Computed tomographic arthrography of the stifle for detection of cranial and caudal cruciate ligament and meniscal tears in dogs.

Abstract

The purpose of this study was to evaluate the utility of single-detector computed tomographic arthrography (CT arthrography) for the diagnosis of cranial and caudal cruciate ligament and meniscal lesions in the dog stifle. Four normal and 25 abnormal stifle joints, determined to have lesions related to intra-articular ligamentous insufficiency based on clinical history, orthopedic examination, and survey orthogonal radiographs, were imaged using a previously developed CT arthrography protocol. Surgery was performed immediately following the CT procedure. Three board-certified radiologists inexperienced at interpreting CT stifle arthrograms reviewed all CT studies independently, and then as a group, without knowledge of surgical or necropsy findings. Sensitivity, specificity, positive predictive value, and negative predictive value for determination of cranial and caudal cruciate and meniscal tears were calculated for each individual reviewer and based on group consensus. All reviewers identified the normal canine stifle joints imaged correctly. Reviewers did well in discriminating normal from torn cranial cruciate ligaments, with sensitivities of 96-100% and specificities of 75-100%. No reviewer correctly identified the solitary caudal cruciate ligament tear and specificity ranged from 89.3% to 100%. Reviewers were less adept at discriminating normal from torn meniscal fibrocartilage, with sensitivities of 13.3-73.3% and specificities of 57.1-100%. Interpretive accuracy improved slightly when consensus scores were compared with surgical findings. Single-detector CT arthrography may be useful for identifying pathology of the canine cruciate ligaments but is of limited value for assessing the menisci.

Full text

Page 1

COMPUTED TOMOGRAPHIC ARTHROGRAPHY OF THE STIFLE FOR
DETECTION OF CRANIAL AND CAUDAL CRUCIATE LIGAMENT AND
MENISCAL TEARS IN DOGS
VALERIE F. SAMII, JONATHAN DYCE, ANTONIO POZZI, WM. TOD DROST, JOHN S. MATTOON, ERIC M. GREEN,
MICHAEL P. KOWALESKI, AMY M. LEHMAN
The purpose of this study was to evaluate the utility of single-detector computed tomographic arthrography (CT
arthrography) for the diagnosis of cranial and caudal cruciate ligament and meniscal lesions in the dog stifle.
Four normal and 25 abnormal stifle joints, determined to have lesions related to intra-articular ligamentous
insufficiency based on clinical history, orthopedic examination, and survey orthogonal radiographs, were imaged
using a previously developed CT arthrography protocol. Surgery was performed immediately following the CT
procedure. Three board-certified radiologists inexperienced at interpreting CT stifle arthrograms reviewed all
CT studies independently, and then as a group, without knowledge of surgical or necropsy findings. Sensitivity,
specificity, positive predictive value, and negative predictive value for determination of cranial and caudal
cruciate and meniscal tears were calculated for each individual reviewer and based on group consensus. All
reviewers identified the normal canine stifle joints imaged correctly. Reviewers did well in discriminating normal
