Surg RadioI Anat (1994) 16 : 199-204
Surgical Radiologm Anatomy Journal of Clinical Anatomy
© Springer-Verlag 1994
Three-dimensional CT analysis of atlantoaxial rotation: results in the normal subject
J-L Dumas 1,2, P Thoreux 3, P Attali 2, D Gotdlust 2 and JP Chevrel t 1 Laboratoire d'Anatomie, UER Bobigny, Universitd Paris XIII, rue Marcel Cachin, F-93000 Bobigny, France 2 Service de Radiologie, 3 Service de Chirurgie Orthop~dique, H~pital Avicenne, 125, route de Stalingrad, F-93000 Bobigny, France
Summary. A three-dimensional analysis of atlantoaxial rotation was made by computed tomography in ten healthy subjects. The method described allowed study of spatial relations between C1 and C2 interspace under both static and dynamic conditions, using a projection in a fixed transverse plane. The results demonstrated an axis of rotation for C2 slightly anterior to that for C I. Asymmetric rotation was observed, the separation between the transverse foramina of Cl and C2 being less marked on the side opposite to the direction of rotation of the head. This use of CT and these anatomic findings may be useful in the study of disorders of atlantoaxial rotation and their effect on the vertebral arteries.
Analyse tridimensionnelle de la rotation atloido-axoidienne : methode tomodensitometrique et rdsultats chez le sujet normal Resume, Une analyse tridimensionhelle de la rotation atlo'/do-axo'idienne est effectuee en tomodensitometrie chez dix sujets sains. La methode
Correspondence to : J-L Dumas
presentde permet une etude des rapports dans l'espace de C1 et C2, en position statique et dynamique, pax projection dans un plan transversal fixe. Les resultats mettent en evidence un axe de rotation de C2 legbrement anterieur ~ celui de C1. Une rotation asymetrique est observee avec une ouverture entre les foramens transversaires de C1 et C2 plus faible du cetd controlateral au sens de rotation de ta t~te. Cette mdthode tomodensitometrique et ces r6sultats anatomiques peuvent ~tre utilises dans l'etude de la pathologie de la rotation atlo'/do-axo~'dienne et de son retentissement sur les arteres vertebrales.
Helical CT is a recent technique which provides three-dimensional reconstructions of excellent quality [6]. This technique is now regarded as standard for study of the bony s~uctures [9]. The three-dimensional bony reconstructions obtained by using static and dynamic helical CT may be expected to improve understanding of the anatomy and pathology of atlantoaxial rotation. This report describes our initial experience with the use of this technique in normal subjects and discusses its use in the demonstration of spatial relations between the atlas (C1) and axis (C2) during rotation of the base of the skull (CO).
Material and methods Key words: Computed tomography (CT), three-dimensional - - Atlas and Axis - - Spine, anatomy - - Spine, CT
Anatomic assessment of rotation of the craniocervical junction has benefited from the use of axial computed t o m o g r a p h y (CT), which allows investigation of the transverse plane of rotation [1, 3, 11]. Dynamic 2-D CT studies with rotation of the head of the patient have also been used to study abnormalities of rotation of the upper cervical spine [2, 5, 7, 10, 12].
