THE ANATOMICAL RECORD 247:542–555 (1997)

Anatomy of the Sheep Spine and Its Comparison to the Human Spine
HANS-JOACHIM WILKE,* ANNETTE KETTLER, KARL HOWARD WENGER,
AND LUTZ EBERHARDT CLAES
Department Unfallchirurgische Forschung und Biomechanik,
Universität Ulm, Ulm, Germany

ABSTRACT Background: The sheep spine is often used as a model for

the human spine, although the degree to which these spines are anatomi-
cally comparable has yet to be categorically established. The purpose of
this study was to investigate the characteristic anatomical dimensions of
the sheep spine and to compare these with existing human data.
Methods: Five complete spines were measured to determine 21 dimen-
sions from the pedicles, spinal canal, transverse and spinous processes,
facets, endplates, and disc.
Results: The results showed that sheep and human vertebrae are most
similar in the thoracic and lumbar regions, although they show substan-
tial differences in certain dimensions. Morphological variations as a
function of spine level typically were well matched in the two species.
Conclusions: Sheep spine may be a useful model for experiments related
to the gross structure of the thoracic or lumbar spine, with certain
limitations for the cervical spine. A thorough database has been provided
for deciding the appropriateness of using the sheep spine as a model for
the human spine. Anat. Rec. 247:542–555, 1997. r 1997 Wiley-Liss, Inc.

Key words: spinal biomechanics; spine; comparative anatomy; human;

sheep; ovine; gross anatomy; in vitro testing

Fresh human specimens are increasingly difficult to its utility as a model for the human spine in in vivo and
obtain for in vitro experiments, and when available in vitro experiments.
such specimens are required in large quantities to
overcome the wide scattering effect associated with MATERIALS AND METHODS
biological variability (Ashman et al., 1989). Specifically, Five spines from 3- to 4-year-old female merino sheep
to provide a model for the human spine, animals such with a weight of 72.1 6 7.3 (57–81) kg were harvested
as sheep, goat, pig, calf, and dog have been used. Such for this study. After storage at 220°C in double-sealed
animal specimens are more readily available (Edmons- bags, the specimens were thawed and then directly
ton et al., 1994) and show much better homogeneity measured. The use of frozen stored specimens is consis-
than do human specimens when selected for breed, sex, tent with the procedures of some investigators (Mor-
age, and weight (Eggli et al., 1992; Gurwitz et al., 1993). oney et al., 1988; Green et al., 1993; Xu et al., 1995),
Sheep in particular are often used as a model for in vivo whereas others have used dried specimens (Francis,
studies concerning, for instance, histomorphology of 1955; Berry et al., 1987; Scoles et al., 1988; Doherty and
the intervertebral disc (Osti et al., 1990; Moore et al., Heggeness, 1994, 1995), but this technique and forma-
1992; Gunzburg et al., 1993) and biomechanical efficacy lin fixation tend to shrink the tissue.
of fusion techniques in the lumbar spine (Ahlgren et al., In general, sheep spines consist of 7 cervical, 12–14
1994). Sheep spines have also been used in vitro to thoracic, and 6–7 lumbar vertebrae (Nickel et al., 1984).
study the initial stabilizing effect of spinal implants in For consistency, only sheep with the most common
the lumbar (Slater et al., 1988; Yamamuro et al., 1990; number of vertebrae—7 cervical, 14 thoracic, and 7
Nagel et al., 1991) and cervical regions (Vazquez- lumbar—were selected. All muscles were dissected, and
Seonae et al., 1993). at first the ligaments and intervertebral discs were
Comprehensive, quantitative data on the characteris- kept intact to maintain the physiological curvature of
tic anatomy of the sheep spine, however, are lacking.
Knowledge of the similarities and differences between
sheep and human spines is essential for interpreting Received 23 August 1996; accepted 3 October 1996.
results from studies using this model and is needed to *Correspondence to: Priv. Doz. Dr. Hans-Joachim Wilke, Depart-
establish for which investigations the sheep model is ment Unfallchirurgische Forschung und Biomechanik, Universität
suitable. Thus, the purpose of this study was to provide Ulm, Helmholtzstra¬e 14, 89081 Ulm, Germany; E-mail: wilke@sirius.
medizin.uni-ulm.de
an anatomical database of the sheep spine and a Contract grant sponsor: University Hospital of Ulm; Contract grant
detailed comparison with the human spine to improve number: P.272.

r 1997 WILEY-LISS, INC.

 SHEEP SPINE ANATOMY 543

Fig. 1. Anatomical definition of reported dimensions. Abbreviations the sheep spine, cranial view. e: C4 of the sheep spine, dorsal view. f:
are listed in Table 1. a: T6 of the sheep spine, lateral view. b: T6 of the Thoracic vertebra, oblique perspective (adapted from Panjabi et al.,
sheep spine, cranial view. c: L4 of the sheep spine, dorsal view. d: L4 of 1993).

the spine. After measuring anterior disc height, the refer to the cranial joint space and were measured on
ligaments and disc were removed to determine verte- the right side.
bral dimensions (Fig. 1). Table 1 lists the nomenclature
key for the various measurements. Angular Dimensions
Angles were measured with a three-dimensional
Linear Dimensions goniometric linkage system consisting of six potentiom-
Lengths, heights, and widths were measured with a eters connected by five rigid rods and yielding a verified
hand-held micrometer. Accuracy of the measurements accuracy of 0.1° and 0.1 mm (Wilke et al., 1994). Assuming
was governed by the users’ definition of the anatomical symmetry, angles of the transverse processes and multiple
landmarks, generally yielding a repeatability of about joint surfaces were measured only on the right cranial side.
0.5 mm. Assuming structural symmetry, pedicle height Various measuring probes were fabricated to maintain
and width, and transverse process length were mea- the reliability and repeatability of the apparatus in its
sured only on the right side. Facet height and width contact with the various bony surfaces.
544 H.-J. WILKE ET AL.