from torn cranial cruciate ligaments, with sensitivities of 96–100% and specificities of 75–100%. No reviewer
correctly identified the solitary caudal cruciate ligament tear and specificity ranged from 89.3% to 100%.
Reviewers were less adept at discriminating normal from torn meniscal fibrocartilage, with sensitivities of
13.3–73.3% and specificities of 57.1–100%. Interpretive accuracy improved slightly when consensus scores were
compared with surgical findings. Single-detector CT arthrography may be useful for identifying pathology of
the canine cruciate ligaments but is of limited value for assessing the menisci.
Ultrasound, Vol. 50, No. 2, 2009, pp 144–150.
Veterinary Radiology &
Key words: canine, CT, stifle.
Introduction
H
cranial cruciate ligament rupture is routinely made on the
basis of physical examination. Patients with suspected
cranial cruciate ligament dysfunction are also usually ra-
diographed to evaluate for osteoarthrosis, joint effusion, or
capsular thickening, but findings are nonspecific. Magnetic
resonance (MR) imaging is the modality of choice in
humans for assessing knee problems.2–5MR arthrography
was used in 11 military dogs and was successful in doc-
umenting a variety of abnormal findings of the cruciate
ligaments, menisci, and collateral ligaments.6
Using computed tomography (CT) arthrography, nor-
mal ligamentous structures of the canine stifle were easily
IND LIMB LAMENESS attributable to cranial cruciate
ligament injury is common in dogs.1A diagnosis of
identified and reconstructions were found to be helpful.7
The purpose of this study was to evaluate CT arthrography
for diagnosis of cruciate ligament and meniscal abnormal-
ities in the dog. We hypothesized that there would be sig-
nificant agreement between CT arthrography and surgical
findings, and that CT arthrography would be both sensitive
and specific in identifying cranial and caudal cruciate
ligament and meniscal fibrocartilage lesions.
Materials and Methods
Dogs determined clinically to have stifle ligamentous
dysfunction were evaluated. Dogs suspected of having
osteochondrosis, fracture, osteomyelitis, septic arthritis, or
neoplasia were not included. All dogs were cared for
according to the principles outlined by the Hospital Exec-
utive Committee of Veterinary Medicine Practice, The
Ohio Sate University. Owner-signed consent forms were
required before admittance into study.
Twenty-five stifles from 25 patients were evaluated. Age
ranged from 1.5 to 9.6 years (mean: 5.8 ? 1.9 years), and
weight from 12 to 54.5kg (mean: 35.7 ? 7.0kg). Breeds
included mixed breed dog (n¼12), Labrador Retriever
(n¼6), boxer (n¼2), Great Pyrenees (n¼2), and one each
Supported by a grant from The Ohio State University Canine Research
Funds.
Address correspondence and reprint requests to Valerie F. Samii,
DVM, at the above address. E-mail: samii.3@osu.edu
Received January 22, 2008; accepted for publication October 2, 2008.
doi: 10.1111/j.1740-8261.2009.01507.x
From the Department of Veterinary Clinical Sciences, College of Vet-
erinary Medicine (Samii, Dyce, Pozzi, Drost, Mattoon, Green, Kowaleski),
and the Center for Biostatistics (Lehman), The Ohio State University,
Columbus, OH 43210.
144

