J Neurosurg Spine 19:217–221, 2013

©AANS, 2013

Risk to the vertebral artery during C-2 translaminar screw

placement: a thin-cut computerized tomography
angiogram–based morphometric analysis
Clinical article

Ron I. Riesenburger, M.D.,1 G. Alexander Jones, M.D., 2 Marie Roguski, M.D.,1

and Ajit A. Krishnaney, M.D. 3
1
Department of Neurosurgery, Tufts Medical Center, Boston, Massachusetts; 2Hudson Valley Brain & Spine
Surgery, Suffern, New York; and 3Department of Neurosurgery, Center for Spine Health, Neurological
Institute, Cleveland Clinic, Cleveland, Ohio

Object. The goal of this study was to characterize the anatomy relevant to placement of crossing C-2 trans-
laminar screws, including morphometric data, and to evaluate the risk of violating the vertebral artery (VA) during
the screw placement. Placement of bilateral crossing C-2 translaminar screws has become an increasingly popular
method for dorsal C-2 instrumentation as it is felt to avoid the known risk of VA injury associated with C1–2 transar-
ticular screw fixation and C-1 lateral mass–C-2 pars screw fixation.
Methods. The source images from 50 CT angiograms of the neck obtained from October to November 2007 were
studied. Digital imaging software was used to measure lamina thickness and maximum screw length, perform angula-
tion of screw trajectories in the axial plane, and evaluate the potential for VA injury. In cases where the VA could be
injured, the distance between the maximal screw length and artery was measured. Logistic regression was performed
to evaluate lamina width, axial angle, and screw length for predicting the potential for VA injury.
Results. Mean lamina thickness, axial angle, and maximal screw length were determined for 100 laminae, and a
potential for VA injury was noted in 55 laminae. The anatomically defined ideal screw length was longer in laminae
with potential for VA injury than in laminae with no apparent risk (35.2 vs 33.6 mm, p = 0.0131). Only increasing op-
timal screw length was noted to be a statistically significant predictor of potential VA injury (p = 0.0159). The “buffer
zone” (the distance between an optimally placed screw and the VA) was 5.6 ± 1.9 mm (mean ± SD, range 1.8–11.4
mm). A screw limited to 28 mm in length appeared to be safe in all laminae studied.
Conclusions. Crossing C-2 translaminar screws have been reported to be safe and effective. In addition to mor-
phometric characteristics, the authors have found that screws placed in this trajectory could jeopardize the vertebral
arteries in the foramen transversarium or the C1–2 interval. A C-2 translaminar screw limited to 28 mm in length
appeared to be safe in all 100 screw trajectories studied in this series.
(http://thejns.org/doi/abs/10.3171/2013.5.SPINE12790)

Key Words      •      radiographic analysis      •      C-2 translaminar screw      •      cervical      •     

bilateral crossing screw      •      vertebral artery

S
everal options are available for posterior atlantoaxi-
al fixation, including wire and cable techniques, pos-
terior C1–2 transarticular screw fixation, pars screw
placement, C-2 pedicle screw placement, and C-2 trans­
laminar screw fixation.26 Screw techniques have largely
supplanted posterior wire and cable techniques be­­cause of
the superior rotational and translational stability of screws.4
In 1992, Jeanneret and Magerl reported the tech-
nique of posterior C1–2 transarticular screw fixation as
an option for reducible subluxations of C-1 on C-2 and
Abbreviations used in this paper: CTA = CT angiography; VA =
vertebral artery.
in patients in whom a decompressive laminectomy is
required.4,10 Posterior C1–2 transarticular screw fixation
provides excellent biomechanical stability but is unfor-
tunately not feasible in up to 20% of patients because
of a high-riding VA foramen in C-2 and an associated
small isthmus.11,14 Because of the above anatomical fac-
tors and the technical skill required in the placement of
C1–2 transarticular screws, the incidence of VA injury
is high and has been reported as 2.2% per transarticular
screw inserted and as 4.1% per operated patient.25 A tech-
nique involving placement of C-1 lateral mass screws and
C-2 pedicle/pars screws was initially described by Goel
in 19946,7 and later modified by Harms in 2001.8 Since