TABLE 1. Anatomical parameters and their appeared to deviate somewhat from columnar. In gen-
abbreviations in this paper eral, the greater width than depth of the sheep end-
Body part Abbreviations Dimension plates—as much as 2.3-fold at L7—was shown by an
oval shape of the vertebral cross section.
Vertebral Body VBH Vertebral body height All vertebrae were taller than wide (1.5-fold at C3
EPW End-plate width and 1.2-fold at T7) and wider than deep, especially in
EPD End-plate depth
Pedicle PDH Pedicle height
the lumbar spine.
PDW Pedicle width
Spinal canal SCD Spinal canal depth
Pedicles
SCW Spinal canal width Pedicle height (PDH) in the cervical region decreased
Spinous and SPL Spinous process length caudally from 38.9 6 1.6 mm at C2 to 15.3 6 1.6 mm at
transverse process SPA Spinous process angle T1 and then increased continuously to 36.8 6 1.4 mm at
TPL Transverse process length L5; L6 and more so L7 were shorter in PDH (Table 3).
TPA Transverse process angle
TPW Transverse process width Pedicle width (PDW) was smallest between C2 and C6
Articular facet FCH Facet height (4.7 6 0.7 to 6.3 6 0.8 mm), although greatest immedi-
FCW Facet width ately adjacent at C7 (14.4 6 0.8 mm). PDW then
IFW Interfacet width generally decreased through T13 (7.6 6 1.1 mm) and
CAX* Card angle about x-axis then increased through L7 (13.4 6 1.1 mm).
CAY* Card angle about y-axis The pedicles always were taller than wide. In the
Intervertebral disc IDH Intervertebral disc height cervical vertebrae, PDW was as little as 14.6% of PDH.
Suffices u upper or cranial A similar relationship was seen in the lumbar verte-
l lower or caudal brae. The difference between VBHp and PDH was least
a anterior
p posterior in the lumbar vertebrae, with a minimum of 4.7 mm at
L5, and greatest in the cervical vertebrae, with a
*The two angles CAX and CAY represent tilting angles of a card, which maximum of 14.1 mm at C4.
was tilted relatively to the transversal plane first around the x-axis (CAX)
and then around the y-axis (CAY) to have the same orientation of the joint
plane.
Spinal Canal
Spinal canal width (SCW) was greatest in the cervi-
cal region, reaching 17.9 6 0.8 mm at C7, then nar-
RESULTS rowed through the thoracic region to 10.9 6 0.8 at T10,
Vertebral Bodies and increased in the lumbar region to 17.5 6 1.2 mm at
L7 (Table 3). Spinal canal depth (SCD) was also largest
Posterior vertebral body height (VBHp) decreased in the cervical spine, with a maximum of 15.2 6 1.0 mm
steadily through the cervical spine from 46.8 6 1.9 mm at C2, then decreased to 8.6 6 0.7 mm at T13, and
at C2 (the most cranial generic vertebra) to 27.0 6 1.1 increased to 10.9 6 1.2 mm at L7.
mm at C7 (Table 2). VBHp in the upper thoracic spine The spinal canal was wider than deep over the whole
was fairly constant at about 26 mm. From mid-thoracic column; thus, as with the vertebral body, it displayed an
through L6, VBHp increased, reaching 41.6 6 1.1 mm oval shape most pronounced in the lumbar vertebrae
at L6; L7 was 5.4 mm shorter. Anterior height (VBHa) and weakest in the thoracic vertebrae.
showed a trend similar to VBHp, measuring 56.3 6 3.1
mm at C2 and 26.6 6 1.4 mm at C7 and then increasing, Spinous and Transverse Processes
beginning in the mid-thoracic spine to a maximum of
39.4 6 1.7 mm at L6. VBHp in the lumbar region was as Spinous process length (SPL) reached a minimum of
much as 3.5 mm (9.3%) larger than VBHa. This differ- 13.0 6 1.2 mm at C3, a maximum of 81.3 6 3.1 mm at
ence likewise was consistent, although smaller, in the T3, and then decreased to 28.8 6 1.7 mm at T13,
thoracic region. However, in the cervical spine VBHa remaining at about this length through the lumbar
typically was larger than VBHp, most notably at C2 spine (Table 4). Minimum transverse process length