Page 2

of the following: Golden Retriever, Siberian husky, and
springer spaniel.
In addition to clinical patients, four normal stifle joints
from three cadaver specimens were evaluated. Age ranged
from 1.5 to 3 years (mean: 2.1 ? 0.4 years), and weight
from 12 to 23kg (mean: 20.3 ? 4.1kg). Breeds represented
included mixed breed dog (n¼2), Staffordshire terrier
(n¼1), and beagle (n¼1). Dogs were euthanized for re-
search unrelated to our study. Survey orthogonal radio-
graphs of each stifle were made to confirm the absence of
abnormal findings.
For CT imaging, each dog was in sternal recumbency
with the affected leg extended caudally. All data were col-
lected using a fourth-generation CT scanner.?A lateral
pilot image was made for planning. To standardize the
imaging plane between stifles, the gantry was positioned
parallel to the tibial plateau. One-millimeter transverse
contiguous prearthrography images were acquired from
just proximal to the patella to 2cm distal to the tibial
plateau. All scans were made using a bone algorithm,
125mAs, 130kVp, and a field of view of 90mm.
For arthrography, stifles were aseptically prepared for
synoviocentesis. A 22-gauge, 1-inch needle was inserted
into the joint medial to the mid-point of the patellar ten-
don. Synovial fluid was aspirated to ensure correct needle
placement. About 3–6ml of iohexol (150mgI/ml)w was in-
jected into the joint until palpable distension was achieved.
The joint was flexed and extended several times to facilitate
distribution of contrast medium throughout the joint. The
limb was repositioned on the CT couch as before and
the CT acquisition protocol repeated. Pre- and postarthro-
graphy CT pilot images were matched as closely as possible
in an attempt to ensure similar patient positioning between
studies.
Tibial plateau leveling osteotomies were performed in all
clinical patients. Surgery was performed immediately fol-
lowing the CT procedure. Arthroscopy or arthrotomy was
performed in each dog to assess intra-articular structures.
Surgeons (J.D., A.P., M.P.K.) were provided with a score
sheet and asked to assess the integrity of the cranial and
caudal cruciate ligaments (normal, torn). Surgeons were
also asked to assess the integrity of the menisci (normal,
medial tear, lateral tear, medial and lateral tears). Gross
examination confirmed joint normalcy in all four cadaver
specimens.
Three board-certified radiologists (W.T.D., E.M.G.,
J.S.M.), with little to no experience at interpreting CT sti-
fle arthrograms reviewed all CT studies independently and
without knowledge of ligament or meniscus status. Before
reviewing the studies, radiologists were provided with the
following instructional materials: four reference articles,7–10
seven normal canine CT stifle arthrograms, three abnormal
canine CT stifle arthrograms from dogs not included in the
study which included examples of partial and complete
cranial cruciate ligament tears and medial meniscal tears.
Instructional images were provided to each radiologist in
DICOM format; the viewing software was at the discretion
of the radiologist.
Radiologists were provided with a score sheet and asked
to assess the integrity of the cranial and caudal cruciate
ligaments (normal or torn) and the meniscal fibrocartilage
(normal, medial tear, lateral tear, medial and lateral tears).
The presence of joint effusion/capsular thickening and
osteoarthrosis (none, mild, moderate, severe) was graded
according to the following criteria. For joint effusion/cap-
sular thickening: Mild—no caudal joint pouch distension.
Increased soft tissue opacity in joint does not extend cra-
nial to margin of extensor fossa. Infrapatellar fat pad
still visible. Moderate—caudal joint pouch distension. In-
creased soft tissue opacity in joint extends cranial to
extensor fossa. Infrapatellar fat pad still visible. Severe—
caudal joint pouch distension. Increased soft tissue opacity
in joint extends cranial to extensor fossa. Infrapatellar fat
pad not visible.
Forosteoarthrosis11,12:
remodeling along trochlear ridges, lateral and medial epi-
condyles, and lateral tibial condyle. Superficial remodeling/
irregularity to apex of the patella. Moderate—plaque-like
proliferative remodeling along trochlear ridges. Prolifera-
tive remodeling on the lateral and medial epicondyles, lat-
eral and medial tibial condyle, tibial plateau, fabellae, and
intercondyloid fossa. Osteophytes (1–2mm) on apex of the
patella. Intracapsular or capsular mineralization may be
apparent. Severe—plaque-like
along trochlear ridges. Proliferative remodeling on the
lateral and medial epicondyles, lateral and medial tibial
condyle, tibialplateau, fabellae,
fossa. Osteophytes (42mm) on apex of the patella. Intra-
capsular or capsular mineralization may be apparent. Ir-
regularity to articular margins and subchondral sclerosis
may be evident.
Radiologists were asked to note their preferred window
width and level settings for image viewing. They were also
asked to comment on the necessity of performing multi-
planar reconstructions for diagnosing cranial and caudal
cruciate ligament and meniscal tears. Radiologists were
given 8 weeks to evaluate the CT examinations, with no
limitations on repeated viewing. A consensus agreement
was also reached in those dogs where individual assess-
ments differed in regards to the cranial and caudal cruciate
ligaments and meniscal fibrocartilage.
Sensitivity, specificity, and positive and negative predic-
tive values, for determination of cranial and caudal cruci-
ate and meniscal fibrocartilage tears (negative or positive)
as compared with surgical or necropsy outcome, were
Mild–irregularproliferative
proliferativeremodeling
andintercondyloid
?Picker PQS, Philips Medical Systems, N.A., Bothell, WA.
wOmnipaque
s, Nycomed Inc., Princeton, NJ.
145
CTARTHROGRAPHY OF THE STIFLE IN DOGS
Vol. 50, No. 2