J Neurosurg: Spine / Volume 19 / August 2013 217


its original description, this technique has been shown to
have biomechanical stability similar to C1–2 transarticu-
lar screws, and several clinical studies have shown excel-
lent fusion rates, low morbidity, and a reduced risk of VA
injury.16 However, in a radiographic study, Yoshida et al.28
evaluated the risk of VA injury from C-2 pedicle screws
and found that placement of a C-2 pedicle screw would
place the VA at risk for injury in approximately 10% of
CT scans examined. This rate did not differ from the rate
found for transarticular screws. The true incidence of
transverse foramen violation with C-2 pedicle screws is
difficult to estimate. In a study with 28 patients, Stulik et
al.18 reported that 2 of 28 C-2 pedicle screws had violated
the transverse foramen. Several other studies have report-
ed a significantly lower incidence of transverse foramen
violation; however, this may represent underreporting as
many of theses studies did not include routine postopera-
tive CT scans to assess screw placement.1,3,6,13,19,22,27
The placement of bilateral crossing C-2 laminar
screws was initially described by Wright23,24 in 2004 and
has been purported to eliminate the risk of VA injury.
Figure 1 shows a representative CT image of the bilateral
crossing C-2 translaminar screw construct. Indications
for placement of C-2 translaminar screws include salvage
of failed C-2 pedicle screw placement and fixation in pa-
tients in whom the VA follows an anomalous course or in
whom the C-2 pedicle is very small. Several biomechani-
cal studies have shown this construct to perform similarly
to C-2 pedicle and C-2 pars screws in the presence of in-
tact atlantoaxial ligaments.5 Several clinical series have
been published that demonstrate the viability of bilateral
crossing C-2 laminar screws in atlantoaxial and subaxial
fixation constructs.5,9,12,15,17,21 No VA injuries were report-
ed in these series. Furthermore, to date, no reports exist
of a compromised C-2 foramen transversarium following
placement of C-2 translaminar screws.
A cadaveric study on this technique has been pub-
lished,20 and radiographic series have provided morpho-
metric data on the anatomy relevant to this technique.2
However, despite the variability of the VA course and dif-
ficulty in evaluating the true incidence of VA injury post-
Fig. 1.  Representative axial CT scan showing bilateral crossing trans­­
laminar screws in C-2.
218
R. I. Riesenburger et al.
operatively, no study has evaluated the risk of transverse
foramen violation and possible concomitant VA injury
associated with the placement of crossing bilateral C-2
translaminar screws. This study was carried out to fur-
ther define the basic radiographic and anatomical charac-
teristics of the C-2 lamina, to elucidate ideal screw trajec-
tories and lengths for this technique, and to characterize
the risk of violation of the VA.
Methods
Cleveland Clinic institutional review board approval
was obtained. A database of patients who had recently
undergone CT angiography (CTA) of the neck was com-
piled. The charts of 60 such consecutive patients were
reviewed. Ten patients were excluded because available
online imaging did not include CTA (n = 9) or because
of the presence of a Klippel-Feil syndrome that had dis-
torted the C-2 lamina (n = 1).
The remaining 50 patients were included in this
study. All studies were completed between October and
November 2007. The dates of the studies and age and sex
of the patients were obtained from the database. Images
were accessed via the Cleveland Clinic online medical
records system, and images were viewed with the online
image viewing software (Syngo Studio Web V20D, Sie-
mens AG Medical Solutions). Thin-cut CTA source imag-
es with intravenous contrast were reviewed by 3 authors
(R.R., G.A.J., and A.A.K.). The contrast and window lev-
els were set to optimize visualization of bony structures,
which also allowed visualization of patent VAs because
of the intravascular contrast.
The Syngo software contains a measuring tool that
was used to obtain linear measurements. A protractor tool
is also available, and this tool was used to obtain angular
measurements. To allow for head rotation of the patients
in the CT scanner gantry, a plumb line was dropped from
the midline point on the ventral body of C-2 and carried
to a point at the midline of the intersection of the C-2
lamina and spinous process. All angular measurements in
the axial plane were then measured from this reference.
The entry point of the screw was the junction of the
lamina and spinous process, and the simulated screw was
placed in the axial plane at an angle that would allow it
to traverse the lamina and enter the lateral mass of C-2.
Maximal screw length, as reported here, refers to the
greatest length of a screw that can be placed without vio-
lating the lateral cortex of the lamina or the lateral mass
of C-2 (Fig. 2).
In some patients, the plotted screw trajectory had the
potential to contact the VA, either in the foramen trans-
versarium or in the interval between C-1 and C-2. In cas-
es where the VA could theoretically be injured by a long
screw placed in the anatomically defined screw trajectory