(9.5 mm, or 20.3%, difference). The width of the cranial (TPL) was 17.9 6 0.9 mm at T5 and maximum at 63.8 6
endplate (EPWu) was typically approximately 25 mm 2.9 mm at L4 and L5; cervical TPL was somewhat
in the upper cervical spine, decreasing caudally to T3, larger than was thoracic TPL.
and then increasing again to 32.7 6 0.6 at L7. C2 with In the cervical spine, it was difficult to determine the
an EPWu of 52.6 6 1.3 mm was exceptionally different. angles of the spinous and transverse processes (SPA
Caudal endplate width (EPWl) was greater than EPWu and TPA) because of their short length, which was also
at each level except at C2, with the largest differences true for TPA in the thoracic spine. SPA from T1 to T7
at the cervical and lumbar ends. EPWl at L7, for increasingly lay flatter, reaching a maximum caudally
example, was 7.7 mm or 23.6% greater than EPWu. directed angle of 44.3 6 2.0°, and then again became
Depth of the endplates (EPD) was the smallest of the more dorsally projected, transitioning to a cranial
vertebral body dimensions, and between the lumbar projection at the thoracolumbar junction. SPA reached
and thoracic regions the most consistent in caudal- a cranially directed maximum of 20.0 6 1.4° at L1. TPA
cranial and regional comparisons. The lowest value was was generally just below 20° in the lumbar region,
17.7 6 0.9 mm at L7, although in the lumbar spine EPD although higher at L7.
was closer in general to 20 mm and the highest value
was 26.5 6 1.2 mm at C4. The caudal-cranial difference Facet Joints
of both endplate width and depth for a given level was The surface of the facets in the sheep cervical spine
remarkable only in the cervical region, where the shape was flat in general, with a dorsolateral orientation. In
 SHEEP SPINE ANATOMY 545
TABLE 2. Dimensions related to the vertebral body of the sheep
(mean 6 S.D. in mm)
Vertebra EPWl EPWu EPDl EPDu VBHa VBHp
C2 28.6 6 1.5 52.6 6 1.3 25.1 6 0.5 17.7 6 0.7 56.3 6 3.1 46.8 6 1.9
C3 29.2 6 1.0 24.7 6 1.2 25.8 6 0.8 20.6 6 0.7 44.6 6 1.2 45.5 6 1.9
C4 29.1 6 1.4 25.8 6 1.0 26.5 6 1.2 20.7 6 1.2 44.3 6 1.7 43.8 6 2.1
C5 29.3 6 1.4 24.4 6 1.4 24.6 6 1.3 21.1 6 1.4 40.3 6 1.2 39.6 6 1.5
C6 25.8 6 1.4 23.4 6 1.1 23.6 6 1.1 20.2 6 1.2 34.2 6 1.1 33.8 6 1.6
C7 24.4 6 1.4 21.8 6 1.6 21.8 6 1.0 20.8 6 1.2 26.6 6 1.4 27.0 6 1.1
T1 19.7 6 0.6 20.7 6 1.4 20.4 6 1.4 20.7 6 1.2 24.0 6 1.2 26.8 6 1.0
T2 19.7 6 1.2 17.8 6 1.1 19.8 6 1.2 19.8 6 0.8 25.1 6 0.8 26.0 6 1.2
T3 20.6 6 0.9 18.9 6 1.2 19.8 6 0.8 19.2 6 0.8 26.1 6 1.1 26.5 6 0.8
T4 21.3 6 0.6 20.5 6 1.2 19.3 6 1.0 18.7 6 0.8 25.6 6 0.4 26.0 6 0.9
T5 21.7 6 0.7 20.4 6 0.4 19.3 6 0.8 18.6 6 0.7 25.0 6 1.1 26.1 6 1.1
T6 21.3 6 1.0 20.6 6 0.8 19.0 6 0.8 18.5 6 0.9 24.3 6 0.9 25.2 6 1.0
T7 21.4 6 0.7 21.1 6 0.8 18.6 6 0.7 18.5 6 0.5 24.3 6 0.7 25.5 6 0.6
T8 20.9 6 0.8 21.2 6 0.8 19.0 6 0.8 18.2 6 0.4 25.2 6 0.8 26.1 6 0.4
T9 22.2 6 0.6 21.2 6 0.6 18.5 6 0.7 18.6 6 0.7 216.2 6 0.6 27.9 6 1.1
T10 22.9 6 0.8 21.9 6 0.4 18.5 6 0.7 18.5 6 0.9 27.1 6 0.7 28.9 6 1.0
T11 24.5 6 0.9 23.4 6 0.7 18.7 6 0.4 18.7 6 0.4 28.9 6 0.7 30.5 6 0.6
T12 25.0 6 1.0 23.9 6 0.4 18.7 6 0.8 18.6 6 0.7 30.1 6 0.4 31.9 6 0.4
T13 26.6 6 0.7 25.2 6 0.4 19.0 6 0.5 18.7 6 0.9 32.4 6 0.7 34.3 6 0.8
L1 26.3 6 1.1 25.0 6 0.0 20.0- 6 0.0 19.5 6 0.7 35.0 6 1.4 37.0 6 0.7
L2 28.9 6 1.2 26.5 6 0.8 20.4 6 0.5 20.4 6 0.7 36.5 6 1.4 38.7 6 0.8
L3 29.8 6 1.3 26.5 6 0.9 20.0 6 0.6 20.5 6 0.7 37.2 6 1.3 40.2 6 1.2
L4 31.0 6 0.6 27.4 6 0.4 20.1 6 0.7 20.8 6 0.8 37.6 6 1.2 41.1 6 0.8
L5 32.1 6 1.1 28.7 6 1.0 19.5 6 0.6 20.4 6 0.8 39.7 6 1.6 41.5 6 1.3
L6 36.4 6 1.6 30.5 6 0.5 18.5 6 0.5 19.7 6 0.4 39.4 6 1.7 41.6 6 1.1
L7 40.4 6 2.0 32.7 6 0.6 17.7 6 0.9 17.6 6 0.7 34.4 6 1.2 36.2 6 1.6