Page 3

calculated for each individual reviewer and as a group
consensus. Kappa statistics were performed to assess ra-
diologist agreement with respect to joint effusion/capsular
thickening and osteoarthrosis using the following scale13:
below 0: poor; 0–0.2: slight; 0.21–0.40: fair; 0.41–0.60:
moderate; 0.61–0.80: substantial; 0.81–1.00: almost perfect.
All analyses were preformed using SAS v9.1.3z and Stat-
aSE 9.y
Results
Each reviewer had a different preferred window for view-
ing (reviewer 1: WW¼1500, WL¼300; reviewer 2: WW¼
3000, WL¼400; reviewer 3: WW¼2700, WL¼600).
Complete or partial cranial cruciate tears were detected
in all 25 clinical patients at surgery. Reviewers did well in
discriminating normal from torn cranial cruciate ligaments,
with sensitivities of 100%, 100%, and 96% and specificities
of 100%, 100%, and 75% for reviewers 1, 2, and 3, re-
spectively (Fig. 1). A partial caudal cruciate tear was iden-
tified at the time of surgery in only one patient, and the
caudal cruciate ligament in all other dogs was normal. All
reviewers diagnosed the torn caudal cruciate ligament as
being normal (sensitivity 0%). Reviewers 1, 2, and 3
diagnosed torn caudal cruciate ligaments in 3, 0, and 1
normal dog, respectively, with calculated specificities of
89.3%, 100%, and 96.4%. Of the 29 stifles imaged, 14 dogs
had medial meniscal tears and the remaining 15 dogs had
normal meniscal fibrocartilage. A lateral meniscal tear
was not diagnosed in any dog at surgery; however, on
consensus review of the CT images, a lateral meniscal tear
was incorrectly diagnosed. Reviewers were less adept
at discriminating normal from torn meniscal fibrocarti-
lage, with sensitivities of 66.7%, 13.3%, and 73.3%
and specificities of 64.3%, 100%, and 57.1% for review-
ers 1, 2, and 3, respectively. Positive and negative predictive
values for individual reviewer scores are presented in
Table 1.
Based on consensus evaluation, the interpretive accuracy
for diagnosing torn cranial cruciate ligaments increased to
100% sensitivity and specificity. Sensitivity remained at 0%
and specificity was 96.4% for diagnosis of caudal cruciate
tears. Sensitivity and specificity for meniscal tears was
64.3% and 73.3%, respectively (Figs. 2 and 3). All normal
stifles (n¼4) were correctly classified as such by all
reviewers. Positive and negative predictive values for
consensus scores are presented in Table 1. Agreement be-
tween reviewers for the presence of effusion (k¼0.4311)
and osteoarthrosis (k¼0.4237) was moderate.
Discussion
CT arthrography proved sensitive and specific for the
diagnosis of cranial cruciate ligament tears but was less so
for diagnosing meniscal tears. There were several limita-
tions to this study that likely affected the statistical out-
come. These limitations were due to the numbers of dogs
represented in the various lesion categories, imaging equip-
ment used, image postprocessing limitations, inconsistent
contrast medium distribution in osteoarthritc joints, and
reviewer’ inexperience.
Twenty-nine dogs, including four normal cadaver spec-
imens, were included in the study population. The intent of
adding the four normal dogs was to assess the reviewers’
abilities to correctly identify normal stifles in a population
of clinically abnormal dogs, which they all did. All clinical
patients not only had tears of the cranial cruciate ligament,
but also varying degrees of stifle osteoarthrosis. It could be
argued that the presence of joint effusion and/or periartic-
ular new bone production may have biased the reviewers
and affected their assessment of cruciate or meniscal
lesions. Only one clinical patient was confirmed as having a
caudal cruciate ligament tear at the time of surgery.
Because all reviewers incorrectly interpreted the caudal
cruciate ligament as being normal, overall reviewer sensi-
tivity was 0%. This result must be interpreted with caution
due to the low sample size.
Examinations were performed using a single-detector
CT scanner. Because of scanner limitations and the thin
slice (1mm) algorithm used, helical acquisitions were not
possible and a conventional acquisition protocol was used.
In a study performed on normal canine stifles using a
nearly identical protocol, multiplanar reconstructions were
helpful in identifying all intra-articular structures.7In the
present study, the added presence of joint disease suffi-
ciently reduced the resolution of multiplanar reconstructed
images, impeding identification of meniscal tears. Review-
ers reported that, with the exception of meniscal lesions,
most diagnoses were made from acquired transverse
images only. Dorsal plane CT arthrography of the canine
stifle following simulated meniscal injuries was assessed in
a cadaver model.14CT arthrography was 90% sensitive
and 100% specific for identifying meniscal lesions. It is
likely that identification of naturally occurring meniscal
lesions may have improved using a dorsal scanning
approach, though the coexistance of degenerative joint
disease may still have impeded detection of tears in some
dogs. Dual-detector helical CT arthrography is a highly
sensitive and specific means of evaluating knee intra-artic-
ular ligamentous disease in people.8–10Dual-detector
helical technology uses submillimeter collimation in a dual-
detector system, allowing for thinner slice acquisitions as
compared with single-detector systems. Because of thinner
slice capabilities and the ability to use helical technology,
zSAS Institute, Inc., Cary, NC.
yStataCorp, College Station, TX.
146
SAMII ET AL.
2009