described above, the lamina was deemed as being “at
risk” for VA injury, and the distance from the entry point
to the wall of the VA was measured. The difference be-
tween this distance and the maximal screw length is the
distance between the tip of a well-placed screw and the
VA. This distance is a measure of the margin of safety,
or “buffer zone,” between the maximal screw length and
the artery.
J Neurosurg: Spine / Volume 19 / August 2013
C-2 translaminar screw fixation
Fig. 2. Axial CT scan through C-2, demonstrating the technique
used for the measurements in this study. The long arrow extending
from the junction of the spinous process and the lamina to the C-2 lat­
eral mass represents the simulated screw trajectory. Its length is the
anatomically defined maximal screw length. The double arrow is the
distance between the tip of the screw and the transverse foramen/VA
(single arrow); this distance, defined as the “buffer zone,” was only mea­
sured for laminae with VAs that were at risk for injury. For angulation of
the screw, the angle (q) was measured between a line dropped from
the midpoint of the vertebral body to the spinous process and the line
of the screw trajectory.
The JMP9 statistical software package (SAS Insti-
tute) was used for statistical analysis. Student t-tests were
used to compare continuous variables. Logistic regres-
sion was used to evaluate for variables predictive of po-
tential VA injury.
Results
In total, 50 patients were included in this study (25
males and 25 females), and their mean age was 60.1 ± 17.3
years (range 55.2–65.0). The average thickness of the left
and right laminae, measured at the thinnest point in the
axial plane, was 6.5 mm (range 4.5–9.1) and 6.3 mm (range
4.2–8.7), respectively. The axial angle of the left and right
laminae as measured from the sagittal plane (as described
above) was 47.2° (range 39.4°–56.1°) and 46.8° (range
36.5°–56.2°), respectively. Maximal screw length on the
left and right sides was 34.9 mm (range 27.8–41.3 mm) and
34.1 mm (range 27.2–41.2), respectively (Table 1).
A potential for VA injury was noted in a total of 55
laminae. The anatomically defined maximal screw length
was longer in laminae with potential risk for VA injury
than in laminae with no apparent risk (35.2 mm vs 33.6
TABLE 1: Morphometric characteristics of laminae
Variable Left
no. of laminae 50
width of lamina (mm)* 6.5 (6.2–6.8)
angle (°)* 47.2 (46.3–48.1)
screw length (mm)* 34.9 (34.0–35.7)
*  Values are presented as the mean (95% CI).
J Neurosurg: Spine / Volume 19 / August 2013
mm, p = 0.0131). In addition, a statistically significant
difference was detected in axial angle between these 2
groups (45.9° and 48.4° for laminae with VAs at risk and
not at risk, respectively, p = 0.0007) (Table 2).
Logistic regression was performed to evaluate lamina
width, axial angle, and screw length for predicting the po-
tential for VA injury. Only the increase in optimal screw
length was noted to be a statistically significant predictor
of potential for VA injury (p = 0.0159). The probability of
risk for VA injury increases with increasing screw length
(Fig. 3).
The buffer zone (the mean distance between the tip
of an optimally placed screw and the VA) was 5.6 mm
(95% CI 5.1–6.1 mm). The buffer zone ranged from as
low as 1.8 mm to as high as 11.4 mm (Fig. 4).
We noted that a screw length of 28 mm or less did not
violate the VA in any C-2 lamina studied in this series.
Discussion
The C-2 translaminar screw technique has become
increasingly popular in recent years as an alternative to
C-2 pars and C-2 pedicle screw fixation. Although VA
injury as a result of C-2 translaminar screw placement has
not been reported in the literature, its incidence may be
unrecognized due to the difficulty in identifying unilat-
eral VA injury postoperatively. The goal of this study was
to further define the anatomy relevant to the placement of
C-2 translaminar screws and to characterize the risk of
transverse foramen violation.
Our findings indicate that all patients studied had suf-
ficiently thick laminae to tolerate placement of 3.5-mm
diameter translaminar screws, and most could tolerate a
4.0-mm diameter screw. Although the axial angle differed
between lamina at risk and lamina not at risk for VA injury,
the small difference of 2.5° is not likely to be clinically
relevant. To a first approximation, using the junction of
the spinous process and lamina as an entry point, an entry
angle of 45° to the sagittal plane is suitable for this tech-
nique. However, this angle should be adjusted according to
the observed trajectory of the lamina at the time of screw
insertion. This finding is similar to that reported by Wang
in his cadaveric analysis of the technique20 and to that re-
ported by Bhatnagar et al. in a radiographic review.2 In
addition, our finding is consistent with the reported screw
trajectories published in clinical series.
Medial cortex violation is a potential complication of
translaminar screw placement. The average thickness of
the lamina at its thinnest point was approximately 6 mm.
The thinnest lamina measured was 4.2 mm thick. These
Right All
50 100
6.3 (6.0–6.6) 6.4 (6.2–6.6)
46.8 (45.6–48.0) 47.0 (46.3–4.8)
34.1 (33.0–35.1) 34.5 (33.8–35.1)
219