the thoracic spine, the facet surface had the character TABLE 3. Dimensions related to the pedicles and the
of a shallow peanut shell bent into a slight crescent spinal canal of the sheep (mean 6 S.D. in mm)
shape facing anteromedially, with more open lips at the
Vertebra PDW PDH SCW SCD
ends to provide slight axial rotation. More caudally and
into the lumbar spine, this shape became slightly bent C2 6.9 6 0.7 38.9 6 1.6 16.9 6 1.4 15.2 6 1.0
in the vertical plane, providing a tilting degree of C3 4.7 6 0.7 32.3 6 2.0 15.2 6 1.2 11.5 6 0.4
freedom (Fig. 1d). Given the complexity of the shape, C4 4.4 6 0.7 30.2 6 1.2 16.0 6 1.1 11.4 6 0.4
standardization of the measurements was difficult, and C5 5.1 6 0.8 27.6 6 1.6 15.7 6 0.8 12.0 6 0.4
C6 6.3 6 0.8 22.3 6 1.4 17.2 6 0.9 13.4 6 0.8
generally larger standard deviations resulted from C7 14.4 6 0.8 16.7 6 1.0 17.9 6 0.8 13.5 6 0.4
these dimensions. Nevertheless, for a quantifiable ori- T1 14.0 6 1.3 15.3 6 0.8 17.2 6 0.6 12.0 6 0.6
entating perspective, the angles relative to the anatomi- T2 13.4 6 1.2 16.3 6 0.3 13.9 6 0.5 10.3 6 0.4
cal planes and the lengths of the major axes are reported. T3 12.6 6 0.8 17.5 6 0.9 13.6 6 0.4 10.0 6 0.5
The angle about the x-axis relative to the coronal T4 10.9 6 0.7 17.7 6 1.0 13.1 6 0.4 9.9 6 0.2
plane (CAX; Fig. 1f), reached a minimum of 64.7 6 3.4° T5 11.2 6 0.9 17.7 6 1.0 12.1 6 0.7 9.4 6 0.4
at C3 and developed to just over 80° in the thoracic T6 10.4 6 0.8 17.8 6 1.3 11.8 6 0.3 9.2 6 0.4
region (Table 5). The angle about the y-axis relative to T7 9.8 6 0.9 18.5 6 1.2 11.0 6 0.4 8.8 6 0.8
the frontal plane (CAY) was oriented dorsolaterally in T8 9.7 6 0.9 19.8 6 1.4 10.9 6 0.4 8.9 6 0.2
T9 10.2 6 0.8 21.7 6 1.0 11.1 6 1.0 8.8 6 0.3
the cervical spine, decreasing from 37.8 6 5.1° at C3 to T10 10.2 6 0.8 23.6 6 1.1 10.9 6 0.8 8.9 6 0.4
19.0 6 2.6° at C6. An abrupt transition of CAY through T11 10.5 6 0.6 24.9 6 1.5 11.4 6 0.8 8.5 6 0.6
the frontal plane occurred between T1 and T2, with the T12 8.7 6 1.2 26.2 6 1.0 11.6 6 0.7 8.5 6 0.6
remaining thoracic facet joints being oriented between T13 7.6 6 1.1 30.1 6 1.0 11.6 6 0.5 8.6 6 0.7
9.4 6 1.4° and 18.6 6 2.6° anterolaterally. L1 8.0 6 0.7 33.0 6 0.7 11.5 6 0.7 8.5 6 0.7
Facet joint height (FCH, measured at the cranial L2 8.6 6 0.5 33.9 6 1.4 12.3 6 0.8 8.4 6 0.4
joint; Fig. 1), was between 13.2 6 1.0 and 15.3 6 1.7 mm L3 9.5 6 0.9 35.7 6 1.4 12.6 6 1.1 8.6 6 0.7
in the cervical spine, with a similar range in the lumbar L4 9.6 6 1.0 36.3 6 1.4 12.9 6 1.1 8.8 6 0.4
spine, and was lower in general in the thoracic spine. A L5 10.2 6 1.0 36.8 6 1.4 13.9 6 1.6 9.0 6 0.6
L6 10.7 6 1.0 35.3 6 1.5 16.1 6 0.9 10.1 6 0.8
relatively large transition occurred between T11 and L7 13.4 6 1.1 29.0 6 1.9 17.5 6 1.2 10.9 6 1.2
T12. Facet joint width (FCW) was very similar to FCH
except in the thoracic region, where, typically, it was
considerably less.
Interfacet width (IFW) was greatest in the cervical
region, reaching 33.7 6 1.4 mm at C7, least in the
thoracic region with a minimum of 8.2 6 0.8 mm at T6, Intervertebral Discs
and then increased up to 30.6 6 1.1 mm in the lumbar Anterior disc height (IDH) was greatest in the cervi-
region. IFW actually exceeded EPW in the cervical cal region, between 6.8 6 0.6 and 7.2 6 0.8 mm, least in
region. the thoracic region, between 2.6 6 0.2 and 4.5 6 0.8
546 H.-J. WILKE ET AL.