Page 4

multiplanar reconstruction detail is much improved as
compared with conventional CT image acquisitions. Dual-
detector helical CT arthrography is as sensitive and specific
as MR in identifying knee intra-articular ligamentous
pathology.9We suspect that the use of dual-detector helical
CT technology in imaging canine stifles with intra-articular
disease will result in improved multiplanar image resolu-
tion, and hence, better accuracy.
In a CT arthrogram study conducted on normal canine
stifle joints, an intra-articular volume of 0.3ml iodinated
contrast medium (150mgI/ml) per cm thickness of the
joint, from medial to lateral, was deemed of sufficient vol-
ume and opacity.7We determined this volume to be inad-
equate in the clinical patients and found that injection of
contrast medium until palpable joint pouch distension was
a better indicator of adequate contrast medium volume
within the joint. Still, even and adequate distribution of
contrast medium within the joint was not attained in all
dogs. This compromised ligamentous and meniscal defini-
tion, impeding differentiation between these structures and
Fig. 1. One-millimeter transverse computed tomographic arthrography images at the level of the intercondyloid fossa of a normal stifle joint (A), a stiflejoint
with a partial cranial cruciate tear (B), and a stifle joint with a complete cranial cruciate tear (C). The caudal cruciate (a), cranial cruciate (b), and
meniscofemoral (c) ligaments and long digital extensor tendon (d) are labeled when identified. Only a small strand of the cranial cruciate ligament (arrow) can
be identified in image B. The cranial cruciate ligament is not identified in image C. M, medial; L, lateral.
Table1. Positive Predictive Value (PPV) and Negative Predictive Value (NPV) for Individual Reviewers and Consensus Score in Diagnosing ‘‘Tear’’ or
‘‘No Tear’’ for the Cranial and Caudal Cruciate Ligaments and Meniscal Fibrocartilage
Reviewer 1Reviewer 2Reviewer 3Consensus
PPV (%) NPV (%)PPV (%) NPV (%)PPV (%)NPV (%)PPV (%)NPV (%)
CrCL
CdCL
Men
100100
96.2
64.3
100100
96.6
51.6
96
0
64.7
75
96.4
66.7
100100
96.4
68.8
000
66.7 10069.2
CrCL, cranial cruciate ligament; CdCL, caudal cruciate ligament; Men, meniscus.
147
CTARTHROGRAPHY OF THE STIFLE IN DOGS
Vol. 50, No. 2