TABLE 2: Characteristics of laminae grouped by risk of transverse foramen violation*
Variable Not at Risk
no. of laminae 45
width of lamina (mm)† 6.5 (6.2–6.9)
angle (°)† 48.4 (47.3–49.4)
screw length (mm)† 33.6 (32.6–34.5)
buffer zone (mm)† NA
*  NA = not applicable.
†  Values are presented as the mean (95% CI).
measurements would suggest that the risk of medial cor­
tex breach when placing a standard 3.5-mm cervical
screw in the appropriate trajectory is very small. Clearly,
the preoperative imaging should be reviewed by the sur-
geon prior to instrumentation to ensure adequate laminar
thickness for screw placement.
The majority (32/50) of virtual screws placed on the
left and 23/50 virtual screws placed on the right could vi-
olate the VA, either in the foramen transversarium or the
interval between C-1 and C-2, if a long screw was placed.
Whether this would lead to dissection, pseudoaneurysm
formation, occlusion of the VA, or be clinically silent
cannot be determined from this study and would likely
depend on individual patient characteristics and the de-
gree of VA violation. Of the laminae in which VA injury
was possible, the mean buffer zone between the maximal
screw length and the VA is 5.6 mm. While this provides
an additional margin of safety in the average lamina, the
buffer zone was only 1.8 mm in one of the studied laminae
in this series. Although VA injury as a result of C-2 trans-
laminar screw placement has not been reported in the lit-
erature, our results indicate that VA injury is possible dur-
ing placement of C-2 translaminar screws. Consequently,
we recommend that surgeons carefully review preopera-
tive images to determine the optimal screw length, assess
for potential for VA injury, and exercise great care during
screw placement, especially when longer screws are to be
placed. Although not routinely used, 3D imaging of the
C-2 may be useful in determining ideal entry points and
trajectories for translaminar screw placement, especially
in patients with aberrant anatomy or cervical deformity.
A screw length of 28 mm did not violate the VA in
any of the patients studied. This information may be use-
Fig. 3.  Left: Distribution of screw length stratified by whether the VA
is at risk for injury.  Right: Probability curve of the VA being at risk for
injury depending on screw length (p = 0.0159).
220
R. I. Riesenburger et al.
At Risk p Value
55
6.3 (5.9–6.6) 0.254
45.9 (44.9–46.8) 0.0007
35.2 (34.3–36.1) 0.0131
5.6 (5.1–6.1)
ful in the rare situation in which preoperative imaging
studies are suboptimal and the length of the C-2 lamina
cannot be properly measured. However, this study is un-
derpowered to eliminate risk of VA injury and there may
be rare patients in whom a screw of less than 28 mm may
result in VA injury.
Interestingly, although the average lamina measures
more than 34 mm, of the 103 C-2 translaminar screws
placed by Dorward and Wright, 5 the mean screw length
was 28.9 mm (range 16–34 mm). Other clinical series did
not publish the mean length of screws placed in their clin-
ical centers; however, if the series by Dorward and Wright
is representative of clinical practice, it may be possible to
improve the biomechanical strength of this construct by
placing longer screws in select patients whose VAs are
not at risk.
Conclusions
Translaminar screw fixation continues to represent a
