Clinical value of fat-suppressed 3D volume isotropic spin-echo (VISTA) sequence compared to 2D sequence in evaluating internal structures of the knee

Abstract

Although many three-dimensional (3D) magnetic resonance imaging (MRI) sequences have been used in the clinical field, there are no studies on fat-suppressed (FS) 3D sequences for the diagnosis of knee abnormalities.
To evaluate the usefulness of FS 3D volume isotropic turbo spin echo acquisition (VISTA) imaging for diagnosis of internal structures of knee.
In 137 patients with 138 knee MRI examinations who had undergone both FS 3D VISTA MRI and two-dimensional (2D) MRI of the knee at 3.0 T, arthroscopic surgery was subsequently performed. Multiplanar reformations (MPR) of FS 3D VISTA images and their MPR sequences were compared with 2D MR images of axial T1-weighted (T1W) images, axial FS T2-weighted (T2W) images, coronal FS T2W images, and sagittal T2W images. Two musculoskeletal radiologists reviewed the images independently.
Arthroscopy revealed 33 anterior cruciate ligament tears, three posterior cruciate ligament tears, 42 lateral meniscus tears, and 68 medial meniscus tears. Five medial collateral ligament tears were clinically confirmed. Image acquisition time was shorter for FS 3D VISTA imaging than 2D imaging. There were no significant differences in diagnostic values between FS 3D VISTA and 2D imaging (P > 0.05). There were excellent inter-observer agreements for both FS 3D VISTA and 2D imaging (κ > 0.84).
FS 3D VISTA imaging could replace 2D imaging because of equal diagnostic ability and shorter scan time.
© The Foundation Acta Radiologica 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.

Full text

Original Article
Clinical value of fat-suppressed 3D volume
isotropic spin-echo (VISTA) sequence
compared to 2D sequence in evaluating
internal structures of the knee
Daekeon Lim
1,2
, Young Han Lee
1
, Sungjun Kim
1
,
Ho-Taek Song
1
and Jin-Suck Suh
1
Abstract
Background: Although many three-dimensional (3D) magnetic resonance imaging (MRI) sequences have been used in
the clinical field, there are no studies on fat-suppressed (FS) 3D sequences for the diagnosis of knee abnormalities.
Purpose: To evaluate the usefulness of FS 3D volume isotropic turbo spin echo acquisition (VISTA) imaging for diagnosis
of internal structures of knee.
Material and Methods: In 137 patients with 138 knee MRI examinations who had undergone both FS 3D VISTA MRI
and two-dimensional (2D) MRI of the knee at 3.0 T, arthroscopic surgery was subsequently performed. Multiplanar
reformations (MPR) of FS 3D VISTA images and their MPR sequences were compared with 2D MR images of axial T1-
weighted (T1W) images, axial FS T2-weighted (T2W) images, coronal FS T2W images, and sagittal T2W images. Two
musculoskeletal radiologists reviewed the images independently.
Results: Arthroscopy revealed 33 anterior cruciate ligament tears, three posterior cruciate ligament tears, 42 lateral
meniscus tears, and 68 medial meniscus tears. Five medial collateral ligament tears were clinically confirmed. Image
acquisition time was shorter for FS 3D VISTA imaging than 2D imaging. There were no significant differences in diagnostic
values between FS 3D VISTA and 2D imaging (P>0.05). There were excellent inter-observer agreements for both FS 3D
VISTA and 2D imaging (k>0.84).
Conclusion: FS 3D VISTA imaging could replace 2D imaging because of equal diagnostic ability and shorter scan time.
Keywords
Magnetic resonance imaging (MRI), knee injuries, three-dimensional, anterior cruciate ligament, posterior cruciate liga-
ment, menisci, tibial
Date received: 26 March 2014; accepted: 16 December 2014
Introduction
Two-dimensional (2D) fast spin-echo (FSE) sequence
magnetic resonance imaging (MRI) has been widely
used as the standard clinical MRI technique for evalu-
ation of the knee because of its favorable contrast and
signal-to-noise ratio. These sequences perform well in
evaluating joint abnormalities such as meniscal tears,
ligamentous injuries, and fibrocartilage damage (1–4).