Ten volunteer subjects (7 men and 3 women) aged between 25 and 58 years (mean 3 7 . t ) were studied. None of the subjects had a history of trauma or disease of the cervical spine. In every subject a lateral radiograph of the craniocervical junction was made initially to determine its neutral position. The subjects were standing and facing horizontally forward with the head in the most natural position. The system used was a scanner with slip-ring technology (a standard,
200
J-L Dumas et al : Three-dimensionalCT analysis of atlantoaxial rotation
Fig. la, b a Lateral radiograph of the upper cervical spine of a 33-year-old woman with the subject's head in the neutral position. The posterior walt of the dens is vertically orientated (dotted line), b Same subject, lateral topogram from the static CT scan when the neutral position of the craniocervical junction is almost reproduced. By comparison, there was a slight anterior incIination of the posterior wall of the dens (clotted line) with respect to the vertical plane. The horizontal lines indicate the spiral scanning volume selected
a Radiographie de profil de la colonne cervicale sup~rieurechez une femme de 33 ans avec la tete en position neutre, l,e mur post6rieur de la dent est orient6 verticalement (ligne pointillde), b M6me st@t, topogramme de profil de l'examen tomodensitomdtriquestafique quand la position neutre de la charni&e crRnio-cervicaleest presque reproduite, En comparaison, une 16ggreinclinaison ant~rieure du mur post~rieurde la dent (ligne pointilI~e) par rapport au plan vertical est observ~e. Les lignes horizontales indiquent le volume de l'acquisition hdlico~dales61ectionn6e
c o n t i n u o u s l y rotating CT scanner: Xpress; Toshiba, Nasu, Japan). The ten subjects were studied in the supine position. The program included an initial examination in the static position (defined by the vertical position of the nasal septum) followed by a second examination in dynamic positioning. The static CT study began with a lateral topogram including the skull b a s e and the c e r v i c a l spine. This topogram was used to adjust the position of the craniocervical junction in the s a g i t t a l p l a n e so as to reproduce its neutral position as prev i o u s l y d e t e r m i n e d (Fig. 1). T h e helical scanning v o l u m e was programmed to cover the volume included between the foramen m a g n u m and the intervertebral space at C2-C3 in an a x i a l p l a n e . T h e n a s e c o n d a c q u i s i t i o n was m a d e at the same level in the d y n a m i c position, the subjects b e i n g a s k e d to turn their head to the left in a completely natural manner, without excessive rotat i o n or s t r a i n i n g . T h e h e a d w a s i m m o b i l i s e d b y a f o r e h e a d band.
The helical CT was made at 120 kV, 150 mAs,with a table feed speed of 1.7 ram/s, a collimation of 5 mm and a continuous exposure time of 30 s. This yielded a 3-D volume acquisition of 4.5 cm. In this program, the dose of skin irradiation was approximately 3.3 cGy per acquisition. The images were reconstructed withl m m increments.The routine Toshiba software program for reconstruction was used to g e n e r a t e the 3-D i m a g e s . Reconstruction of these images required obliteration of the adjacent o c c i p i t a l b o n y structures. The CT scan t o o k 5 to 10 rain, and i m a g e processing occupied 20 to 40 min. The reconstructed images obtained were subsequently analysed in different planes of space to determine the most informative views for study of the C0-C1-C2 complex. The p r o g r a m of analysis of the images c o m p r i s e d the f o l l o w i n g s t a g e s . In the first stage, the static examination was analysed. The ventral view a l l o w e d m e a s u r e m e n t o f the starting tilt o f the transverse plane
of C2 in relation to the horizontal (Fig. 2a). The cranial view allowed for correction of the sagittal angulation o f C2 so as to r e p r o d u c e its neutral sagittal position as measured on the lateral radiograph of the cervical spine (Fig. 2b). The most useful parameter was the position of the posterior border of the dens in relation to the vertical plane. This sagittally projected angle was transferred to the caudal view (Fig. 2c). In the dynamic studies, the first stage was to measure the angle o f rotation of CO, C1 and C2 in the original axial 2-D images [1]. (Lines were drawn through the anteroposterior axes of CO, C1 and C2. The axis o f CO was defined by projection o f the line j o i n i n g the axis o f the n a s a l s e p t u m to t h e i n t e r n a l occipital protuberance. The axes of C1 and C2 were defined by projecting the line j o i n i n g the middle of the posterior border o f the vertebral body to the m i d d l e o f the spinous process; it was perpendicular to the plane of the posterior border of the
J-L Dumas et al : Three-dimensional CF analysis of atlantoaxial rotation
201