safe alternative to other C-2 fixation techniques; however,
the risk of VA injury should be understood by any sur-
geon performing this procedure. Careful review of pre-
operative images with attention to the position of the VA
in relation to the C-2 lamina is highly recommended. In
this study, a C-2 translaminar screw length of 28 mm ap-
peared to be safe in all 100 laminae.
Fig. 4.  Distribution of the buffer zone among laminae shown as a
box plot (top of the graph) and as a column graph. Although the buffer
zone in the vast majority of laminae at risk for VA injury was between 4
and 7 mm long, some laminae had very short buffer zones, indicating a
small margin for error.
J Neurosurg: Spine / Volume 19 / August 2013
C-2 translaminar screw fixation
Disclosure
The authors report no conflict of interest concerning the mate-
rials or methods used in this study or the findings specified in this
paper.
Author contributions to the study and manuscript preparation
include the following. Conception and design: Krishnaney, Jones.
Acquisition of data: Riesenburger, Jones. Analysis and interpreta-
tion of data: Riesenburger, Jones. Drafting the article: Riesenburger,
Roguski. Critically revising the article: all authors. Reviewed sub-
mitted version of manuscript: all authors. Approved the final ver-
sion of the manuscript on behalf of all authors: Krishnaney. Study
supervision: Krishnaney.
References
  1.  Aryan HE, Newman CB, Nottmeier EW, Acosta FL Jr, Wang
VY, Ames CP: Stabilization of the atlantoaxial complex via
C-1 lateral mass and C-2 pedicle screw fixation in a multi-
center clinical experience in 102 patients: modification of the
Harms and Goel techniques. J Neurosurg Spine 8:222–229,
2008
  2.  Bhatnagar R, Yu WD, Bergin PF, Matteini LE, O’Brien JR:
The anatomic suitability of the C2 vertebra for intralaminar
and pedicular fixation: a computed tomography study. Spine
J 10:896–899, 2010
  3.  Chen JF, Wu CT, Lee SC, Lee ST: Posterior atlantoaxial trans-
pedicular screw and plate fixation. Technical note. J Neuro-
surg Spine 2:386–392, 2005
  4.  Coyne TJ, Fehlings MG, Wallace MC, Bernstein M, Tator CH:
C1-C2 posterior cervical fusion: long-term evaluation of re-
sults and efficacy. Neurosurgery 37:688–693, 1995
  5.  Dorward IG, Wright NM: Seven years of experience with C2
translaminar screw fixation: clinical series and review of the
literature. Neurosurgery 68:1491–1499, 2011
  6.  Goel A, Desai KI, Muzumdar DP: Atlantoaxial fixation using
plate and screw method: a report of 160 treated patients. Neu-
rosurgery 51:1351–1357, 2002
  7.  Goel A, Laheri V: Plate and screw fixation for atlanto-axial
subluxation. Acta Neurochir (Wien) 129:47–53, 1994
  8.  Harms J, Melcher RP: Posterior C1-C2 fusion with polyaxial
screw and rod fixation. Spine (Phila Pa 1976) 26:2467–2471,
2001
  9.  Hong JT, Yi JS, Kim JT, Ji C, Ryu KS, Park CK: Clinical and
radiologic outcome of laminar screw at C2 and C7 for poste-
rior instrumentation—review of 25 cases and comparison of
C2 and C7 intralaminar screw fixation. World Neurosurg 73:
112–118, 2010
10.  Jeanneret B, Magerl F: Primary posterior fusion C1/2 in odon-
toid fractures: indications, technique, and results of transar-
ticular screw fixation. J Spinal Disord 5:464–475, 1992
11.  Madawi AA, Casey ATH, Solanki GA, Tuite G, Veres R,
Crockard HA: Radiological and anatomical evaluation of the