Although these standard clinical MRI sequences have
been widely used to evaluate the internal structures of
the knee, they involve the use of relatively thick slice
1
Department of Radiology, Research Institute of Radiological Science,
Medical Convergence Research Institute, and Severance Biomedical
Science Institute, Yonsei University College of Medicine, Seoul, Republic
of Korea
2
Department of Radiology, Catholic Kwandong University International
St. Mary’s Hospital, Incheon, Republic of Korea
Corresponding author:
Jin-Suck Suh, Department of Radiology, Research Institute of Radiological
Science, Medical Convergence Research Institute, and Severance
Biomedical Science Institute, Yonsei University College of Medicine, 50
Yonsei-ro, Seodaemun-gu, Seoul, 120-752, Republic of Korea.
Email: jss@yuhs.ac
Acta Radiologica
2016, Vol. 57(1) 66–73
!The Foundation Acta Radiologica
2015
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0284185114567560
acr.sagepub.com
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

sections and interslice gaps that can lead to partial
volume artifacts. In addition, these sequences employ
a non-isotropic voxel that requires obtaining sequences
in several imaging planes such as axial, sagittal, cor-
onal, and sometimes obliquely planes for imaging the
anterior cruciate ligament (ACL) or the posterior cru-
ciate ligament (PCL). Performing several 2D scan
sequences takes a considerable amount of time.
Three-dimensional (3D) isotropic images have fewer
partial volume artifacts through thin contiguous sec-
tions and can be used in creating multiplanar reforma-
tions (MPR). Despite these advantages, utilization of
3D isotropic sequences in MRI has a limited clinical
significance because of the relatively long scan and
reconstruction times.
Gradient-recalled echo (GRE) sequences have
recently been introduced in 3D imaging for evaluating
cartilage; however, these sequences have a limited role
in imaging of other structures such as the menisci, liga-
ments, and bone marrow edema, which are sensitive to
susceptibility artifacts. Nevertheless, with the develop-
ment of parallel imaging methods – such as SENSE
(sensitivity encoding sequence) in Philips MRI, and
advanced workstation and postprocessing software –
evaluation of the knee by 3D isotropic FSE sequences
has recently become clinically feasible (3,5). Advanced
MRI sequence techniques and hardware can reduce the
image acquisition time and make patients more com-
fortable in the MRI scanner (6).
Several isotropic 3D FSE sequences have been intro-
duced into clinical practice. Studies have shown that the
diagnostic performance of these isotropic 3D images
compare favorably with that of conventional T2 FSE
images for evaluating cartilage, menisci, ligaments, and
bone marrow edema (7–9). Additionally, there have
been several reports of 3D isotropic data with MPR
made by gradient-recalled-echo sequences and spin-
echo (SE) sequences that have shown diagnostic per-
formance similar to that of conventional 2D sequences
in assessing internal derangement of the knee joint
(5–8). Recent studies have shown that the isotropic
3D images perform superiorly in assessing particular
knee lesions (10,11).
However, to our knowledge, there have been no stu-
dies published comparing the diagnostic performance
of fat-suppressed (FS) 3D FSE isotropic intermediate-
weighted (IM) MR images and 2D FSE images in the
diagnosis of knee abnormalities. The volume isotropic
turbo spin echo acquisition (VISTA, Philips Medical
Systems) imaging sequence was developed on the
basis of intermediate-weighted (IM) 3D isotropic FSE
MRI sequences. We hypothesized that the FS 3D
VISTA imaging at 3.0 T has the same or superior diag-
nostic performance in evaluation of the knee joint. The
purpose of our study was to evaluate the clinical value
of FS 3D VISTA imaging at 3.0 T MRI in the evalu-
ation of internal structures of the knee.
Material and Methods
Patients
After approval from the institutional review board and
receiving a waiver of consent for the retrospective
review, we reviewed all knee MRI examinations from
November 2010 to August 2011. Six hundred and forty-
six symptomatic patients underwent both conventional
2D and FS 3D VISTA imaging.