Page 5

intracapsular effusion.
distribution within joints was likely due to the presence
of intra-articular fibrosis in more chronically affected
joints. We suspect this to be the reason for failure in di-
agnosing the solitary caudal cruciate ligament tear in this
study. This tear was located at the origin of the caudal
cruciate ligament in the intercondyloid fossa. The severity
of joint osteoarthrosis was deemed moderate by all re-
viewers. On later review of this particular patient, inter-
condyloidstenosiswasalso
retrospective evaluation of all joints, there was no
Inconsistentcontrastmedium
noted. Onsubjective
correlation between the severity of joint osteoarthrosis
and pattern of intra-articular contrast distribution.
Although an attempt was made to remove all joint effu-
sion before contrast medium injection, this was not always
achieved, resulting in dilution of contrast medium and
reduced intra-articular contrast resolution. Images of all
dogs in this study were considered of diagnostic quality,
though some had greater intra-articular contrast resolution
and distribution than others.
Although criteria for grading the severity of joint effu-
sion and osteoarthrosis were well established, only
Fig. 2. One-millimeter transverse computed tomographic (CT) and CT arthrography images of a normal stifle joint at the level of the menisci (A, B) and an
osteoarthritic stifle joint with a Type 1 (folded caudal horn) medial menscal tear (C–E). Contrast medium outlines the folded caudal horn of the medialmeniscus
(arrows) in image D. A vertically oriented accumulation of contrast medium (arrow) is identified at the caudal medial horn of the meniscus on the sagittal
reconstructed image (E). Two of the three reviewers correctly identified the presence of a medial meniscal tear and the consensus ruling was torn medial
meniscus. M, medial; L, lateral; a, intercondyloid eminence of the tibia; b, long digital extensor tendon; c, periarticular new bone formation.
148
SAMII ET AL.
2009

Page 6

moderate agreement was reached between reviewers. A
plausible explanation for this result is not obvious. It is
possible that additional subjective appraisal made by each
reviewer influenced their interpretation.
None of the reviewers had any prior experience inter-
preting CT arthrography studies of the canine stifle.
Reviewers were provided with reference material to famil-
iarize themselves with the procedure before interpreting the
study population. They were also not restricted as to how
many times they could review patients or amend their
findings before submittal of their final report. Still, it is
likely that inexperience affected the performance nega-
tively. Improved accuracy was noted based on consensus
scoring of cranial cruciate tears. Only one dog was incor-
rectly assessed as having a caudal cruciate ligament tear on
consensus scoring. A false-positive diagnosis of a lateral
mensical tear was made on consensus review of the CT
images. A curvilinear collection of contrast medium was
observed at the caudoaxial margin of the lateral meniscus
on transverse images. On retrospective evaluation, this was
identified as contrast medium situated between the men-
iscofemoral ligament and caudal horn of the lateral me-
niscus. False-positive diagnoses of lateral meniscal tears
have been similarly made with MR imaging.15z
All clinical patients had arthrotomy or arthroscopy per-
formed, in conjunction with tibial plateau leveling osteo-
tomy, to confirm CT arthrography findings. Even if CT
findings matched surgical outcome, it is possible that the
lesions identified on imaging were not the same as those
described at surgery. Arthroscopy may be superior to ar-
throtomy for identifying meniscal lesions.16,17Arthroscopy
allows examination of the joint under magnification and
aqueous distension, which may be a profound advantage
over arthrotomy for assessing small tears or fissures.16
Also, arthroscopy enables improved visualization of the
lateral meniscus. However, the tibial surface of the menis-
cus is not readily seen with arthroscopy and lesions may be
missed in this location.15Most arthrotomies are performed
using either a craniomedial or caudomedial approach, se-
verely limiting the visual evaluation of the lateral meniscus.
Although surgical outcome was the gold standard in this
study, it is conceivable that small meniscal fissures may
have not been seen at surgery. It would be highly unlikely
that a lesion this small would have been identified with CT.
Three patients were diagnosed as having torn medial men-
isci on CT that were normal at surgery. False-positive
diagnoses may have been reached secondarily to poor res-
olution on reconstructed images or partial volume artifact
with either the medial femoral or tibial condyles.
Fig. 3. One-millimeter transverse and reconstructed dorsal plane images
of a Type 7 (transverse) tear of (A) the medial meniscus (arrow). (B) Two of
the three reviewers correctly identified a medial meniscal tear; however, the
consensus ruling was normal meniscus. The insertion of the caudal cruciate
ligament (?) on the tibial condyle is noted. M, medial; L, lateral; d, long
digital extensor tendon.
zMerge eFilm, Milwaukee, WI.
149
CTARTHROGRAPHY OF THE STIFLE IN DOGS
Vol. 50, No. 2

