Bio Mechanics of Occipitocervical Fixation

SPINE Volume 31, Number 7, pp 755–761©2006, Lippincott Williams & Wilkins, Inc.

Biomechanics of Occipitocervical Fixation

Paul A. Anderson, MD,* Ashish L. Oza, MS,* Thomas J. Puschak, MD,‡and Rick Sasso, MD†

Study Design. A human cadaveric biomechanical studycomparing occipital fixation techniques.

Objectives. To compare ranges of motion betweenmidline and lateral occipital fixation and between rigidand nonrigid occipital fixation of an unstable craniocervi-cal spine.

Summary of Background Data. New fixation tech-niques using rods and screws increase surgical choice onwhere fixation is placed onto the occiput. Lateral fixationtheoretically gives improved resistance to deformationbecause of its increased effective moment arm and bilat-eral purchase. Midline fixation allows significantly longerscrew purchase. This study compares these two fixationlocation.

Methods. Cadaveric occipital cervical spine specimenswere tested biomechanically intact and under six differentfixation techniques. Range of motion between the skulland C2 at 1.5 N-m and 2 N-m bending moments wasmeasured in flexion-extension, lateral bending, and axialrotation. Mechanical testing of different rod diametersand a reconstruction plate was performed and comparedwith biomechanical testing. Results were compared be-tween the intact condition and all fixations, between themedial and lateral fixations, and between the rigid andnonrigid fixations by analysis of variance.

Results. The range of motion of all constructs wassignificantly reduced compared with intact. Significantdifferences between groups were only seen in lateralbending in fixation placed laterally. Mechanical testingdemonstrated that construct stiffness was predicted byarea moment of inertia of the rod and plate to a greaterdegree than variation in placement of occipital screws orlocking of the implant.

Conclusion. The choice of location of occipital fixationshould be based more on the ease of use and instabilitypattern. The decreased stiffness of the newer small rodsystems should be considered.

Key words: occiput, occipitocervical instability, inter-nal fixation, cervical fusion, screw fixation. Spine 2006;31:

755–761

Occipitocervical fusion is indicated when the craniocer-vical junction is rendered unstable by various diseaseprocesses. Foerster first described reconstruction of theoccipitocervical junction using a fibular strut graft in1927.1 Historically, stabilization of the craniocervicaljunction consisted of onlay bone grafting with posteriorwiring. In time, techniques involving wiring of the occi-put and upper cervical spine to contoured rods or pinswere developed.2–5 These techniques often required ex-tended fusion segments and significant prolonged post-operative immobilization due to insufficient constructstability. Despite prolonged external immobilization(halo vest or Minerva casting), failure rates of up to 30%have been reported.6,7

Over the past decade, rigid fixation of the occiputusing plate-screw or rod constructs with screw fixationhas become popular.8–11 The increased rigidity of theseconstructs allows fusion of fewer segments and decreasesthe amount of external immobilization required aftersurgery. Techniques involving use of transarticularC1–C2 screws, C1 lateral mass screws, and C2 pediclescrews as caudal anchors have been shown to increaseconstruct rigidity especially in extension, rotation, andanteroposterior translation.12–14

Recent anatomic studies have shown that the externaloccipital protuberance is the thickest part of the occiputand that occipital thickness decreases in a radial fashionlaterally and inferiorly from this point.15–17 The variableregional bony thickness of the occiput, combined withthe location of the dural sinuses, dictates the safe work-ing length occipital screws, thus potentially affectingconstruct strength and rigidity. Occipital screws arecommonly placed parasagittally in line with the C1 andC2 lateral masses (rods or plates) or are placed in themidline (special “T” or “Y” shaped plates). Parasagittalscrew placement aligns with cervical instrumentation,allows placement of 6 screws, and is laterally placedfrom the midline, theoretically increasing its effectivemoment arm. Midline screw fixation, through specialmidline plates, allows for longer screws in the thickeroccipital keel, but usually only allows a maximum ofthree screws.