Of the 646 patients, 117 patients were excluded
because of a history of knee surgery. Finally, 137
patients who underwent arthroscopic surgery after the
MRI examination were enrolled with a total of 138
knee MRI studies, using arthroscopy results as the out-
come reference (Fig. 1). Our sample of 137 patients
included 59 men (age range, 16–66 years; mean age,
36.4 years) and 78 women (age range, 19–85 years;
mean age, 49.3 years) patients. All arthroscopic oper-
ations of the knee were performed by one of the two
Fig. 1. Schematic illustration of patient selection.
Lim et al. 67
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

experienced orthopedic surgeons who had 6 and 20
years of arthroscopy experience, respectively.
Magnetic resonance imaging
All MRI examinations were performed on a 3.0 T MR
system (Achieva, Philips Healthcare, Best, The
Netherlands) with a dedicated eight-element phase
array coil. The time interval between MR examination
and arthroscopic surgery was 1–50 days (mean, 40.4
days). The details of the MR protocol were the same
as those in Table 1 in the study by Lim et al. (11). The
imaging protocol consisted of 2D images, including
axial T1-weighted (T1W) images, axial FS T2-weighted
(T2W) images, coronal FS T2W images, and true sagit-
tal T2W images, which were obtained using conven-
tional 2D MR sequences of 3 mm thickness and
0.3 mm interslice gap. For FS 3D VISTA imaging, FS
isotropic sagittal 3D FSE IM MRI with a voxel size of
0.5 0.5 0.5 mm were obtained. Multiplanar refor-
matted axial, coronal, and oblique coronal images
along the axis of the ACL, with a slice thickness of
0.5 mm, were obtained by one radiologist immediately
after image acquisition using Aquaris NET 1.8.2.6 soft-
ware (TeraRecon, SanMateo, CA, USA) embedded in
the workstation in the reading room. The images were
sent to the picture archiving and communication system
(PACS) (Centricity Radiology RA 1000; GE
Healthcare, Milwaukee, WI, USA) server and saved.
Reformatted images were generated semi-automatically.
The total reformation time of the 3D images was within
3 min. When ligament injury was suspected during
review of the FS 3D VISTA images, reformatted oblique
images, according to the structure of interest, were also
generated by each reader using Aquaris NET 1.8.2.6
software (TeraRecon) embedded in the reading com-
puter. However, the oblique reformatted images along
the structures of interest generated by each reader were
not saved to avoid bias for the following reviewer.
Image analysis
Two musculoskeletal radiologists (LYH and LDK,
with 6 and 2 years of experience in musculoskeletal
imaging, respectively) retrospectively reviewed the
MR studies of each case by using two separate and
independent MRI image sets: conventional 2D images
and FS 3D VISTA images. They were blinded to the
original reports of the knee MRI and the arthroscopic
surgery. By using a PACS, the readers evaluated only
the conventional 2D MR images of each patient at the
first session. To reduce recall and learning bias, the
readers evaluated only the FS 3D VISTA images of
each patient 1 month later at a second session. The
2D and FS 3D VISTA images from each patient were
randomly presented at separate reading sessions. The
presence or absence of ACL, PCL, medial meniscus
(MM), lateral meniscus (LM), medial collateral liga-
ment (MCL), and lateral collateral ligament (LCL)
tears were recorded. Complete discontinuity of liga-
ment bundles or indistinct margins of the ligaments
with or without abnormal signal intensity were used
as the diagnostic criteria for cruciate and collateral liga-
ments (12–14). The diagnostic standard for meniscal
tear was defined as definite extension of intrameniscal
increased signal intensity to the superior or inferior
meniscal surface (15,16).
The presence of a meniscal tear was defined by the
side of the meniscus that was torn and not by the type
of tear. The readers recorded one meniscal tear even if
multiple meniscal tears were noted.
The arthroscopic findings were regarded as a refer-
ence for cruciate ligament and meniscal tears. For cru-
ciate ligaments, decreased tension during probing or
grossly or focally disrupted bundles indicated a tear.