TABLE 4. Dimensions and orientation of the spinous and transverse processes

of the sheep (mean 6 S.D. in mm and °)
Vertebra TPW mm SPL mm SPA ° TPL mm TPA °
C2 53.3 6 2.6 20.5 6 1.4 22.3 6 0.8
C2 58.9 6 2.9 12.9 6 1.1 26.8 6 1.2
C4 65.0 6 3.0 13.0 6 1.2 30.8 6 1.6
C5 64.5 6 2.6 16.7 6 2.7 29.4 6 1.3
C6 58.2 6 1.7 20.5 6 1.3 26.1 6 1.8
C7 63.8 6 1.0 27.4 6 1.5 24.8 6 0.8
T1 59.7 6 2.5 69.1 6 2.7 234.9 6 2.5 21.8 6 1.0
T2 54.0 6 1.8 75.5 6 4.2 236.2 6 3.5 20.4 6 1.1
T3 52.1 6 1.0 81.3 6 3.1 239.3 6 2.0 19.9 6 1.0
T4 50.0 6 1.7 79.3 6 3.5 240.1 6 2.6 19.2 6 0.6
T5 48.5 6 1.0 75.8 6 3.7 242.4 6 2.2 17.9 6 0.9
T6 47.0 6 2.0 70.9 6 3.3 244.3 6 2.0 18.3 6 0.3
T7 47.0 6 2.2 68.9 6 4.4 244.3 6 2.0 18.0 6 0.5
T8 48.2 6 2.1 62.0 6 2.3 241.2 6 2.9 19.3 6 1.2
T9 48.5 6 2.5 56.6 6 3./1 240.8 6 2.9 19.2 6 1.0
T10 49.2 6 0.8 43.8 6 2.0 231.8 6 2.0 19.7 6 0.4
T11 49.5 6 2.3 35.7 6 2.0 220.5 6 1.5 19.6 6 0.7
T12 51.5 6 2.4 30.2 6 2.1 0.4 6 3.5 21.0 6 1.1
T13 56.3 6 2.8 28.8 6 1.7 15.3 6 2.0 23.8 6 1.3
L1 102.0 6 0.0 29.0 6 1.4 20.0 6 1.4 46.0 6 0.0 17.0 6 2.8
L2 121.2 6 2.5 32.2 6 2.8 15.6 6 1.2 54.5 6 2.4 19.9 6 1.2
L3 131.4 6 4.5 30.7 6 1.5 13.8 6 2.0 60.9 6 3.3 19.6 6 4.0
L4 136.4 6 3.9 29.7 6 1.7 14.7 6 1.4 63.8 6 2.7 18.9 6 2.7
L5 138.7 6 3.5 29.2 6 1.9 15.1 6 1.1 63.8 6 2.9 19.3 6 3.2
L6 140.3 6 4.3 28.1 6 2.0 13.6 6 1.8 62.8 6 3.7 19.3 6 2.8
L7 116.4 6 5.7 27.0 6 1.7 11.2 6 1.3 50.5 6 2.1 26.3 6 4.7

TABLE 5. Dimensions and orientation of the cranial joint surfaces of the sheep
(mean 6 S.D. in mm and °)
Vertebra FCW mm FCH mm CAX ° CAY ° IFW mm
C2
C3 12.6 6 1.4 14.1 6 1.9 264.7 6 3.4 237.8 6 5.1 29.1 6 1.2
C4 14.1 6 1.0 13.8 6 1.3 271.1 6 4.7 236.6 6 4.3 29.9 6 1.8
C5 12.5 6 1.2 13.2 6 1.0 272.0 6 3.5 230.5 6 3.3 30.8 6 1.9
C6 12.2 6 0.8 14.2 6 1.6 268.3 6 2.7 222.1 6 2.7 32.3 6 1.4
C7 11.9 6 0.7 15.3 6 1.7 263.8 6 4.1 219.0 6 2.6 33.7 6 1.4
T1 12.9 6 1.1 11.8 6 1.8 282.4 6 4.0 230.3 6 4.7 30.7 6 1.1
T2 8.4 6 0.8 11.8 6 1.2 284.5 6 3.9 16.0 6 2.3 12.0 6 0.9
T3 8.0 6 1.2 11.6 6 1.3 283.1 6 2.5 18.6 6 2.6 11.2 6 0.8
T4 7.0 6 1.1 10.2 6 1.6 284.8 6 2.2 10.8 6 3.0 9.6 6 1.0
T5 5.4 6 1.0 9.4 6 1.6 284.9 6 2.1 12.3 6 2.6 9.2 6 1.4
T6 4.6 6 0.4 10.2 6 1.0 280.9 6 3.01 1.3 6 2.1 8.2 6 0.8
T7 4.1 6 0.7 10.4 6 1.0 278.1 6 4.2 9.4 6 1.4 8.7 6 0.8
T8 3.9 6 0.4 10.9 6 0.9 277.3 6 3.6 9.6 6 2.5 9.1 6 0.8
T9 4.0 6 0.7 13.3 6 1.0 280.1 6 3.5 12.4 6 1.9 9.5 6 1.1
T10 4.9 6 0.7 14.1 6 2.0 280.0 6 4.6 11.1 6 4.4 10.9 6 1.0
T11 5.1 6 0.8 15.8 6 1.6 279.1 6 4.6 16.1 6 3.4 11.5 6 0.4
T12 8.3 6 0.9 11.5 6 1.7 14.5 6 0.9
T13 9.3 6 0.9 11.9 6 1.3 16.1 6 1.0
L1 12.8 6 1.1 14.3 6 0.4 18.8 6 1.8
L2 12.4 6 0.9 13.9 6 1.3 20.5 6 0.7
L3 13.0 6 0.8 14.0 6 1.2 22.0 6 1.1
L4 13.6 6 1.1 15.0 6 1.6 23.1 6 1.7
L5 14.1 6 1.3 13.2 6 1.5 25.1 6 1.6
L6 13.6 6 1.2 13.8 6 1.0 27.2 6 1.4
L7 14.3 6 1.4 15.2 6 2.0 30.6 6 1.1