Fig. 2a-c 3-D CT reconstruction of the C0-C 1-C2 joint complex in the same subject, with the head in the static position, a Ventral and slightly inferior projection (anterio-posterior view). C2 is slightly tilted towards the left; the angle of inclination with respect to the horizontal plane is 4 ° . The line for measurement is drawn through symmetrical landmarks (foramina, pedicles, etc.) and connected with a parallel projection of the lower border of the film. b Cranial projection (view from above). The posterior wall of the dens is vertically positioned to reproduce its neutral orientation. The angle of correction is 1° in the sagittal plane (c~ = 1). e Caudal projection (view from below). The previously determined sagittal angle of projection has been transferred ({~ = 179), and the tilt of C2 with respect to the horizontal plane has been corrected ([I = 4). The transverse foramina of C2 lie slightly anterior to those of C1. Lines for measurement are drawn through the centers of the transverse foramina Reconstruction TDM 3-D da complexe C0-C1-C2 chez le mSme sujet, avec la t~te en position statique, a Projection ventrale et 16g~rement inf&ieure (rue antgro-postdrieure). C2 est 16gtrement inclinte vers la gauche ; l'angte d'inclinaison est de 4 ° par rapport ~tl'horizontal. La ligne de mesure est tracde par des rep~es symttriques (foramens, pddicules, etc.) et rapportte ~t une t~ojection parall~le au bord inf6fieur du film. b Projection crfimale (vue cr6niate). Le tour postdrieur de la dent est positiorm6 verticalement pour reproduire son orientation en position neatre. L'angle de correction est de 1°dans le plan sagittal (c~= 1). c tSojection caudale (rue caudale). L ' a @ e sagiax~ de projection prdalablement ddtermin6 a 6t6 report6 (c¢ = 179) et l'inctinaison de C2 par rapport au plan horizontal a dt6 corrigte (9 = 4). Les foramens transversaires de C2 sont situts 16gbrement en avant de ceux de C 1. Les lignes de mesure sont tractes par Ie centre des foramens transversaires
vertebral body. The angles formed
r o t a t i o n o f CO, C 1 a n d C 2 in r e l a -
tral v i e w w e r e o b t a i n e d a f t e r c o r r e c -
by these lines and the vertical border of the films were determined.
t i o n to t h e sagittal p l a n e . ) In the dynamic studies,
r o t a t i o n (Fig. 3a). T h e c r a n i a l v i e w
These angles indicated the degree of
r e c o n s t r u c t i o n s o f C1 a n d C 2 in v e n -
ting their respective angles of axial the
was analysed
with the transverse
202
J-L Dumas et al : Three-dimensional CT analysis of atlantoaxial rotation
Fig. 3a-c 3-D CT reconstruction of the C0-CI-C2 joint complex in the same subject, when her head is rotated to the left. The axial rotations are: CO: 48 °, CI: 47 °, C2:17 °. a Ventral and slightly inferior projection (anterio-posterior view) after correction of the axial rotation of C2 (y = -163). The tilt of C2 is 11°o it is slightly greater than the tilt of C1 (9°; measured when y = -133, not shown). b Cranial projection (view from above) when C2 is horizontally ([3 = -11) and sagittally (y = 17) orientated, The posterior wall of the dens is vertically orientated to reproduce its neutral position. The angle of correction is -2 ° in the sagittal plane (¢x -- -2). c Caudal projection (view from below). The previously determined sagittal angle of projection has been transfered ((z = -178). The initial tilt of C2 has been corrected ([3 = 4) with respect to the horizontal plane; in this condition the transverse plane of projection is the starting plane of C2. The distance between the right transverse foramina of CI and C2 (short arrow) is smaller than that between those on the left (long arrow). Reconstruction TDM 3-D du complexe C0-C1-C2 chez le mEme sujet quand la rate est tournre vers la gauche. Les rotations axiales sont ; CO : 48 °, C1 : 47 °, C2 : t7 °. a Projection ventrale et Rgrrement inf~rieure (rue antdro-postdrieure) aprbs correction de la rotation axiale de C2 (y = -163). L'inclinaison de C2 est de 11 °, elle est 16g~rement sup~rieure ~t celle de C1 (9 ° ; mesur6e pour 5' = -133, non montrr), b Projection crfiniale (rue crdniale) quand C2 est orientre horizontalement (~ = -11) et sagittalement (Y = 17). Le tour postrrieur de la dent est orient6 verticalement pour reproduire sa position neutre. L'angle de correction est de -2 ° darts le plan sagittal (e~ = -2). e Projection caudale (rue caudale). L'angle sagittal de projection prraIablement d~tennin6 a 6t6 report6 (tz = -178). L'inclinaison de d~part de C2 par rapport ~tl'horizontate a 6t6 corrigr, e ([3 = 4) ; dans cette condition, le plan transversal de projection correspond au plan de d~part de C2. La distance entre les foramens transversaires droits de C1 et C2 (¢Rche courte) est plus petite que ta distance entre les foramens transversaires ganches (y'l~chelongue)
of the dens was
t h e c a u d a l v i e w w a s t h e initial trans-
determined (Fig. 3b). This sagittal projection angle was transferred on
v e r s e p l a n e o f C 2 in n e u t r a l p o s i t i o n , It w a s o b t a i n e d b y a d d i n g t h e a n g l e
sagittal a n g l e o f p r o j e c t i o n r e q u i r e d
to t h e c a u d a l v i e w . T h e h o r i z o n t a l
r e q u i r e d to c o r r e c t t h e initial a n g u l a -
to reproduce
plane of projection determined
t i o n o f C 2 (Fig. 3c).