Occipital screws are placed through round and slottedholes in plates or connectors. Traditional plate designsallow screw head rotation or toggle in relation to theplate that theoretically can decrease construct stiffnessand lead to corrosion or screw back-out. This phenom-enon was observed commonly in anterior plate designs,such as Caspar and Orozco, but was mitigated usingbicortical screw purchase. Alternatively, using a rigidlylocked connection between the plate and screw improved

From the *Departments of Orthopedic Surgery and Rehabilitation andNeurologic Surgery, University of Wisconsin, Madison, WI; †IndianaSpine Group, Indianapolis, IN; and ‡Golden, CO.Acknowledgment date: February 28, 2005. First revision date: April 2,2005. Acceptance date: April 20, 2005.The device(s)/drug(s) that is/are the subject of this manuscript is/are notFDA-approved for this indication and is/are not commercially avail-able in the United States.Corporate/Industry funds were received in support of this work. Al-though one or more of the author(s) has/have received or will receivebenefits for personal or professional use from a commercial party re-lated directly or indirectly to the subject of this manuscript, benefitswill be directed solely to a research fund, foundation, educational in-stitution, or other nonprofit organization which the author(s) has/havebeen associated.Address correspondence and reprint requests to Paul A. Anderson,MD, Departments of Orthopedic Surgery and Rehabilitation and Neu-rologic Surgery, University of Wisconsin, 600 Highland Ave K4/736,Madison, WI 53972. E-mail: [email protected]

755

construct stiffness, required only unicortical fixation,and avoided screw back-out. The effect of having a com-pletely rigidly locked rod-connector-screw system com-pared with a nonrigid system where screw heads cantoggle is unknown.

The objective of this study is to compare ranges ofmotion between midline and lateral occipital fixationand between rigid and nonrigid occipitocervical fixationof an unstable craniocervical spine. We hypothesize thatlateral fixation is mechanically superior despite short-ened screw lengths. A second hypothesis is that the rig-idly locked systems are biomechanically superior to non-rigidly locked devices.

Materials and Methods

Specimen Preparation. Six fresh frozen human cadaveric oc-cipitocervical specimens were thawed overnight and cleared ofsoft tissue maintaining ligaments and capsules. The caudal end,at C7 and T1, was potted in epoxy resin and the occiput wasscrewed to a wooden block. The specimens were mounted ver-tically in a special fixture in a MTS (Material Testing System)858 servo hydraulic testing machine (MTS Systems Corp.,Eden Prairie, MN).

Biomechanical Protocol. Pure 2.0 N-m bending momentswere applied in flexion-extension, lateral bending, and axialrotation as previously described.13 Motors attached to both thecranial and caudal spinal ends applied the torques while a loadcell under the specimen recorded axial compression and axialtorque. Bending moments in flexion-extension and lateralbending were measured from strain gauges mounted on themotor shaft. The input load was a cosine wave with a 10 sec-onds period for 5 cycles. On the fifth cycle, data were recorded.

Displacements were recorded from extensometers rigidlyaffixed to the occiput and to C2. The data were reduced toangular displacements using trigonometry as previously re-ported by Sutterlin et al.13

Testing Sequence. After the specimens were tested intact, theunstable upper cervical spine was destabilized by osteotomy ofthe odontoid process with an oscillating saw. Other investiga-tors have used this injury model to examine fixation of thecraniocervical spine.11,13,18,19 Six different occipitocervicalstabilization techniques were applied and tested in random or-der, Table 1.

Fixation in the cervical spine was never altered during test-ing. Cervical fixation used C1 lateral masses screws and C2

pedicles screws. No loosening of the cervical fixation was notedat the end of the testing. In all test conditions, 3.2-mm-diameterrods were used. Thus, only the type and location of the occip-ital fixation were altered.

Occipital Fixation. The occipital fixation techniques areshown in Figure 1. Each occipital fixation used three screws perside except for the midline plate, which had only three midlinescrews.

Variable axis screws (Vertex, Medtronic, Memphis, TN)were placed in line with the lateral masses of C1 and C2, whichis defined as lateral (Figure 1a). The variable angle screws were

Figure 1. a, Dorsal view of placement of variable angle screws(Vertex) parasagittally in a line with C1 and C2 screws. The implantis rigidly locked at all junctions. The drawing demonstrates thelateral parasagittal location of the occipital screws. The occipitalscrews were also placed medially (not shown), which requiredmore rod contouring. b, Dorsal view of lateral connectors placedlateral so that the rod aligns with the C1 and C2 screws. Theconnectors were also located far laterally and medial, whichrequired more rod contouring (not shown). The implant is rigidlylocked at the connector–rod junction but not the occipital screws.c, Dorsal view of the midline plate. Three occipital screws areplaced in the midline, and the rods are locked laterally by setscrews. The implant is rigidly locked at the connector–rod junctionbut not the occipital screws.