mm, and somewhat larger again in the lumbar region, spine. The small standard deviations encourage their
between 4.2 6 0.4 and 4.5 6 0.4 mm (Table 6). use as a model where appropriate to overcome the
inconclusiveness of studies in which comparative behav-
DISCUSSION ior among study groups is subtle, and thus the inherent
The results of this study provide a database of the scatter in the data for each group may obscure real
anatomy of the 3- to 4-year-old female merino sheep differences in treatment effect or other variables.
 SHEEP SPINE ANATOMY 547
TABLE 6. Anterior disc height of the sheep spine, with pedicle and vertebra height actually being
(mean 6 S.D. in mm) more closely correlated in the sheep than in the human
Disc IDHa
spine.
A better correlation is found between the two species
C2–3 6.7 6 0.8 in pedicle width (Fig. 6), where lumbar and cervical
C3–4 6.9 6 0.8 results are each largely overlapping, although diver-
C4–5 7.2 6 0.6 gence is seen in the mid-thoracic region.
C5–6 6.8 6 0.6
C6–7 7.2 6 0.8 Spinal Canal
C7–T1 5.3 6 0.4
T1–2 4.5 6 0.8 The trends of the two species in spinal canal width
T2–3 3.3 6 0.3 are nearly identical, although they differ in value,
T3–4 2.8 6 0.3 again mostly toward the extremes (Figs. 7 & 8). Rela-
T4–5 2.7 6 0.3 tive consistency of the depth interregionally occurs in
T5–6 2.6 6 0.2
T6–7 2.6 6 0.2
both species.
T7–8 2.6 6 0.2 Spinous and Transverse Processes
T8–9 2.7 6 0.3
T9–10 2.8 6 0.3 The spinous process in the cervical and thoracic
T10–11 3.0 6 0.4 regions is 2–3 times longer in the sheep, with a notable
T11–12 3.2 6 0.3 difference of more than 50 mm in the upper thoracic
T12–13 3.7 6 0.4 region (Fig. 9). However, in the lumbar spine, the sheep
T13–L1 4.3 6 0.4
L1–2 4.4 6 0.2
and human spinous processes are about the same
L2–3 4.2 6 0.4 length. Conversely, the tip-to-tip transverse process
L3–4 4.5 6 0.4 width is very similar between the two species in the
L4–5 4.3 6 0.3 cervical and thoracic regions but diverges somewhat in
L5–6 4.5 6 0.5 the lumbar spine (Fig. 10).
L6–7 4.5 6 0.0
Facet Joints
In the cervical spine, the facet orientation in sheep is
The following sections provide a structure-by-struc- consistently dorsolateral and leans somewhat forward,
ture anatomical comparison of the sheep and human whereas in the human it is on average directly lateral
spine, which may be helpful for a given study to suggest and leans farther forward (Figs. 11 & 12). In the
the appropriateness of using the sheep spine as a model thoracic spine, the transition is most abrupt for the
for the human spine. sheep spine, with the facets turned somewhat anterolat-
erally and standing almost vertically. The human facets
Vertebral Bodies likewise are more vertical than in the cervical region
A fundamental difference between the species is that and have virtually the same slightly anterior orienta-
the human vertebra is characteristically wider than tion in the transverse plane as those of the sheep. From
tall, by as much as a factor of 2, whereas the sheep the mid-thoracic region to the thoracolumbar junction,
vertebra is taller than wide, especially in the cervical the two species become almost identical in facet orienta-
spine (Figs. 2–4). However, for both species, vertebra tion. Comparative data are not available for the lumbar
width is greater than depth, thus producing a typically region due to the large curvature and uncertainty in the
oval shape that is most pronounced in the lumbar measurement of the sheep facet angles.
region. In the sheep spine, height, width, and depth are Facet height and width mostly overlap between the
greater in the cervical spine than in the other regions, sheep and human, except for width in the thoracic
in contrast to the human spine, where these dimensions region, which is considerably less in sheep (Figs. 13 &
are the least; height is also greater in the sheep than in 14). Variation in width from the cervical to thoracic
the human spine in the thoracic and lumbar regions, region is somewhat stronger in the sheep but not as
but width and depth are less. strong through the lumbar region. Interfacet distance
Another difference is the curvature in the lumbar followed the same trend in the sheep and human, with
spine, which for the sheep is slightly kyphotic rather smaller dimensions but larger transitions for the sheep
than lordotic. The atlas and axis of the two species are (Fig. 15). This trend follows the result of larger width in
also fundamentally different. However, these dimen- general in the human vertebrae.
sions are not included to avoid detracting from the more
Discs
characteristic comparisons of the vertebra.
Anterior disc height in the sheep lumbar spine is as
Pedicles much as 5 mm lower than the 11-16 mm height in
Pedicle height comparison (Fig. 5) is qualitatively humans (Nachemson et al., 1979; Green et al., 1993).
similar to that of vertebral height, and the differences This relationship is reversed in the cervical spine,
are likewise most pronounced in the cervical spine. The where the 5–7 mm sheep disc height is 2–3 mm greater
trend through the thoracic spine is comparable between than that in humans (Moroney et al., 1988).
the two species, but the lower transition to a decreasing
pedicle height, which occurs at T11–T12 in the human CONCLUSIONS
spine, does not occur until L5 in the sheep spine. Thus, Similarities in the major dimensions of the sheep and
the differences again become exaggerated in the lumbar human spine overall are strongest in the thoracic and
 Fig. 2. Posterior vertebral body height (VBHp) of the sheep spine from C2 to L7 (mean 6 S.D.) in
comparison with reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.

