Neurosurg Rev (2006) 29: 114–117
DOI 10.1007/s10143-005-0003-7
ORIGINA L ARTI CLE
Peter Spangenberg . Volker Coenen .
Joachim Michael Gilsbach . Veit Rohde
Virtual placement of posterior C1-C2 transarticular screw
fixation
Received: 26 April 2005 / Revised: 7 August 2005 / Accepted: 18 September 2005 / Published online: 1 November 2005
# Springer-Verlag 2005
Abstract We wanted to evaluate how often safe and ef-
fective posterior C1-C2 transarticular screw placement is
realizable when it is performed according to guidelines
given in the literature. In 50 adult patients, computerized
tomography scan data from C0 to C3 were transformed into
a 3D spine model. Virtually, bilateral screws were placed
from the medial third of the C2-C3 facet joint towards
the rim of the C1 anterior arc parallel to midline. Three
categories of virtual screw position were rated: optimal
(virtual screw inside the C2 pars interarticularis, transver-
sing the middle third of the atlantoaxial joint, and sparing
the vertebral artery canal), suboptimal (virtual screw
violating the C2 pars interarticularis, and/or transversing
the lower or upper third of the C1-C2 joint, and sparing
vertebral artery canal), and unacceptable (virtual screw
breaching the vertebral artery canal). Optimal placement
was seen in 74, suboptimal placement in 11, and unac-
ceptable locations in 15 sites. We conclude that due to the
variability of the anatomy of the upper cervical spine, op-
timal transarticular C1-C2 screw placement is not possible
in up to 26%, and even hazardous in up to 15%.
Keywords Atlantoaxial instability . Transarticular screw
fixation . 3D model . Virtual screw placement
Introduction
Since the introduction of atlantoaxial transarticular screw
fixation by Magerl and Seemann [14], this technique has
become the preferred method for fixation of C1-C2 in-
This paper was presented in part at the Jahrestagung der Deutschen
Gesellschaft für Neurochirurgie, May 25–28, 2003, Saarbrücken,
Germany
P. Spangenberg (*) . V. Coenen . J. M. Gilsbach . V. Rohde
Neurochirurgische Klinik, Universitätsklinik RWTH,
Aachen, Germany
e-mail: pspangenberg@ukaachen.de
Tel.: +49-241-8088481
Fax: +49-241-8080420
stability. It has proven to be effective [3, 10] and biome-
chanically superior to other methods [8, 11, 15, 16, 19, 23].
Unfortunately, dorsal transarticular C1-C2 screw fixation is
more risky than other C1-C2 fixation techniques. Risks
include hypoglossal nerve and vertebral artery (VA) injury
[6, 24]. Especially VA injury could be a serious, potentially
life-threatening complication. Today, the operative stan-
dard is set with a screw entry point at the medial third of the
C2-C3 facet joint and a screw trajectory parallel to midline
towards the upper rim of the C1 anterior arc.
We wanted to evaluate how often C1-C2 transarticular
screw placement following these accepted standards led
finally to optimal and less optimal screw positions and to
screw positions which endanger the VA.
Materials and methods
The study was approved by the local ethics committee.
Fifty consecutive adult patients (31 male, 19 female, age
20–68 years) admitted after head trauma without injury of
the cervical spine were enrolled into the study. A spiral CT
scan of 1-mm increments from C0 to C3 was obtained; this
imaging was part of the diagnostic routine. The data were
transferred into a workstation that generated standard two-
dimensional orthogonal (axial, sagittal, and coronal) scans
as well as a 3D reconstruction (Silicon graphics). A screw
with a diameter of 4 mm was placed virtually into the 3D
spine model on both sides. The virtual screw’s entry point
was set into the medial third of the C2-C3 facet joint, the
trajectory was placed parallel to midline, and the tip was
aimed for the upper rim of the C1 anterior arc (Fig. 1). The
obtained screw positions were assigned for one of the
following three categories: (1) optimal: screw completely
inside the C2 pars interarticularis, crossing the middle third
of the C1-C2 joint, not penetrating the canal of the VA in
the C2-vertebra, (2) suboptimal: screw not completely
inside the C2 pars interarticularis, and/or crossing the C1-
C2 joint in the lower or upper third and outside the VA
canal, (3) unacceptable: screw trajectory penetrates the VA
canal. If the virtual screw placement was not acceptable,
 115

Fig. 1 Coronal, sagittal, axial and 3D reconstruction with two 4-mm-thick outlines representing virtual screws parallel to midline in
standard position with optimal placement on both sides. The red dot represents the localization of the screw at the level of the visible section

we investigated if and how screw positioning had to be
changed to meet the safe screw position criteria. In the
unacceptable cases, we tested how the screw had to be
readjusted to avoid the VA canal.
Results
Standard parallel to midline screw placement with a screw
entry point at the medial third of the C2-C3 facet joint
turned out to be optimal in 74 sites of 100 C1-C2 joints
(74%). In 11 sites (11%) screw positioning was suboptimal
because of violating the C2 pars interarticularis in seven
and crossing the C1-C2 joint not inside the middle third in
four cases.
In 15 sites (15%) the virtual screw trajectory was un-
acceptable because of penetrating the VA canal (Fig. 2 and
Table 1). In those cases, the VA was high-riding or located
far medially in the C2 vertebra. In seven patients the VA
canal had an aberrant course unilaterally and in four pa-