Table 1. Fixation Groups

Fixation Type Location

IntactVariable axis screw, medial Rigid MedialVariable axis screw, lateral Rigid LateralConnectors, medial Nonrigid MedialConnectors, lateral Nonrigid LateralConnectors, far lateral Nonrigid LateralMidline plate Nonrigid Medial

The variable axis screws have rigid locking mechanisms, whereas the L-connectors and plate are nonrigid. The far lateral and lateral groups werecombined into lateral and the medial and midline as medial fixation.

756 Spine • Volume 31 • Number 7 • 2006

also placed medial as close to the midline as possible so that thepaired screws touched, (not shown). After tightening of thelocking nut, this implant is rigidly locked at all metal junctions.The variable axis screw instrumentation is not designed for usein this location but was used in this study for purposes ofhaving a rigid locking system so that the hypothesis statedabove could be examined.

L-shaped connectors have nonlocking holes for the occipitalscrews and at an orthogonal angle a second hole for the3.2-mm rod. A set screw locks the rod to the L-connector butthe occipital screws are not rigidly locked to the connectors.The L-connectors were placed in three positions: laterally inline with the lateral masses of C1 and C2 (Figure 1b), far lateral(approximately 5 mm) from the lateral position (not shown),and medially as close to the midline as possible (not shown).

The occipital plate has three nonrigidly locked holes formidline screw placement and arms that extend laterally, whichcontain grooves for rod attachment. The rod is rigidly lockedwith two set screws on each side, although the occipital screwsare not rigidly locked to the plate (Figure 1c).

Surgical Technique. Bicortical 2.4-mm drill holes wereplaced into the lateral masses of C1 and into C2 pedicles usingthe technique described by Harms and Melcher.20 Variableaxis screws ranging 40 to 50 mm were then inserted gainingbicortical purchase. Two 3.2-mm-diameter rods were cut tolength and contoured. They were fixed to C1 and C2 and thenused to locate screw holes for: far lateral, lateral, medial, andmidline screw placement. Three screws were placed in eachplane starting just below the external occipital protuberanceand proceeding caudal as allowed by the implants. The holeswere tapped and 3.5-mm screws were placed into the L-connectors or the plates. In the case of variable axis screws, thescrews themselves were placed. Screw purchase was bicorticalin all cases. After screw and rod placement, all connectionswere fully tightened.

Screw Depth Measurement. After all testing, the implantswere removed and the occipital screw depth was measured withthe reconstruction system depth gauge provided by the manu-facturer (#6905744, Medtronic, Memphis TN). The locationof the screws relative to the midline was measured with a digitalvernier caliper.

Mechanical Testing. Paired 3.0-, 3.2-, 3.5-mm diameter ti-tanium rods and 3.5-mm AO stainless steel reconstructionplates were contoured similarly to those fixations used in the

biomechanical testing. They were separated 30 mm and fixedusing grips at the approximate location of C1 and occipitalfixation positions. The construct was then tested using the sameprotocol as above.

Statistical Analysis. The range of motion between the occiputand C2 at �1.5 N-m and �2.0 N-m of torque for all three axeswas calculated. The differences between intact and all fixationsand between groups were measured by analysis of variance. Forstatistical purposes, the lateral and far lateral fixations weregrouped together as were the medial and midline groups. Asignificance level of P �0.05 was chosen. All analyses wereperformed with SAS (SAS Institute, Inc. Cary, NC).

Stiffness in N-m/degree was calculated using least squaresregression of load displacement plots in each loading mode.These results were compared with a previously performedstudy, which used the same biomechanical protocol. In thatstudy, axis plates, titanium 1⁄4-inch rods, and a Y-plate werecompared.13

Results

Overall Range of MotionThe overall ranges of motion at �1.5 N-m and �2 N-mtorque are given in Table 2. A statistically significantdecrease of motion in all planes was observed betweenthe intact and all fixation methods. At �2 N-m torque,the variable axis screws placed laterally had the leastrange of motion, 1.21° and 1.51°, in flexion-extensionand lateral bending, respectively. In axial rotation, thevariable axis screws placed medially had the least rangeof motion, 1.55°.