Fig. 3. Caudal endplate width (EPW) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with
reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.
 Fig. 4. Caudal endplate depth (EPD) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with
reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.

Fig. 5. Pedicle height (PDH) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with reported
values for the human spine from C2 to L5. Vertebrae images are at equal scale.
 Fig. 6. Pedicle width (PDW) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with reported
values for the human spine from C2 to L5. Vertebrae images are at equal scale.

Fig. 7. Spinal canal width (SCW) of the of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison
with reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.
 Fig. 8. Spinal canal depth (SCD) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with
reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.

Fig. 9. Spinous process length (SPL) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with
reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.
 Fig. 10. Transverse process width (TPW) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison
with reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.

Fig. 11. Joint surface inclination relative to the coronal plane CAX (adapted from Panjabi et al., 1993) of
the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with reported values for the human spine from
C2 to L5. Vertebrae images are at equal scale.
 Fig. 12. Joint surface inclination relative to a vertical plane, CAY (adapted from Panjabi et al., 1993) of
the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with reported values for the human spine from
C2 to L5. Vertebrae images are at equal scale.

Fig. 13. Facet height (FCH) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with reported
values for the human spine from C2 to L5. Vertebrae images are at equal scale.
 Fig. 14. Facet width (FCW) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with reported
values for the human spine from C2 to L5. Vertebrae images are at equal scale.