Fig. 2 a Sagittal CT, and b 3D

reconstruction with a virtual
screw passing through the
vertebral artery canal
116
Table 1 Evaluation of virtual standard C1-C2 transarticular screw
position in 100 sites
Screw position Percentage (%)
Inside C2 pars interarticularis, middle 74
third of C1-C2 joint, outside of VA canal
Violating C2 pars interarticularis 7 individual variation consists of medializing the standard
Crossing C1-C2 joint outside middle third 4 trajectory by 10–20°. We strongly believe that it is almost
Crossing VA canal 15
tients bilaterally. In eight of these 15 sites, angulating the
screw’s trajectory by 10–20° medially allowed to position
the screw without risk for the VA.
Discussion
Posterior C1-C2 transarticular screw fixation has become
the preferred technique to achieve atlantoaxial arthrodesis
[2, 4, 21]. Clinical and cadaver studies have shown that C1-
C2 transarticular screw fixation provides significantly
greater biomechanical stability in axial rotation, lateral
bending, flexion and extension than conventional posterior
fusion [8, 16, 17]. Fusion rates of 96% [12] could be
obtained, which is significantly better than the 70–90% rate
obtained by other posterior atlantoaxial fusion techniques
[5, 7, 20].
However, this technique is surgically demanding. In
1991, Grob et al. [9] published a series of 161 patients who
underwent C1-C2 transarticular screw fixation with a
malposition rate of 15%. The risks associated with inac-
curately placed screws are substantial and include injury of
the VA, the spinal cord or cranial nerves, respectively [6,
13, 24]. The risk of VA injury increases in patients with
anomalies of the VA course in the C2 vertebra. According
to Madawi, such aberrant vertebral arteries can be seen in
20% of the patients [13].
Despite the high rate of VA anomalies, operation
guidelines were given with an entry point for the screw
at the medial third of the C2-C3 facet joint, a trajectory
parallel to the midline, and aiming towards the upper rim of
the anterior C1 arch. The data from the literature seem to
support the adherence to these surgical guidelines: Grob et
al. [9] in a series of more than 160 patients and Haid et al.
[10] in a series of 75 patients reported no VA injury.
However, the low rate of VA injury is possibly the result
of a bias. In Haid’s series, 66 patients had bilateral screws.
In nine patients (12%) only unilateral screws were placed,
because of anatomic limitations. Possibly, placing the
second screw in these nine patients would have resulted in
VA injury. This compares very favorable with the 15% rate
of unacceptable screw placements in our series, and with
the results of Paramore et al. [18]. These authors made a
computed tomography (CT) study of 94 patients, in which
18–23% had a high-riding VA groove on at least one side
precluding safe placement of transarticular screws. The
other reason for the reported low rates of VA injury might
be that the artery does not fill the osseous canal completely,
with the result, that a screw that crosses the canal does not
necessarily injure the artery.
In our study the individual variation of the standard
trajectory in eight cases with aberrant VA course allowed a
safe screw placement, which spares the VA canal. This
impossible to variate the trajectory exactly according to the
individual needs during surgery with only lateral fluoro-
scopic guidance. At least in those cases, the intraoperative
use of a navigation system should be considered. In a cadaver
study, Bloch et al. [1] reported 30 screws that were positioned
while using frameless stereotactic guidance. All those 30
screws were shown to be placed exactly as they were
planned. However, the actual intraoperative C1-C2 position
in patients with instability could differ significantly from that
visualized on the preoperative CT scan, which explains that
in the series of Weidner, image-guidance significantly
reduced but did not eliminate the risk of unintended screw
placement [22].
As we were positioning the screws in a virtual model
without control of an anatomical specimen, our study has
some limitations. However, we believe that our results may
be transferred to our clinical routine because the evaluation
of the screw positioning is a result of the individual ana-
tomy of the C1-C2 complex which should be represented
reliably by CT data.
Conclusion
From the results of our study, we conclude that there exists
no general trajectory for a transarticular screw of the C1-C2
articulation. Following the accepted operative guidelines
does not preclude suboptimal and even dangerous screw
placements. Due to the variable anatomy, the optimal screw
trajectory may be parallel to midline, angulated or even
may not exist. Preoperative study of fine-cut CT scans or—
if possible-virtual planning of screw trajectory in a 3D
model might help to avoid risky screw placements.
References
1. Bloch O, Holly LT, Park J, Obasi C, Kim K, Johnsom JP (2001)
Effect of frameless stereotaxy on the accuracy of C1-2 trans-
articular screw placement. J Neurosurg 95:74–79
2. Campanelli M, Kattner KA, Stroink A, Gupta K, West S (1999)
Posterior transarticular screw fixation in the treatment of
displaced type II odontoid fractures in the geriatric population:
review of seven cases. Surg Neurol 5:596–600
3. Coyne TJ, Fehlings MG, Wallace MC, Bernstein M, Tator CH
(1995) C1-C2 posterior cervical fusion: long-term evaluation of
results and efficacy. Neurosurgery 37:688–692
4. Dickmann CA, Foley KT, Sonntag VK, Smith MM (1995)
Cannulated screws for odontoid screw fixation and atlantoaxial
transarticular screw fixation. Technical note. J Neurosurg
83:1095–1100
5. Dickmann CA, Sonntag VK (1998) Posterior C1-C2 transartic-
ular screw fixation for atlantoaxial arthrodesis. Neurosurgery
43:275–280