For menisci, a visible fissure or cleft extending over
the meniscus or substantial surface fraying indicated a
tear. In consensus, using conventional 2D images and
clinical symptoms and signs such as positive valgus or
varus stress tests, the two radiologists established diag-
nostic standard definitions of MCL and LCL tears. A
collateral ligament tear was defined as a complete dis-
continuity or an indistinct margin of the ligament fibers
on conventional coronal FS T2W imaging. For lateral
collateral ligaments, only fibular collateral ligaments
were included.
Statistical analysis
All statistical analyses were carried out with the
Predictive Analytics Software (PASW) Statistics ver-
sion 18 (SPSS Inc., Chicago, IL, USA). Statistical sig-
nificance was set at P<0.05. The sensitivity, specificity,
and accuracy of both methods for evaluating tears of
the cruciate ligaments and both menisci were calcu-
lated. The difference in sensitivity and specificity
between conventional 2D imaging and FS 3D VISTA
imaging was tested using the McNemar test. The sensi-
tivity, specificity, and accuracy of FS 3D VISTA
Table 1. Inter-observer agreement analyzed with unweighted
kappa.
Conventional 2D FS 3D VISTA
ACL 0.885 0.844
PCL 0.85 0.85
LM 0.937 0.903
MM 0.914 0.928
Data are k-values.
68 Acta Radiologica 57(1)
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

imaging for diagnosing the tears of both medial and
lateral collateral ligaments were also calculated. By
non-weighted kappa (k) statistics, inter-observer agree-
ment was tested between conventional 2D and FS 3D
VISTA images. A kvalue of less than or equal to 0.20
indicated poor inter-observer agreement, a kvalue
of 0.21–0.40 indicated fair agreement, a kvalue of
0.41–0.60 indicated moderate agreement, a kvalue of
0.61–0.80 indicated good agreement, and kvalue
of 0.81–1.00 indicated excellent agreement (17).
Results
Arthroscopic surgery revealed 33 ACL tears (30 com-
plete, 3 partial), three PCL tears, 42 LM tears (3 verti-
cal, 7 horizontal, 3 radial, 3 bucket-handle, 1 oblique,
and 25 complex tears), and 68 MM tears (7 vertical, 15
horizontal, 7 radial, 3 bucket-handle, 1 oblique, and 35
complex tears). In consensus the two radiologists using
conventional 2D images along with clinical information
found five MCL tears. However, no LCL tears were
found when conventional 2D images and clinical infor-
mation were used.
Inter-observer agreement was excellent for all eval-
uated lesions with both conventional 2D and FS 3D
VISTA images and ranged from k¼0.85 (PCL tear
evaluated with both conventional 2D images and FS
3D VISTA images) to k¼1.0 (MCL tear evaluated
with FS 3D VISTA image). The kappa values are tabu-
lated in Table 1.
The sensitivity, specificity, and accuracy of conven-
tional 2D and FS 3D VISTA images that were used in
the diagnosis of tears of the cruciate ligaments, both
menisci, and MCL by each reader are tabulated in
Table 2. Because the inter-observer agreement was
excellent for ligaments and meniscal tears, the mean
Table 2. Diagnostic performance in evaluating ligaments, meniscal and MCL tears for each reader.