atlantoaxial transarticular screw fixation technique. J Neuro-
surg 86:961–968, 1997
12.  Meyer D, Meyer F, Kretschmer T, Börm W: Translaminar
screws of the axis—an alternative technique for rigid screw
fixation in upper cervical spine instability. Neurosurg Rev
35:255–261, 2012
13.  Ondra SL, Marzouk S, Ganju A, Morrison T, Koski T: Safety
and efficacy of C2 pedicle screws placed with anatomic and
lateral C-arm guidance. Spine (Phila Pa 1976) 31:E263–E267,
2006
J Neurosurg: Spine / Volume 19 / August 2013
14.  Paramore CG, Dickman CA, Sonntag VKH: The anatomical
suitability of the C1-2 complex for transarticular screw fixa-
tion. J Neurosurg 85:221–224, 1996
15.  Parker SL, McGirt MJ, Garces-Ambrossi GL, Mehta VA, Sci­
ubba DM, Witham TF, et al: Translaminar versus pedicle screw
fixation of C2: comparison of surgical morbidity and accu-
racy of 313 consecutive screws. Neurosurgery 64 (Suppl 2):
343–348, 2009
16.  Ringel F, Reinke A, Stüer C, Meyer B, Stoffel M: Posterior
C1-2 fusion with C1 lateral mass and C2 isthmic screws: accu-
racy of screw position, alignment and patient outcome. Acta
Neurochir (Wien) 154:305–312, 2012
17.  Sciubba DM, Noggle JC, Vellimana AK, Conmay JE, Kretzer
RM, Long DM, et al: Laminar screw fixation of the axis. J
Neurosurg Spine 8:327–334, 2008
18.  Stulik J, Vyskocil T, Sebesta P, Kryl J: Atlantoaxial fixation
using the polyaxial screw-rod system. Eur Spine J 16:479–
484, 2007
19.  Vilela MD, Jermani C, Braga BP: C1 lateral mass screws for
posterior segmental stabilization of the upper cervical spine
and a new method of three-point rigid fixation of the C1-C2
complex. Arq Neuropsiquiatr 64:762–767, 2006
20.  Wang MY: C2 crossing laminar screws: cadaveric morphomet-
ric analysis. Neurosurgery 59 (1 Suppl 1):ONS84–ONS88,
2006
21.  Wang MY: Cervical crossing laminar screws: early clinical
results and complications. Neurosurgery 61 (5 Suppl 2):311–
315, 2007
22.  Wang S, Wang C, Wood K, Yan M, Zhou H: Radiographic
evaluation of the technique for C1 lateral mass and C2 pedicle
screw fixation in three hundred nineteen cases. Spine (Phila
Pa 1976) 36:3–8, 2011
23.  Wright NM: Posterior C2 fixation using bilateral, crossing C2
laminar screws: case series and technical note. J Spinal Dis-
ord Tech 17:158–162, 2004
24.  Wright NM: Translaminar rigid screw fixation of the axis.
Technical note. J Neurosurg Spine 3:409–414, 2005
25.  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 nerve. J
Neurosurg 88:634–640, 1998
26.  Yanni DS, Perin NI: Fixation of the axis. Neurosurgery 66 (3
Suppl):147–152, 2010
27.  Yeom JS, Buchowski JM, Park KW, Chang BS, Lee CK, Riew
KD: Undetected vertebral artery groove and foramen viola-
tions during C1 lateral mass and C2 pedicle screw placement.
Spine (Phila Pa 1976) 33:E942–E949, 2008
28.  Yoshida M, Neo M, Fujibayashi S, Nakamura T: Compari-
son of the anatomical risk for vertebral artery injury associ-
ated with the C2-pedicle screw and atlantoaxial transarticular
screw. Spine (Phila Pa 1976) 31:E513–E517, 2006
Manuscript submitted August 16, 2012.
Accepted May 7, 2013.
Please include this information when citing this paper: pub-
lished online June 7, 2013; DOI: 10.3171/2013.5.SPINE12790.
Address correspondence to: Ajit Krishnaney, M.D., Department
of Neurosurgery, Neurological Institute, Cleveland Clinic, 9500
Euclid Ave., S-40, Cleveland, OH 44195. email: krishna@ccf.org.
221