plane of C2 orientated horizontally and the medial
plane
orientated
sagittally. In this cranial view, the the starting
neutral
sagittal position
for
J-L Dumas et al : Three-dimensional CT analysis of atlantoaxial rotation
(
203
Table 1. Results of measurements in ten normal subjects with head rotated to left R~sultats des valeurs mesur~es chez dix sujets t~moins avec la t~te tourn6e vers la gauche %
=,~--,,,B~=_. .
. . . . .
Subject
Sex
Age
Rotation of C0 a
Angulation C1-C2 a
Difference between dR and dLb (dR _
t 2 3 4 5 6 7 8 9 10
M M M F M M F M F F
34 45 55 31 58 31 25 29 34 29
48 ° 43 ° 51 ° 50 ° 49 ° 53 ° 50 ° 54 ° 48 ° 47 °
30 ° 33 ° 38 ° 31 ° 29 ° 25 ° 38 ° 32° 30° 31 °
16.6% 16.6% 15.3% 20% 16.6% 41.6% 16.6% 18.I% 27.2% 33.3%
L1
~'~ '~.~L = dR + 2 7 %
Fig. 4 Diagram showing the rotationaJ parameters on the 3-D CT reconsm~ction in candaI projection (view from below, as shown in Figs. 2c, 3c, same subject) in the static position (thin lines) and after rotation of the head to the left (heavy tines). L1, line joining the centers of the transverse foramina of C1 (dotted lines); L2, line joining the centers of the transverse foramina of C2 (solid lines); dR, distance between the right foramina of C1 and C2; dL, distance between the left tbrmnens of C1 and C2.47 ° = angle of C1 rotation, The observed transverse projection of the entire motion of the C1-C2 complex indicates that the rotational centers of C 1 and C2 lie in the middle of L1 and L2 respectively. In the static position L2 lies in front of L1, This results in an asymmetric rotation; dR is approximately 27% shorter than dL
Schdma mettant en 6vidence les pmam~tres de la rotation sur la reconstruction TDM 3-D en projection caudale (rue caudate, comme cela est itlustrg par les figs. 2c, 3c, chez le m~me sujet). En position statique (traits fins) et aprbs rotation de la tote vers la gauche (traits gras). L1, ligne joignant le centre des foramens transversaires de C1 (ligne pointiIl~e) ; L2, Iigne joignant le centre des foramens transversaires de C2 (trait plein) ; dR, distance entre les foramens droits de C1 et C2 ;dL, distance entre les foramens gauche de C1 et C2.47 ° = angle de rotation de C1. La projection tranversale observ6e de l'ensemble du mouvement de C1 - C2 indique que le centre de rotation de C1 et de C2 est situ6 respectivement jusle au milieu de L1 et de L2. En position statique, L2 est en position ventrale par rapport h de L1. [1 en r6sulte une rotation asym6trique : dR est plus courte que dL de 27 %
Results
T h e e n t i r e v o l u m e o f the C 0 - C 1 - C 2 c o m p l e x w a s i m a g e d in e v e r y s u b ject and good quality 3-D reconstructions were obtained. These clearly demonstrated the anatomy of
a Measurements made in 2-D CT scans b Distances between transverse foramina of C1 and C2 on right (dR) and left (dL)
a Mesures effectu6es sur les examens TDM 2-D b Distances entre les forarnens transverses de C1 et de C2, du c6t6 droit (dR) et du cot6 gauche (dL)
the bony structures and the spatial relations between the vertebral b o d i e s . T h o s e p r o j e c t i o n s in c a u d a l (inferior) v i e w s w e r e s e l e c t e d w h i c h were the most informative as regards the anatomy. In the static studies, the caudal views in every subject showed the slightly anterior position o f a line join i n g the centers o f the transverse foram i n a o f C 2 in relation to a n equivalent line j o i n i n g those o f C I (Fig. 2c). I n t h e 2 - D d y n a m i c studies, the rotational values were c o m p a r a b l e to t h e results r e p o r t e d in the l i t e r a t u r e (Table 1) [1, 11]. T h e m e a n rotation o f CO was 49.3 ° (range 43-53°). T h e m e a n difference in axial rotation betw e e n C1 a n d C 2 was 31,8 ° (range 2538°). In the three-dimensional reconstructions the rotational d i s p l a c e m e n t o f C1 o n C 2 w a s p a r t i c u l a r l y w e l l s h o w n . W i t h the h e a d t u r n e d to the left, t h e r i g h t a r t i c u l a r m a s s o f t h e atlas m o v e d anteriorly a n d medially, w h e r e a s t h e left lateral m a s s u n d e r w e n t posteromedial rotation. T h e ventral v i e w s s h o w e d a simultaneous tilt o f b o t h C t and C 2 towards the side o f the rotation in relation to the horizon-