Reader 1 Reader 2
Joint abnormality Conventional 2D FS 3D VISTA Conventional 2D FS 3D VISTA
ACL tear Sensitivity 93.94% (31/33) 96.97% (32/33) 100% (33/33) 93.94% (31/33)
95% CI 82.6–98.8 86.2–99.8 89.9–100.0 82.6–98.8
Specificity 97.8% (132/135) 97.78% (132/135) 97.0% (131/135) 97.78% (132/135)
95% CI 95.0–99.0 95.2–98.5 94.6–97.0 95.0–99.0
Accuracy 97% 97.62% 97.6% 97.02%
PCL tear Sensitivity 100% (3/3) 100% (3/3) 100% (3/3) 100% (3/3)
95% CI 34.8–100.0 34.8–100.0 34.8–100.0 34.8–100.0
Specificity 99.39% (164/165) 99.39% (164/165) 99.39% (164/165) 99.39% (164/165)
95% CI 98.2–99.4 98.2–99.4 98.2–99.4 98.2–99.4
Accuracy 99.4% 99.4% 99.4% 99.4%
LM tear Sensitivity 90.48% (38/42) 95.24% (40/42) 95.24% (40/42) 95.24% (40/42)
95% CI 79.1–96.7 84.9–99.1 84.6–99.2 84.8–99.2
Specificity 89.68% (113/126) 91.27% (115/126) 88.10% (111/126) 90.48% (114/126)
95% CI 85.9–91.8 87.8–92.6 84.5–89.4 87.0–91.8
Accuracy 89.88% 92.26% 89.88% 91.67%
MM tear Sensitivity 94.12% (64/68) 94.12% (64/68) 92.65% (63/68) 95.59% (65/68)
95% CI 96.8–98.0 86.9–98.0 85.1–97.1 88.7–98.8
Specificity 85.00% (85/100) 87.00% (87/100) 85.00% (85/100) 88.00% (88/100)
95% CI 80.0–87.6 82.1–89.6 79.9–88.0 83.3–90.2
Accuracy 88.69% 89.35% 88.1% 91.07%
MCL tear Sensitivity NA 100% (5/5) NA 100% (5/5)
95% CI 55.1–100 55.1–100
Specificity NA 100% (163/163) NA 100% (163/163)
95% CI 98.6–100 98.6–100
Accuracy NA 100% NA 100%
ACL, anterior cruciate ligament; CI, confidence interval; LM, lateral meniscus; MCL, medial collateral ligament; MM, medial meniscus; PCL, posterior
cruciate ligament.
Lim et al. 69
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

sensitivities, specificities, and accuracies of both readers
were calculated for these aspects (Table 3). The mean
sensitivity, specificity, and accuracy of both the conven-
tional 2D MR images and the FS 3D VISTA images
used in the diagnosis of cruciate ligament tears was
95.45% or more, and there was no statistically signifi-
cant difference in sensitivity and specificity between the
conventional 2D MR images and the FS 3D VISTA
images (Fig. 2, Table 3).
In the diagnosis of an LM tear, there was no signifi-
cant difference in the mean sensitivity and specificity
between the FS 3D VISTA images and the conven-
tional 2D images. For the diagnosis of MM tears, the
mean sensitivity between conventional 2D images and
the FS 3D VISTA images (Fig. 3) was not statistically
significant, however, there was a significant difference
in the mean specificity (P<0.05).
The mean sensitivity, specificity, and accuracy were
all 100% for detecting MCL tears in FS 3D VISTA
images (Fig. 4).
Discussion
Isotropic 3D MR images can be obtained with a size of
0.5 0.5 0.5 mm voxels. This can reduce the partial
volume artifacts associated with conventional 2D MR
images (18). MPR can be utilized in all necessary
planes, including the standard orthogonal planes,
within only a few seconds. Therefore, it can be helpful
for the detection and analysis of complex or variably-
positioned internal structures of the knee. Lim et al.
have shown that the 3D isotropic MR images with
reader-defined MPR images have better diagnostic per-
formance in detecting small radial and root tears of the
menisci than conventional 2D images (11). This was
Fig. 2. Images of a 25-year-old male patient with surgically proven ACL complete tear that was detected by both readers and both
conventional 2D MRI and FS 3D VISTA imaging. (a) Sagittal conventional T2W FSE image show a discontinuity of ACL (arrow).
(b) Corresponding FS 3D VISTA image show an ACL tear (arrow). (c) Oblique coronal reformatted image shows an ACL tear (arrow)
Table 3. Diagnostic performance in evaluating ligaments,
meniscal, and MCL tears.