Rigid Versus Nonrigid ConstructsRigidly locked constructs include medially and laterallyplaced variable axis screws, while the nonrigid con-structs were the midline plate and the L-connectorsplaced in three different locations. The range of motionat �2 N-m between rigid and nonrigid constructs isshown in Figure 2. Rigid constructs had less motion in allplanes compared with the nonrigid constructs, but thedifferences were not statistically significant.

Medial Versus Lateral Screw PlacementMedial screw placement includes the medial L-connectors,the midline plate, and medial variable axis screws. Lateral

Table 2. Range of Motion at 1.5 N-m and 2 N-m Torque for Different Occipitocervical Stabilization Techniques

Range of Motion (°) Under 2 N-m Range of Motion (°) Under 1.5 N-m

Flexion-Extension

LateralBending Axial Rotation

Flexion-Extension

LateralBending Axial Rotation

Mean SE Mean SE Mean SE Mean SE Mean SE Mean SE

Intact 15.76 2.79 12.65 2.1 97.54 7 15.69 2.8 12.41 2.06 97.25 7.27Lateral vertex 1.21 2.79 1.51 0.47 3.28 0.98 1.08 1.9 1.39 0.45 2.57 0.79Parasagittal connectors 2.34 0.67 2.81 1.0 3.82 1.57 2.26 0.65 2.62 1.0 3.09 1.27Plate 2.15 0.52 2.39 0.43 1.96 0.86 1.96 0.44 2.13 0.43 1.75 0.79Midline vertex 2.36 0.77 3.66 1.45 1.55 0.48 2.21 0.78 3.3 1.39 1.24 0.42Lateral connectors 1.98 0.39 1.95 0.86 3.46 1.29 1.85 0.39 1.8 0.84 2.87 1.07Midline connectors 2.63 0.77 3.22 1.1 2.43 0.89 2.51 0.74 2.92 1.43 1.85 0.69

757Occipitocervical Fixation • Anderson et al

screw placement includes the L-connectors placed laterallyand far laterally and lateral variable axis screws. Compar-ison of range of motion at �2 N-m between the two loca-tions is shown in Figure 3. There were differences in therange of motion depending on the loading directions.There was greater motion between the segments for me-dial than lateral placed screws in flexion-extension andlateral bending. In axial rotation, however, range of mo-tion was greater for laterally placed screws. Motion dur-ing lateral bending for laterally placed screws was statis-tically significant (P � 0.05), but the differences were notstatistically significant for flexion-extension and axial ro-tation.

Screw DepthThe average screw depth is shown in Table 3. The screwdepth in the midline and more rostral was always greaterthan lateral and caudal. There was less than 3-mm dif-ference in length between the midline and medial screws.Therefore, there is likely only a small difference in rigid-ity between the midline and medially placed screws, es-pecially when considering that six total screws can beplaced medially and only three in the midline.

StiffnessThe results of comparing stiffness between rigid and non-rigid, and medial and lateral fixation were similar to theresults as reported by range of motion.

Mechanical TestingThe stiffness of 3.5-mm reconstruction plates, 3.0-mm,3.2-mm, and 3.5-mm rods is given in Figure 4. Similar tothe biomechanical study, the plate was significantly, (2 to3 times stiffer) than the 3- or 3.5-mm rods. Diminishedrod diameter resulted in reduced stiffness to the inversesquare of the rod diameter. Additionally, the plate testedwas stainless steel, which is approximately twice as stiffas titanium, which would account for some of the ob-served differences.

Discussion

The occiput is a flat bone that protects the contents of theposterior fossa and encircles on three sides the foramenmagnum. The anterior and posterior borders of the fo-ramen magnum are called the basion and opisthion, re-spectively. Convex occipital condyles border the fora-men magnum laterally and articulate with the lateral

Figure 2. Comparison of range ofmotion (degrees) between rigidand nonrigid implants under flex-ion-extension, lateral bending, andaxial rotation.