tal plane, In the majority o f cases, the tilt o f C 2 was greater than that of C1, r e s u l t i n g in a r e l a t i v e d i s p l a c e m e n t d o w n w a r d s o f the right articular m a s s o f C1 i n r e l a t i o n to that o f C 2 (Fig. 3a). T h e c a u d a l v i e w s d e m o n s t r a t e d an a s y m m e t r i c increase o f the distance b e t w e e n the ipsilateral t r a n s v e r s e f o r a m i n a o f C 2 and C 2 (Fig. 3c). This distance was smaller on the side o p p o s i t e to the d i r e c t i o n o f r o t a t i o n (right side) than o n the side o f rotation (left side). T h e m e a n distance o n the right side was 2 2 % smaller t h a n that o n the left side (range 15% to 4 1 % ) (Table 1 ) , T h e lines joining the right transverse foramina of C1 and C2 and the left transverse foramina of C 1 and C 2 were parallel, c o n f i r m i n g that the caudal projection h a d been m a d e in the transverse p l a n e at the start of m o v e ment. These observations were reproducible in all the subjects e x a m i n e d . Discussion
During rotation of the atlantoaxial complex, we observed a displacem e n t o f the b o n y c o m p o n e n t s in the t h r e e p l a n e s o f space. W h e r e a s t h e
204
basic 2-D transverse sections allow analysis only of the transverse rotational displacement of each single bony segment independent of the others [1, 11], the 3-D reconstructions demonstrate the relative displacements of one bony segment in relation to another. The difficulty in analysing the three-dimensional shift of C t and C2 is due to the fact that their spatial relationships appear to vary with each modification of angle in the view plane chosen. It is therefore necessary, for each plane of spatial displacement, to deten-nine in advance a plane of reference which is the start plane of the bony segments in neutral position, and to observe the bony components after their rotation from this reference plane. This was the principle on which our analysis of rotation of C1 and C2 in the transverse plane was based. The transverse plane of reference chosen was the transverse start plane of C2 (which was not always horizontal in neutral position). Moreover, to obtain the transverse projection of rotation in this fixed reference plane, it was also necessary to reestablish its initial sagittal tilt. The results presented are therefore those obtained by analysing the projection of the relative rotation of C 1 and C2 in a fixed transverse plane which was the start plane of the movement. Finally, the position of the subject on the examination table should reproduce as nearly as possible the normal neutral position of the craniocervical junction as evaluated radiologically with the subject in the upright position. When a minimal difference of sagittal inclination of axis of the dens was observed on the topogram of the static CT examination, this was corrected on the three-dimensional reconstruction. Thus, the projection of the bony segments obtained gave the best indication of their relationships in the normal neutral position. The p r i m a r y f u n c t i o n of the atlantoaxial joint is rotation [1, 11]. The cranial views confirm that,
J-L Dumas et al : Three-dimensional CT analysis of atlantom'dal rotation