Page 7

Conclusion
Single-detector CT arthrography may be useful for iden-
tifying abnormalities of the canine cruciate ligaments but is
of questionable value for assessing the menisci. The use of
multidetector technology, enabling submillimeter slice
acquisitions, may improve the quality of dorsal and sag-
ittal reconstructions often used in cruciate and meniscal
tear assessment. Lack of interpreter experience in assessing
canine stifle CT arthrograms and poor contrast medium
distribution within some joints may have negatively
affected the statistical results.
ACKNOWLEDGMENTS
The authors would like to acknowledge the veterinary radiology
residents and CT technologists at The Ohio State University for their
technical assistance in this project.
REFERENCES
1. Vasseur PB. The stifle joint. In: Slatter DH (ed): Textbook of small
animal surgery, 2nd ed. Philadelphia: WB Saunders Co., 1993;1817–1866.
2. Crues JV III, Mink J, Levy TL, Lotysych M, Stoller DW. Meniscal
tears of the knee: accuracy of MR imaging. Radiology 1987;164:
445–448.
3. Mink JH, Levy T, Crues JV III. Tears of the anterior cruciate lig-
ament and menisci of the knee: MR imaging evaluation. Radiology
1988;167:769–774.
4. Rubin DA. MR imaging of the knee menisci. Radiol Clin North Am
1997;35:21–44.
5. Ha TP, Li KC, Beaulieu CF. Anterior cruciate ligament injury: fast
spin echo MR imaging with arthroscopic correlation in 217 examinations.
Am J Roentgenol 1998;170:1215–1219.
6. Banfield CM, Morrison WB. Magnetic resonance arthrography of the
canine stifle joint: technique and applications in eleven military dogs. Vet
Radiol Ultrasound 2000;41:200–213.
7. Samii VF, Dyce J. Computed tomographic arthrography of the
normal canine stifle. Vet Radiol Ultrasound 2004;45:402–406.
8. Vande Berg BC, Lecouvet FE, Poilvache P, Maldague B, Malghem J.
Spiral CT arthrography of the knee: technique and value in the assessment of
internal derangement of the knee. Eur Radiol 2002;12:1800–1810.
9. Vande Berg BC, Lecouvet FE, Poilvache P, Dubuc JE, Maldague B,
Malghem J. Anterior cruciate ligament tears and associated meniscal lesions:
assessment
2002;223:403–409.
10. Vande Berg BC, Lecouvet FE, Poilvache P, et al. Dual-detector
spiral CT arthrography of the knee: accuracy for detection of meniscal
abnormalities and unstable tears. Radiology 2000;216:851–857.
11. Tigari M. Changes in the canine stifle joint following rupture of the
anterior cruciate ligament. J Small Anim Pract 1971;19:17–26.
12. Tigari M, Vaughan LC. Arthritis of the canine stifle joint. Vet Rec
1975;96:394–399.
13. Landis JR, Koch GG. The measurement of observer agreement for
categorical data. Biometrics 1977;33:159–174.
14. Tivers MS, Mahoney P, Corr SA. Canine stifle positive contrast
computed tomography arthrography for the assessment of caudal horn
meniscal injury: a cadaver study. Vet Surg 2008;37:269–277.
15. Martig S, Konar M, Schmokel HG, et al. Low-field MRI and
arthroscopy of meniscal lesions in ten dogs with experimentally induced cra-
nial cruciate ligament insufficiency. Vet Radiol Ultrasound 2006;47:512–522.
16. Ralphs SC, Whitney WO. Arthroscopic evaluation of menisci in
dogs with cranial cruciate ligament injuries: 100 cases (1999–2000). J Am Vet
Med Assoc 2002;221:1601–1604.
17. Pozzi A, Hildreth BE, Rajala-Schultz PJ. Comparison of arthros-
copy and arthrotomy for the diagnosis of medial meniscal pathology: an ex
vivo study. Vet Surg (in press).
at dual-detectorspiralCT arthrography.Radiology
150
SAMII ET AL.
2009