al.[[14]] reported in a prospective study on the evaluation of 1256 lateral mass screws positioned
in 178 consecutive patients at their institution. One screw revealed in the follow-up
CT violating the foramen transversarium without penetrating the vertebral artery required
no further intervention.
Figure 3: Showing violation of the foramen transversarium without that level was skipped on
the left side penetration of the vertebral artery
In our study, all patients were thought to be stable, based on the absence of motion
on postoperative radiographs and the absence of hardware breakage or migration, coupled
with the maintenance of alignment at the end of the follow-up.
We agreed with Al-Barbarawi et al.,[[13]] based on the postoperative radiograph, no pseudoarthrosis was noted.
There was almost no increased morbidity due to added instrumentation. That was agreed
with the study conducted by McAllister et al.[[8]] they advocated combining cervical laminectomy with fusion, as laminectomy alone
can result in postoperative kyphosis and late deterioration. They found that cervical
fusions are associated with relatively low complication rates and as they say “the
morbidity of lateral mass screw/plate fixation is low.”
As regard to the functional outcome, we found that there is no ideal metric method
to judge the functional outcomes; with different minimum magnitude of clinical differences
can be detected for varying CSM severity categories.
In our study, we found significant improvement between the preoperative and the postoperative
Nurick score, with no correlation between the preoperative and the postoperative Nurick
score, in which 23 (88.5%) patients of 26 (with preoperative weakness) showed at least
one grade up while only 3 patients (11.5%) of 26 remained the same with significant
improvement.
Our opinion was supported by the study of Rajshekhar and Kumar,[[15]] who found clinical improvement in 100% of patients at follow-up of 17 patients
with Nurick scores of 5.
While Macdonald et al.[[16]] studied eight cases of CSM with Nurick scores of 5 undergoing surgery and found
50% clinical improvement (improvement of at least 1° in the Nurick scale), also Matsunaga
et al.[[17]] considered 31 patients with Nurick scores of 5 and reported improvement in only
16.2%.
On the other side, Holly et al.[[18]] found that there was a correlation between the preoperative and the postoperative
Nurick score, and had shown that patients with preoperative lower Nurick's grade had
a better outcome.
We found that functional outcome is better in younger patients than older one. We
think that may be due to the presence of several comorbidities such as atherosclerosis,
hypertension, and diabetes mellitus in older ages. Our results aggress with the study
conducted by Naderi et al.[[19]] concluded that there was no relation between surgical outcome and age of the patient.
On the other side Nakashima et al.[[20]] concluded that “surgical decompression results in superior functional status in
younger patients compared with elderly patients and confirms that the elderly are
less effective at translating neurological recovery into functional improvements.”
The duration of symptoms in our cases ranged from 4 to 30 months with the mean duration
of 12.9 months. In our study, we found that there was significant correlation between
the duration and the functional outcome. Our results agreed with Ebersold et al.[[21]] who studied 100 patients with CSM, they underwent surgical decompression and found
that poor neurological outcome was related to increased duration of preoperative symptoms
On the other side, a study published by Fehlings et al.[[22]] who concluded that there's no correlation between duration of symptoms and the
surgical outcome.
Suri et al.[[23]] prospectively evaluated 146 consecutive patients with CSM over a 2-year period.
Assessed patients clinically and with the Nurick grading system preoperatively, and
at 3 and 6 months' postoperatively, he found that patients with symptoms >2-year duration
of showed significantly less improvement in their postoperative Nurick score.
In the literature, We did not find standard values for “normal” curvature, several
arguments about the value of the normal cervical lordosis measurements including the
study conducted by Guo et al.,[[24]] who calculated normal cervical sagittal alignment, obtained by measuring the C2–C7
Cobb angle, on cervical lateral radiographs in the neutral position to be −12.7 ±
6.6 in female patients and −16.3 ± 7.3 in male patients in 414 asymptomatic volunteers.
Benzel,[[25]] defined an effective cervical lordosis “as the configuration where no dorsal component
of C3 to C7 crosses a line from the posterior caudal corner of C2 to an identical
point on C7.” Hamanishi and Tanaka,[[26]] considered a minimum lordosis of 10° required for adequate dorsal migration of
the spinal cord.
Yamazaki et al.[[27]] noted in a study conducted in patients presented with ossification of the posterior
longitudinal ligament that ventral compression after posterior decompression continued
when the lordosis was >10° and if the extent of ventral canal encroachment exceeded
7 mm. On the same topic, McAviney et al.[[28]] published a study, in which almost 300 cervical X-rays were examined after dividing
the patients into groups with and without cervical pain. The authors conclude that
they found a statistically significant association between cervical pain and lordosis
<20° and a “clinically normal” range for cervical lordosis of 31°–40°.
Kuntz et al.[[29]] accumulated data from multiple studies and in a total of 464 patients, found the
pooled mean and variance of cervical lordosis to be –17 ± 14 standard deviation.
Overall, we found that patients with preoperative cervical lordosis have a better
functional outcome than those who have kyphotic one. That was agreed with a study
conducted by Shamji et al.[[30]] as they studied the association between cervical spine alignments with neurologic
recovery and concluded that the majority of patients with CSM showed postoperative
neurologic improvement. However, patients with preoperative lordotic alignment exhibited
greater improvement than those with preoperative kyphotic alignment. In our study,
we found that patients with maintained cervical lordosis (Group 1) showed significant
improvement in the functional outcome with nonsignificant change in cervical lordosis
indicating that there is no correlation between the change in the cervical lordosis
and the functional outcome in patients with maintained lordotic curve, however, in
patients with a preoperative kyphotic curve who shifted toward lordosis postoperatively
(Subgroup B) showed significant improvement in the functional outcome than the patients
who remained kyphotic postoperatively (subgroup A). Our results agreed with Sielatycki
et al.[[31]] in a study of patients with maintained cervical lordosis who underwent posterior
decompression and lateral mass fixation, they found no correlations between cervical
sagittal alignment and postoperative outcomes. Neither baseline nor postoperative
myelopathy severity was associated with either the amount of SVA or degree of lordosis.
Also a study conducted by Roguski et al.,[[32]] they found that when baseline kyphosis is present, any increase in the sagittal
vertical axis and degree of kyphosis is associated with worse myelopathy and poor
outcomes.
There was no neurological deterioration detected of the functional grading at the
end of our follow-up, this proves the importance of cervical fixation using lateral
mass fixation in resolving the dynamic compressive factors, whose role as the main
etiology for the progressive neurological deterioration in long-term follow-up of
patients with CSM.