Figure 3. Comparison of range ofmotion (degrees) between medi-ally and laterally placed screwsunder flexion-extension, lateralbending, and axial rotation.


masses of the atlas. The posterior margin of the foramenmagnum, the external occipital protuberance, and thesuperior and inferior nuchal lines serve as bony land-marks during occipital screw placement. The externaloccipital protuberance is located in the midline approx-imately 4 cm cranial to the posterior edge of the foramenmagnum.15 The internal occipital protuberance lies op-posite of the external protuberance on the inner table.The external occipital protuberance is the thickest partof the occiput ranging from 11 to 17.5 mm and mainlyconsists of thick dense cortical bone with little cancellousbone.15–17 The outer cortical table contributes 45% ofthe total occipital thickness, while the inner cortical tabletends to be thin, measuring 4 to 6 mm, and representsonly 10% of occipital thickness. Zipnick et al17 andEbraheim et al15 have shown in separate studies that thethickness of the occiput is greatest at the external protu-berance and decreases in a radial fashion moving later-ally and inferiorly from the external protuberance. Thisassociation is consistent with the results of our screwlength measurements (Table 3). The occiput is thinnest inthe region of the cerebellar fossa.

The superior and inferior nuchal lines and the coarseintervening cortical bone serve as muscular attachmentsites. The trapezius and portions of the sternocleidomas-

toid attach to the superior nuchal line, while the rectuscapitis posterior major and minor insert on the inferiornuchal line. The semispinalis capitis and oblique capitissuperior muscles insert on the occiput between the supe-rior and inferior nuchal lines. The ligamentum nuchaeinserts at the external occipital protuberance.

Thorough knowledge of the location of the large ve-nous sinuses and the relative regional thickness of theocciput is required for safe placement of occipital screws.The transverse venous sinuses run horizontally at thelevel of the superior nuchal line and converge near themidline to form the confluence of sinuses, which is lo-cated opposite the external occipital protuberance. Theright transverse sinus tends to be larger than the left andis more likely to be located just inferior to the superiornuchal line. The sagittal sinus runs superiorly in the oc-cipital midline from the confluence, while the straightsinus runs anteriorly through the tentorium cerebelli.The internal occipital crest runs from the posterior fora-men magnum to the internal occipital protuberance andcontains the small occipital venous sinus which joins theconfluence.

The occiput has a relatively broad area for screw pur-chase and is relatively safe unless screw placement is toorostral, above the external occipital protuberance, or too

Table 3. Mean Screw Depth and Ranges of Depth at Variable Locations From the Midline

Left FarLateral(mm)

LeftParasagittal

(mm)Left Medial

(mm)Midline

(mm)Right Medial

(mm)

RightParasagittal

(mm)

Right FarLateral(mm)

Distances from the midline 22.8 15.95 6.91 8.23 16.67 26.28SE 0.93 0.6 1.06 1.3 2.07 2.25Top

Mean 7.17 7.17 9.83 12.33 8.17 7.17 6.67Range 4–10 5–9 7–16 9–17 4–11 4–10 5–8

MidMean 5.17 5.67 7.17 8.83 6 5.5 5Range 4–7 4–7 5–10 6–13 4–9 4–8 4–6

LowerMean 3.83 4.17 6.5 8.5 5 4.5 3.83Range 3–4 3–5 5–10 6–10 3–8 3–6 3–5

Figure 4. Results of mechanicaltesting of 3.0-, 3.2-, 3.5-mm rodsand 3.5-mm reconstruction plates.The much larger stiffness of theplates is from a much greater areamoment of inertia, which is muchmore metal at a greater distancefrom the axis of rotation.


long screws are used. Aligning the occipital screws withthe cervical fixation and maintaining satisfactory screwlength are the major surgical hurdles. This study demon-strates that occipital screw fixation can be placed eitherlaterally, in-line with C1 and C2 fixation points, or moremedial. The choice may depend on the instability patternwhere rotational patterns may have increased stabilitywith midline fixation and anteroposterior and lateralpatterns of instability by more lateral placed constructs.Instability in the vertical direction such as from rheuma-toid arthritis was not tested in this model, although hav-ing two parallel constructs laterally may have biome-chanical advantages.

We hypothesized that rigidly locked constructs, wherethe screw head cannot rotate in the rod/plate connector,would have greater stiffness and therefore less range of mo-tion. This was not observed as the smaller rod size with itsgreater elasticity had a much greater influence then the ri-gidity of the connectors. Although the polyaxial screws ofthe Vertex instrumentation afforded ease of use, theirprominence when placed on the occiput precludes their usein this application.