Fig. 15. Interfacet width (IDH) of the sheep spine from C2 to L7 (mean 6 S.D.) in comparison with
reported values for the human spine from C2 to L5. Vertebrae images are at equal scale.
 SHEEP SPINE ANATOMY 555
lumbar regions. The strongest difference in trend is in approaches for lumbar burst fractures using short-segment instru-
vertebral body height, which is greatest in the cervical mentation. Spine, 18:977–982.
Moore, R.J., O.L. Osti, B. Vernon-Roberts, and R.D. Fraser 1992
spine in sheep but in the lumbar spine in humans. Changes in endplate vascularity after an outer anulus tear in the
Nevertheless, regional trends are similar in most mea- sheep. Spine, 17:874–878.
surements. Moroney, S.P., A.B. Schultz, J.A.A. Miller, and G.B.J. Andersson 1988
These results provide data that may be helpful to Load-displacement properties of lower cervical spine motion
segments. J. Biomech., 21:769–779.
plan future studies that contemplate the use of sheep as Nachemson, A., A.B. Schultz, and M.H. Berkson 1979 Mechanical
a model for the human spine. Regarding spinal implant properties of human lumbar spine motion segements—Influences
tests, these results suggest that the sheep may be a of age, sex, disc level, and degeneration. Spine, 4:1–8.
reasonable anatomical model for instrumentation affect- Nagel, D.A., P.C. Kramers, B.A. Rahn, J. Cordey, and S.M. Perren
1991 A paradigm of delayed union and nonunion in the lumbosa-
ing the thoracic and lumbar regions. cral joint—A study of motion and bone grafting of the lumbosacral
spine in sheep. Spine, 16:553–559.
ACKNOWLEDGMENTS Nickel, R., A. Schummer, and E. Seiferle 1984 Lehrbuch der Anatomie
der Haustiere I. Verlag Paul Parey, Berlin und Hamburg.
We specially thank Herrn Albert Aigner for the Osti, O.L., B. Vernon-Roberts, and R.D. Fraser 1990 Anulus tears and
preparation of the sheep spines. The maceration of our intervertebral disc degeneration. An experimental study using an
exemplar specimen was done by Gustav Reiter. This work animal model. Spine, 15:431–435.
Panjabi, M.M., J. Duranceau, V. Goel, T. Oxland, and K. Takata 1991a
was supported by the University Hospital of Ulm (P.272). Cervical human vertebrae—Quantitative three-dimensional
anatomy of the middle and lower regions. Spine, 16:861–869.
LITERATURE CITED Panjabi, M.M., K. Takata, V. Goel, D. Federico, T. Oxland, J. Du-
ranceau, and M. Krag 1991b Thoracic human vertebrae—
Ahlgren, B.D., M.S. Vasavada, R.S., Brower, C. Lydon, M.D. Herkow- Quantitative three-dimensional anatomy. Spine, 16:888–901.
itz, and M.M. Panjabi 1994 Anular incision technique on the Panjabi, M.M., V. Goel, T. Oxland, K. Takata, J. Duranceau, M. Krag,
strength and multidirectional flexibility of the healing interverte- and M. Price 1992 Human lumbar vertebrae—Quantitative three-
bral disc. Spine, 19:948–954. dimensional anatomy. Spine, 17:299–306.
Ahmed, A.M., N.A. Duncan, and D.L. Burke 1990 The effect of facet Panjabi, M.M., T. Oxland, K. Takata, V. Goel, J. Duranceau, M. Krag
geometry on the axial torque-rotation response of lumbar motion 1993 Articular facets of the human spine. Quantitative three-
segments. Spine, 15:391–401. dimensional anatomy. Spine, 18:1298–1310.
Ashman, R.B., J.E. Bechtold, W.T. Edwards, C.E. Johnston, P.C. Putz, R. 1981 Funktionelle Anatomie der Wirbelgelenke. In: Normal
McAfee, and A.F. Tencer 1989 In vitro spinal arthrodesis implant and Pathological Anatomy, vol. 43. Doerr, W., and H. Leonhardt,
mechanical testing protocols. J. Spinal Dis., 2:73–80. eds. Stats Uni, Pilezhausen.
Berry, J.L., J.M. Moran, W.S. Berg, and A.D. Steffee 1987 A morphomet- Scoles, P.V., A.E. Linton, B. Latimer, M.E. Levy, and B.F. Digiovanni
ric study of human lumbar and selected thoracic vertebrae. Spine, 1988 Vertebral body and posterior element morphology: the
12:362–367. normal spine in middle life. Spine, 13:1082–1085.
Cain, C.C.J.M., and R.D. Fraser 1995 Bony and vascular anatomy of Slater, R., D. Nagel, and R.L. Smith 1988 Biochemistry of fusion mass
the normal cervical spine in the sheep. Spine, 20:759–765. consolidation in the sheep spine. J. Orthoped. Res., 6:138–144.
Cotterill, P.C., J.P. Kostuik, G. D’Angelo, G.R. Fernie, and B.E. Maki Tominaga, T., C.A. Dickman, V.K.H. Sonntag, and S. Coons 1995
1986 An anatomical comparison of the human and bovine thoraco- Comparative anatomy of the baboon and the human cervical
lumbar spine. J. Orthoped. Res., 4:298–303. spine. Spine, 20:131–137.
Doherty, B.J., and M.H. Heggeness 1994 The quantitative anatomy of Van Schaik, J.P.J., H. Verbiest, and F.D.J. Van Schaik 1985 The
the atlas. Spine, 19:2497–2500. orientation of laminae and facet joints in the lower lumbar spine.
Doherty, B.J., and M.H. Heggeness 1995 Quantitative anatomy of the Spine, 10:59–63.
second cervical vertebra. Spine, 20:513–517. Vazquez-Seonae, P., J. Yoo, D. Zou, L.A. Fay, B.E. Fredrickson, J.C.
Edmonston, S.J., K.P. Singer, R.E. Day, P.D. Breidahl, and R.I. Price Handal, H.A. Yuan, and W.T. Edwards 1993 Interference screw
1994 Formalin fixation effects on vertebral bone density and fixation of cervical grafts—A combined in vitro biomechanical and
failure mechanics: An study of human and sheep vertebrae. Clin. in vivo animal study. Spine, 18:946–954.
Biomech., 9:175–179. Wilke, H.-J., G. Ostertag, and L. Claes 1994 Three-dimensional
Eggli, S., F. Schläpfer, M. Angst, P. Witschger, and M. Aebi 1992 goniometer linkage system for the analysis of movements with six
Biomechanical testing of three newly developed transpedicular degrees of freedom. Biomed. Tech., 39:149–155.
multisegmental fixation systems. Eur. Spine J., 1:109–116. Xu, R., M.C. Nadaud, N.A. Ebraheim, and R.A. Yeasting 1995 Morphol-
Francis, C.C. 1955 Dimensions of the cervical vertebrae. Anat. Rec., ogy of the second cervical vertebra and the posterior projection of
122:603–609. the C2 pedicle axis. Spine, 259–263.
Green, T.P., M.A. Adams, and P. Dolan 1993 Tensile properties of the Yamamuro, T., J. Shikata, H. Okumura, T. Kitsugi, Y. Kakutani, T.
annulus fibrosus. Ii. Ultimate tensile strength and fatigue life. Matsui, and T. Kokubo 1990 Replacement of the lumbar vertebrae
Eur. Spine J., 2:209–214. of sheep with ceramic prostheses. J. Bone Joint Surg., 72-B:889–
Gunzburg, R., R.D. Fraser, R. Moore, and B. Vernon-Roberts 1993 An 893.
experimental study comparing percutaneous discectomy with Zindrick, M.R., L.L. Wiltse, A. Doornik, E.H. Widell, G.W. Knight, A.G.
chemonucleolysis. Spine, 18:218–226. Patwardhan, J.C. Thomas, S.L. Rothmann, and B.T. Fields 1987
Gurwitz, G.S., J.M. Dawson, M.J. McNamara, C.F. Federspiel, and Analysis of morphometric characteristics of the thoracic and the
D.M. Spengler 1993 Biomechanical analysis of three surgical lumbar pedicles. Spine, 12:160–166.