Study Design. Destructive compression tests and finite element analyses wer
e conducted to investigate the biomechanical strength at the graft-endplate
interface in anterior cervical fusion.
Objectives. To investigate The effect of endplate thickness, endplate holes
, and bone mineral density of the vertebral body on the biomechanical stren
gth of the endplate-graft interface in an anterior interbody fusion of the
Summary of Background. Subsidence of the graft into the vertebral body is a
well-known complication in anterior cervical fusion. However, there is no
information in the literature regarding the compressive strength of the gra
ft-endplate interface in relation to the endplate thickness, holes in the e
ndplate, and bone mineral density of the vertebral body.
Methods. Biomechanical destructive compression tests and finite element ana
lyses were performed in this study. Cervical vertebral bodies (C3-C7) isola
ted from seven cadaveric cervical spines (age at death 69-86 years, mean 79
years) were used for compression tests. Bone mineral density of each verte
bral body was measured using a dual energy radiograph absorptiometry unit.
Endplate thickness was measured using three coronal computed tomography ima
ges of the middle portion of the vertebral body obtained using a computer-a
ssisted imaging analysis. Then each vertebral body was cut into halves thro
ugh the horizontal plane. A total of 54 specimens, consisting of one endpla
te and half of the vertebral body, were obtained after excluding eight vert
ebrae with gross pathology on plain radiograph. Specimens were assigned to
one of three groups with different endplate conditions (Group I, intact; Gr
oup II, partial removal; and Group ill, complete removal) so that group mea
n bone mineral density became similar. Each endplate was slowly compressed
until failure using an 8-mm-diameter metal indenter, and the load to failur
e was determined as a maximum force on a recorded force-displacement curve.
The effect on the strength of the graft-endplate interface of various hole
patterns in the endplate was studied using a finite element technique. The
simulated-hole patterns included the following: one large central hole, tw
o lateral holes, two holes in The anterior and posterior portion of the end
plate, and four holes evenly distributed from the center of the endplate. S
tress distribution in the endplate was predicted in response to an axial co
mpressive force of 110 N, and the elements with von Mises stress greater th
an 4.0 MPa were determined as failed.
Results, The endplate thickness and bone mineral density were similar at al
l cervical levels, and the superior and inferior endplates had similar thic
kness at all cervical levels. There was no significant association between
bone mineral density and endplate thickness. load to failure was found to h
ave a significant association with bone mineral density but not with endpla
te thickness. However, load to failure tends to decrease with incremental r
emoval of the endplate, and load to failure of the specimens with an intact
endplate was significantly greater than that of the specimens with no endp
late. Finite element model predictions showed significant influence of the
hole pattern on the fraction of the upper endplate exposed to fracture stre
ss. A large hole was predicted to be more effective than the other patterns
at distributing a compressive load across the remaining area and thus mini
mizing the potential fracture area.
Conclusion. Results of this study suggest that it is important to preserve
the endplate as much as possible to prevent graft subsidence into the verte
bral body, particularly in patients with poor bone quality. It is preferabl
e to make one central hole rather than multiple smaller holes in the endpla
te for vascularity of the bone graft because it reduces the surface area ex
posed to fracture stresses.