when rotation of CO is less than 63 ° [8], the projection into the atlantic ring of the articular mass of C2 opposite to the direction of rotation is such that the spinal canal remains wide enough to prevent compression of the neuraxis (Fig. 3b). The caudal views show that the projection of the rotation of C 1 and C2 in a transverse plane defines an asymmetric increase of the distance between the transverse foramina of C1 and C2 (Fig. 3c). This is probably to maintain the blood- flow in the vertebral aa. This mechanism prevents excessive anterior stretching of the vertebral a. of the side opposite to the direction of rotation, while the vertebral a. on the side of rotation remains directed upward and backward. This projection in the transverse plane of rotation of C1 mad C2 may be s c h e m a t i c a l l y o b t a i n e d when the center of rotation of each vertebra is situated exactly at the middle of the line joining the centers of its transverse foramina. This asymmetric displacement develops because the transverse foramina of C2 are initially in an anterior position relative to those of C1 (Fig. 4). It is clear that 2-D functional CT imaging is useful in evaluating abnormalities of rotation of the upper cervical spine [2, 5, 10, 12]. In the absence of functional studies, it may be impossible to distinguish a fixed atlantoaxial rotation from an acute transient torticollis [7]. It is important to recognise fixed rotation because of its inherent neurologic risks [4]. Functional 2-D CT studies of cadavers have demonstrated hypermobility in rotation of the upper cervical spine after section of the alar ligaments [3]. Abnormally extensive atlantoaxial rotation may severely endanger the vertebral aa., particularly when combined with anterior displacement of C1 [4]. The present study shows that 3-D images significantly add to the understanding of rotational aflantoaxial shift as compared with routine axial CT imaging.
Functional 3-D CT studies should promote the investigation of such disorders. The method and the anatomic results reported here may provide a basis for the three-dimensional CT investigation of abnormalities of rotation in the upper cervical spine. References 1.
Dumas JL, Sainte Rose M, Dreyfus P, Goldlust D, Chevrel JP (1993) Rotation of the cervical spinal column: a computed t o m o g r a p h y in vivo study. Surg Radiol Anat 15 : 333-339 2. Dvorak J, Hayek J, Zehnder R (1987) CTfunctional diagnostics of the rotator 5, instability of the upper cervical spine: an evaluation on h e a l @ adults and patients with suspected instability. Spine I2 : 726-731 3. Dvorak J, Panjabi M, Gerber M, Wichmann W (1987) CT-functional diagnostics of the rotatory instability of the upper cervical spine: an experimental study on cadavers. Spine 12 : 197-205 4. Fielding JW, Hawkins RJ (1977) Arianto-axial rotatory fixation. J Bone Joint Surg [Am] 59-A : 37-44 5. Fielding JW, Stillwell WT, Chynn KY, Spyropoulos EC (1978) Use of computed t o m o g r a p h y for the diagnosis of atlanto-axial rotatory fixation. J Bone Joint Surg [Am] 6 0 - A : 1 1 0 2 - 1 1 0 4 6. Fishman EK, Magid D, Ney DR, et al (1991) T h r e e - d i m e n s i o n a l i m a g i n g . Radiology 181 : 321-337 7. Kowalski HM, Cohen WA, Cooper P, Wisoff JH (1987) Pitfalls in the CT diagnosis of atlantoaxial rotary subluxation. Am J Roentgenol 149 : 595-600 8. Mazzara JT, Fielding JW (1988) Effect of C I - C 2 rotation on canal size. Clin Orthop 237 : 115-119 9. Ney DR, Fishman EK, Kawashima A, Robertson DD, Scott W W (1992) Comparison of helical and serial CT with regard to three-dimensional imaging of musculoskeletal anatomy. Radiology 185 : 865-869 10. Ono K, Yonenobu K, Fuji T, Okada K (1985) Atlantoaxial rotatory fixation: radiographic study of its mechanism. Spine 10 : 602-608 11. Penning L, Wilmink JT (1987) Rotation of the cervical spine: a CT study in normal s u b j e c t s . Spine 12 : 7 3 2 - 7 3 8 12. Rinaldi I, Mullins WJ, Delaney WF, Fitzer PM, Tornberg DN (1979) Computerized tomographic demonstration of rotational atlanto-axial fixation. J Neurosurg 50 : 115-119
Received March 22, 1994/Accepted in final form April 29, 1994