Joint
abnormality
Conventional
2D FS 3D VISTA Pvalue
ACL tear Sensitivity 96.97% 95.45% 0.99
95% CI 90.5–99.5 88.8–98.7
Specificity 97.41% 97.78% 0.99
95% CI 95.8–98.0 96.1–98.6
Accuracy 97.32% 97.32%
PCL tear Sensitivity 100% 100% 0.99
95% CI 57.3–100.0 57.3–100.0
Specificity 99.39% 99.39% 0.99
95% CI 98.6–99.4 98.6–99.4
Accuracy 99.4% 99.4%
LM tear Sensitivity 92.86% 95.24% 0.5
95% CI 85.8–96.9 99.8–98.4
Specificity 88.89% 90.84% 0.063
95% CI 86.6–90.3 88.7–91.9
Accuracy 89.88% 91.94%
MM tear Sensitivity 93.38% 94.85% 0.5
95% CI 88.6–96.5 90.3–97.6
Specificity 85% 87% <0.05
95% CI 81.7–87.2 83.9–88.9
Accuracy 88.4% 91%
MCL tear Sensitivity NA 100%
95% CI 55.1–100.0
Specificity NA 100%
95% CI 98.6–100.0
Accuracy NA 100%
ACL, anterior cruciate ligament; CI, confidence interval; LM, lateral
meniscus; MCL, medial collateral ligament; MM, medial meniscus;
PCL, posterior cruciate ligament.
70 Acta Radiologica 57(1)
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

probably due to a smaller interslice gap and the readers’
desire for reconstructed images along the structures of
interest.
MRI of the knee has been well documented as
having high sensitivity and specificity for the diagnosis
of cruciate ligament and meniscal tears. Recently, many
studies showed that the isotropic 3D images have a
diagnostic performance that is similar to that of the
conventional 2D images (1–5,8,11,19). The 3D VISTA
sequence provides high-resolution volumetric
intermediate-weighted (IW) images acquired with
TSE. Acquisition time and inter-echo spacing are opti-
mized by applying reduced flip angles in combination
with non-selective refocusing pulses. Combining paral-
lel acquisitions using robust factors in both directions
of the encoding phase ensures faster acquisition times.
Isotropic 3D FSE images have a similar diagnostic per-
formance in meniscal and ligament tears and perform
superiorly in the evaluation of cartilaginous defects in
the knee joint (10). Jung et al. compared isotropic 3D
Fig. 4. Images of a 21-year-old male patient with MCL tear detected at coronal FS 2D MRI which was detected by both readers at FS
3D VISTA imaging. (a) Coronal FS T2W FSE MR image show complete rupture of MCL (arrow). (b) Corresponding at FS 3D VISTA
imaging show MCL complete rupture (arrow).
Fig. 3. Images of a 50-year-old male patient with surgically proven MM tear which was detected by both readers and FS 3D VISTA
imaging. However, both readers interpreted as meniscal degeneration on conventional 2D MRI. (a) Sagittal T2W FSE image of knee
joint show a focal high signal lesion at MM (arrow), both readers interpreted as degeneration of meniscus. (b) Corresponding FS 3D
VISTA MR imaging show cleft like signal intensity extent to periphery of MM (arrow).
Lim et al. 71
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

IW FSE, isotropic 3D balanced fast field-echo (FFE),
and conventional 2D FSE and did not find any signifi-
cant differences in performance between the 2D FSE,
3D IW FSE, and 3D balanced FFE in the evaluation of
menisci and ligament tears (10). The result of that study
was similar to our results in that there was no signifi-
cant difference in diagnostic performance in evaluating
the tears of the ACL, PCL, and both menisci, there was
high inter-observer agreement, but no evaluation of the
collateral ligaments. We can obtain fat-suppressed 3D
MR images by applying a frequency selective fat satur-
ation technique. A fat-saturated image has an advan-
tage of good visualization of small anatomic details
particularly in areas with a large amount of fat such
as joint structures (20).
Because of the anatomical complexity of the LM, it
has been reported that LM tears are more likely to be
missed in conventional 2D MR images if the tear
involves only one-third of the meniscus, or is in the
posterior horn (21). However, although the FS 3D
VISTA sequence has isotropic voxels and a smaller
interslice gap compared with conventional 2D images,
it did not have a better diagnostic performance in LM
evaluation regardless of less partial volume artifacts.
Although we did not classify the extent of the tear in
the meniscus, we found that FS 3D VISTA imaging has
a higher specificity in medial meniscal tears regardless
of the type or extent.