The location of fixation, lateral compared with mid-line, had only small effect on range of motion and wasdependent on load direction. Lateral fixation had signif-icantly less motion in lateral bending. No other signifi-cant differences were observed in this small sample size.In rotation, there appeared a trend toward less motion inrotation using medial fixation. We could not explain thisobservation based on any analytic calculations.

A prior occipital cervical fixation study from our lab-oratory compared the stiffness of axis plates, a Y-plate,and 6.4-mm stainless steel rods fixed by cable.13 Theobserved stiffness in flexion extension was 8.5, 5.7, and4.3 N-m/degree. These implants had 2 to 3 times greaterstiffness than the current study despite using the samemethodology and injury model (Figure 5). From obser-vations during testing, the decreased stiffness was sec-ondary to increased elasticity of the 3.2-mm rods com-pared with the significantly stiffer rods and plates. Some

of the difference was also related to the stiffer stainlesssteel used in the plates. These conclusions were validatedbased on the mechanical testing which showed that thestiffness in flexion-extension was related to I, area mo-ment of inertia for rods and beams. The tendency of abody to resist angular acceleration is moment of inertia,which depends not only on the mass of a body but thedistribution of mass around its axis. So bodies with agreater moment of inertia are more difficult to deformthan ones with lower moment of inertia. The three rodshad a circular cross section with different diameters. Thearea moment of inertia of a body with a circular crosssection is I � �D4/64 where D is the diameter of thecross section.

Figure 2 shows that a rod with a greater moment ofinertia has less motion under same applied torque. Theplate on the other hand has a rectangular cross sectionand the area moment of inertia is I � bh3/12 where b isthe width and h is the thickness of the plate.

In the case of a 3.5-mm reconstruction plate, b � 10mm and h � 3 mm so the I � 22.5 mm4. In the case of therod with D � 3.5 mm, the I of the rod � 7.36 mm4; thus,the ratio of plate to rod stiffness in the sagittal planeshould be about 3. This analytical calculation did accu-rately predict both the mechanical and cadaveric biome-chanical results.

Another mechanical factor of these small rod con-structs not tested in this study is their fatigue life. Underfatigue, titanium is notch sensitive, especially if bent overat a sharp angle as required at the craniocervical junc-tion. When bending titanium rods, it is important tocreate a gentle curve over a longer segment rather than asingle acute bend.

Excellent outcomes with high fusion successes havebeen reported using rigid plate-screw constructs.8,10,21,22

These constructs have significantly greater stiffness thanthe new generation of implants that have rod sizes rang-ing from 3.0 to 3.5 mm. Whether this decreased stiffnesswith its increased motion will affect bony healing is un-known.

Figure 5. Comparison of stiffnessvalues (Nm/°) of axis plate,13 ag-gregate of different types of in-strumentation in our study andlateral connectors in our study.


Compared with other studies analyzing occipitalscrew placement, depth was measured with a instrumentavailable at surgery. The lengths were therefore in nom-inal increments. Similar to the other studies, the screwdepth was greatest, mean 12.3 mm, in the midline justbelow the external occipital protuberance. However, awide range of depths was observed, ranging 9 to 17 mm.The keel of bone, the internal occipital crest separatingthe cellebellar fossi, which allows greater screw depth, isrelatively narrow, as the screw depth decreases to 9.3mm only 7 mm off the midline.23 Shorter depths areencountered as screws are placed caudally, even in themidline. In the midline, minimum lengths are 6 mm,whereas laterally they are as little as 3 mm. Screws of thisshort length are not routinely available in current instru-mentation sets. To gain purchase with short screws, bi-cortical lengths and a finer thread pitch will be required.

Conclusion

The location of occipital fixation and its rigidity haveonly a small influence on biomechanical properties.Screw depth of the caudal most screws, especially whenplaced off the midline into the cerebellar fossa, are ex-tremely short and are often less than the available screwlengths. The 3.0- to 3.5-mm rod screw devices for occi-put cervical stabilization have significantly less stiffnessin flexion-extension and a greater range of motion com-pared with plate and screw systems. This difference isfrom the greater area moment of inertia of the latterdevices. The biologic effect of decreased stiffness withthese constructs is unknown.

Key Points

● The location of occipital fixation has a small in-fluence on stiffness and range of motion followinginternal fixation of rod–screw constructs and de-pends on load direction.● Greater rod diameter and use of plates are asso-ciated with greater stiffness.● Screw depth laterally and caudally toward foramenmagnum is shorter than available screw lengths.