6. Ebraheim NA, Misson JR, Xu R, Yeasting RA (2000) The
optimal transarticular C1-2 screw length and the location of the
hypoglossal nerve. Surg Neurol 53:208–210
7. Farey ID, Nadkarni S, Smith N (1999) Modified Gallie
technique versus transarticular screw fixation in C1-C2 fusion.
Clin Orthop 35:126–135
8. Grob D, Crisco JJ III, Panjabi MM, Wang P, Dvorak J (1992)
Biomechanical evaluation of four different posterior atlantoax-
ial fixation techniques. Spine 17:480–490
9. Grob D, Jeanneret B, Aebi M, Markwalder TM (1991) Atlanto-
axial fusion with transarticular screw fixation. J Bone Joint
Surg Br 73:972–976
10. Haid RW, Subach BR, McLaughin MR, Rodts GE Jr, Wahlig
JB Jr (2001) C1-C2 transarticular screw fixation for atlantoaxial
instability: a 6-year experience. Neurosurgery 49:65–70
11. Henriques T, Cunningham BW, Olerud C, Shimamoto N, Lee GA,
Larsson S (2000) Biomechanical comparison of five different
atlantoaxial posterior fixation techniques. Spine 25:2877–2883
12. Liang ML, Huang MC, Cheng H, Huang WC, Yen YS, Shao
KN, Huang CI, Shih YH, Lee LS (2004) Posterior transarticular
screw fixation for chronic atlanto-axial instability. J Clin
Neurosci 11:368–372
13. Madawi AA, Casey TH, Solanki GA, Tuite G, Veres R,
Crockard HA (1997) Radiological and anatomical evaluation of
the atlantoaxial transarticular screw fixation technique. J Neu-
rosurg 86:961–968
14. Magerl F, Seemann PS (1987) Stable posterior fusion of the
atlas and axis by transarticular screw fixation. In: Kehr P,
Weidner A (eds) Cervical spine I. Springer, Berlin Heidelberg
New York, pp 322–327
15. Mitchell TC, Sadasivan AL, Ogden AL, Mayeux RH, Mukherjee
DP, Albright JA (1999) Biomechanical study of atlantoaxial
arthrodesis: transarticular screw fixation versus modified Brooks
posterior wiring. J Orthop Trauma 13:483–489
117
16. Naderi S, Crawford NR, Song GS, Sonntag VK, Dickmann CA
(1998) Biomechanical comparison of C1-C2 posterior fixa-
tions. Cable, graft, and screw combinations. Spine 23:1946–
1955
17. Oda I, Abumi K, Sell LC, Haggerty CJ, Cunningham BW,
McAfee PC (1999) Biomechanical evaluation of five different
occipito-atlanto-axial fixation techniques. Spine 24:2377–2382
18. Paramore CG, Dickmann CA, Sonntag VKH (1996)The
anatomical suitability of the C1-2 complex for transarticular
screw fixation. J Neurosurg 85:221–224
19. Richter M, Schmidt R, Claes L, Puhl W, Wilke HJ (2002)
Posterior atlantoaxial fixation: biomechanical in vitro compar-
ison of six different techniques. Spine 27:1724–1732
20. Taggard DA, Kraut MA, Clark CR, Traynelis VC (2004) Case-
control study comparing the efficacy of surgical techniques for
C1-C2 arthrodesis. J Spinal Disord Tech 17:189–194
21. Wang J, Vokshoor A, Kim S, Elton S, Kosnik E, Bartkowski H
(1999) Pediatric atlantoaxial instability: management with
screw fixation. Pediatr Neurosurg 30:70–78
22. Weidner A, Wähler M, Chiu ST, Ulrich C (2000) Modification
of C1-C2 transarticular screw fixation by image-guided surgery.
Spine 25:2667–2674
23. Wilke H-J, Fischer K, Kugler A, Magerl F, Claes l, Worsdorfer
O (1992) In vitro investigations of internal fixation systems of
the upper cervical spine: stability of posterior atlanto-axial
fixation techniques. Eur Spine J 1:191–199
24. Wright NM, Lauryssen C (1998) 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.
J Neurosurg 88:634–664