Yoon et al. (22) reported that the sensitivity, speci-
ficity, and accuracy were 95%, 100%, and 98%,
respectively, using 1-mm thick 2D fast SE IW imaging
performed using the postprocessed MPR technique
with a 3.0 T MR scanner for the diagnosis of meniscal
tears. The sensitivity, specificity, and accuracy were
100% for the diagnosis of ACL and PCL tears.
However, their scanning method was not isotropic
and yielded some degree of image degradation during
reformation. Jung et al. (7) reported that the perform-
ance of 3D isotropic FSE IM MRI with MPR was
similar to that of conventional 2D imaging in the diag-
nosis of cruciate ligament tears and meniscal tears of
the knee. However, that study did not evaluate the col-
lateral ligaments. Similar to our results, those of
Kijowski et al. (8) showed that FSE-CUBE sequences
have a similar diagnostic ability as conventional 2D
MR images for detecting cruciate ligament tears, col-
lateral ligament tears, meniscal tears, and bone marrow
edema, as well as having excellent inter-observer
agreement.
Our results show that the FS 3D VISTA images have
a similar diagnostic value compared to the conventional
2D images and that the 3D VISTA images may replace
the conventional 2D images. In addition, the imaging
time was shorter in the FS 3D VISTA images than in
the conventional 2D MR images. The MPR images can
be semi-automatically performed within a few seconds
at our institute with little mouse clicking. With MPR
focused on suspicious abnormal lesions without partial
volume or summation artifacts due to adjacent struc-
tures, the diagnostic performance could be increased.
At our institution, the imaging time was shorter for FS
3D VISTA images (7 min 5 s) than for conventional 2D
MR images (15 min 7 s). With shorter imaging time, the
patient can be more comfortable, which might reduce
motion artifacts and improve the image as a collateral
effect.
Our study had several limitations. First, we analyzed
the MR images retrospectively. Second, the readers
were aware that the patients had undergone arthro-
scopic surgery when they analyzed the MR findings.
This factor may have introduced reader bias and there-
fore potentially influenced those results that indicated
high diagnostic performance or overestimation in
lesions. Third, we did not evaluate cartilage, muscle,
or bone marrow. A study of FS 3D isotropic images
will be necessary for detection of bone marrow lesion or
marrow edema. More studies that investigate knee car-
tilage are necessary. Fourth, we did not classify the type
of meniscal tear such as radial tears. A study on the
diagnostic value of 3D isotropic MR images in meniscal
tear classification is needed. Finally, we did not com-
pare the diagnostic performance between the non-FS
3D images and the FS 3D images. In addition, refor-
matting according to structures of interest may affect
the diagnostic performances in several cases. More stu-
dies about diagnostic performance or image quality of
non-FS 3D images versus FS 3D images are needed.
In conclusion, FS 3D VISTA imaging with MPR
may replace conventional 2D FSE MRI because of
the similar diagnostic ability and shorter image acqui-
sition time in evaluating tears of the ACL, PCL, MCL,
and both menisci.
Conflict of interest
None declared.
Funding
This work was supported by the National Research
Foundation of Korea (NRF) grant funded by the Korea gov-
ernment (MEST) (2012R1A2A1A01011328).
References
1. Escobedo EM, Hunter JC, Zink-Brody GC, et al.
Usefulness of turbo spin-echoMR imaging in the evalu-
ation of meniscal tears: comparison with a conventional
spin-echo sequence. Am J Roentgenol 1996;167:
1223–1227.
2. Jee WH, McCauley TR, Kim JM, et al. Meniscal tear con-
figurations: categorization with MR imaging. Am J
Roentgenol 2003;180:93–97.
72 Acta Radiologica 57(1)
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from

3. Kijowski R, Lu A, Block W, et al. Evaluation of the
articular cartilage of the knee joint with vastly under-
sampled isotropic projection reconstruction steady-state
free precession imaging. J Magn Reson Imaging 2006;24:
168–175.
4. Oei EH, Nikken JJ, Verstijnen AC, et al. MR imaging of
the menisci and cruciate ligaments: a systematic review.
Radiology 2003;226:837–848.