References

1. Foerster O. Die Leitungsbahnen des Schmerzgefuhls und die chirurgischeeBehandlung der Schmerzzustande. Berlin: Urbin and Schwarzenberg, 1927.

2. Wertheim SB, Bohlman HH. Occipitocervical fusion: indications, technique,and long-term results in thirteen patients. J Bone Joint Surg Am 1987;69:833–6.

3. Hamblen DL. Occipito-cervical fusion: indications, technique and results.J Bone Joint Surg Br 1967;49:33–45.

4. Robinson RA, Southwick WO. Indications and techniques for early stabili-zation of the neck in some fracture dislocations of the cervical spine. SouthMed J 1960;53:565–79.

5. Elia M, Mazzara JT, Fielding JW. Onlay technique for occipitocervical fu-sion. Clin Orthop 1992;280:170–4.

6. Robertson SC, Menezes AH. Occipital calvarial bone graft in posterior oc-cipitocervical fusion. Spine 1998;23:249–54.

7. Lowry DW, Pollack IF, Clyde B, et al. Upper cervical spine fusion in thepediatric population. J Neurosurg 1997;87:671–6.

8. Smith MD, Anderson P, Grady MS. Occipitocervical arthrodesis using con-toured plate fixation: an early report on a versatile fixation technique. Spine1993;18:1984–90.

9. Itoh T, Tsuji H, Katoh Y, et al. Occipito-cervical fusion reinforced by Lu-que’s segmental spinal instrumentation for rheumatoid diseases. Spine 1988;13:1234–8.

10. Grob D, Dvorak J, Panjabi M, et al. Posterior occipitocervical fusion: apreliminary report of a new technique. Spine 1991;16(suppl):17–24.

11. Richter M, Wilke HJ, Kluger P, et al. Biomechanical evaluation of a newmodular rod-screw implant system for posterior instrumentation of the oc-cipito-cervical spine: in-vitro comparison with two established implant sys-tems. Eur Spine J 2000;9:417–25.

12. Hurlbert RJ, Crawford NR, Choi WG, et al. A biomechanical evaluation ofoccipitocervical instrumentation: screw compared with wire fixation. J Neu-rosurg Spine 1999;90:84–90.

13. Sutterlin CE III, Bianchi JR, Kunz DN, et al. Biomechanical evaluation ofoccipitocervical fixation devices. J Spinal Disord 2001;14:185–92.

14. Currier BL, Papagelopoulos PJ, Neale PG, et al. Biomechanical evaluation ofnew posterior occipitocervical instrumentation system. Clin Orthop 2003;411:103–15.

15. Ebraheim NA, Lu J, Biyani A, et al. An anatomic study of the thickness of theoccipital bone: implications for occipitocervical instrumentation. Spine1996;21:1725–9.

16. Haher TR, Yeung AW, Caruso SA, et al. Occipital screw pullout strength: abiomechanical investigation of occipital morphology. Spine 1999;24:5–9.

17. Zipnick RI, Merola AA, Gorup J, et al. Occipital morphology: an anatomicguide to internal fixation. Spine 1996;21:1719–24.

18. Puttlitz CM, Melcher RP, Kleinstueck FS, et al. Stability analysis of cranio-vertebral junction fixation techniques. J Bone Joint Surg Am 2004;86:561–8.

19. Oda I, Abumi K, Sell LC, et al. Biomechanical evaluation of five differentoccipito-atlanto-axial fixation techniques. Spine 1999;24:2377–82.

20. Harms J, Melcher RP. Posterior C1–C2 fusion with polyaxial screw and rodfixation. Spine 2001;26:2467–71.

21. Vale FL, Oliver M, Cahill DW. Rigid occipitocervical fusion. J Neurosurg1999;91:144–50.

22. Fehlings MG, Errico T, Cooper P, et al. Occipitocervical fusion with a five-millimeter malleable rod and segmental fixation. Neurosurgery 207;32:198–207.

23. Roberts DA, Doherty BJ, Heggeness MH. Quantitative anatomy of the oc-ciput and the biomechanics of occipital screw fixation. Spine 1998;23:1100–7.


Documents

Bio Mechanics of Occipitocervical Fixation