5. Duc SR, Pfirrmann CW, Koch PP, et al. Internal knee
derangement assessed with 3-minute three-dimensional
isovoxel true FISP MR sequence: preliminary study.
Radiology 2008;246:526–535.
6. Gold GE, Busse RF, Beehler C, et al. Isotropic MRI of
the knee with 3D fast spin-echo extended echo-train
acquisition (XETA): initial experience. Am J
Roentgenol 2007;188:1287–1293.
7. Jung JY, Yoon YC, Kwon JW, et al. Diagnosis of inter-
nal derangement of the knee at 3.0-T MR imaging: 3D
isotropic intermediate-weighted versus 2D sequences.
Radiology 2009;253:780–787.
8. Kijowski R, Davis KW, Woods MA, et al. Knee joint:
comprehensive assessment with 3D isotropic resolution
fast spin-echo MR imaging–diagnostic performance com-
pared with that of conventional MR imaging at 3.0 T.
Radiology 2009;252:486–495.
9. Ristow O, Steinbach L, Sabo G, et al. Isotropic 3D fast
spin-echo imaging versus standard 2D imaging at 3.0 T of
the knee–image quality and diagnostic performance. Eur
Radiol 2009;19:1263–1272.
10. Jung JY, Yoon YC, Kim HR, et al. Knee derangements:
comparison of isotropic 3D fast spin-echo, isotropic 3D
balanced fast field-echo, and conventional 2D fast spin-
echo MR imaging. Radiology 2013;268:802–813.
11. Lim D, Lee YH, Kim S, et al. Fat-suppressed volume
isotropic turbo spin echo acquisition (VISTA) MR ima-
ging in evaluating radial and root tears of the meniscus:
focusing on reader-defined axial reconstruction. Eur J
Radiol 2013;82:2296–2302.
12. Robertson PL, Schweitzer ME, Bartolozzi AR, et al.
Anterior cruciate ligament tears: evaluation of multiple
signs with MR imaging. Radiology 1994;193:829–834.
13. Barnett MJ. MR diagnosis of internal derangements of
the knee: effect of field strength on efficacy. Am J
Roentgenol 1993;161:115–118.
14. Mink JH, Levy T, Crues JV 3rd. Tears of the anterior
cruciate ligament and menisci of the knee: MR imaging
evaluation. Radiology 1988;167:769–774.
15. De Smet AA, Norris MA, Yandow DR, et al. MR diag-
nosis of meniscal tears of the knee: importance of high
signal in the meniscus that extends to the surface. Am J
Roentgenol 1993;161:101–107.
16. De Smet AA, Tuite MJ, Norris MA, et al. MR diagnosis
of meniscal tears: analysis of causes of errors. Am J
Roentgenol 1994;163:1419–1423.
17. Seigel DG, Podgor MJ, Remaley NA. Acceptable values
of kappa for comparison of two groups. Am J Epidemiol
1992;135:571–578.
18. Yao L, Pitts JT, Thomasson D. Isotropic 3D fast spin-
echo with proton-density-like contrast: a comprehensive
approach to musculoskeletal MRI. Am J Roentgenol
2007;188:W199–201.
19. Jung JY, Jee WH, Park MY, et al. Meniscal tear config-
urations: categorization with 3D isotropic turbo spin-
echo MRI compared with conventional MRI at 3 T.
Am J Roentgenol 2012;198:W173–180.
20. Delfaut EM, Beltran J, Johnson G, et al. Fat suppression
in MR imaging: techniques and pitfalls. Radiographics
1999;19:373–382.
21. Jones AO, Houang MT, Low RS, et al. Medial meniscus
posterior root attachment injury and degeneration: MRI
findings. Australas Radiol 2006;50:306–313.
22. Yoon YC, Kim SS, Chung HW, et al. Diagnostic efficacy
in knee MRI comparing conventional technique and mul-
tiplanar reconstruction with one-millimeter FSE PDW
images. Acta Radiol 2007;48:869–874.
Lim et al. 73
at Medical Lib / Yonsei University on January 28, 2016acr.sagepub.comDownloaded from