Content uploaded by Mikael Boesen
Author content
All content in this area was uploaded by Mikael Boesen on Apr 26, 2021
Content may be subject to copyright.
quantify the spatial variations in T2 time and the T2 time versus CTh
relationship in OA knees.
Figure 1. Mean ±SD T2 relaxation time (a) and CTh (b) per VOI. Asterisks
indicate significant differences compared to the same VOI on the medial
plateau: *P<0.05, ** P<0.005. Number symbols indicate differences
between VOIs in the same compartment: #P<0.01, ## P<0.001.
835
ASSOCIATIONS AMONG A SPECTRUM OF KNEE OSTEOARTHRITIS
FEATURES ON MRI INFORM ABOUT POTENTIAL PATHOGENIC
MECHANISMS, DATA FROM THE OAI/FNIH BIOMARKERS
CONSORTIUM
W.E. van Spil y, L.A. Deveza z, D.J. Hunter z.
y
Univ. Med. Ctr. Utrecht,
Utrecht, Netherlands;
z
Royal North Shore Hosp. and Inst. of Bone and
Joint Res., Kolling Inst., Sydney Univ., Sydney, Australia
Purpose: To investigate associations among osteoarthritis (OA) features
on MRI of the patellofemoral (PF) and tibiofemoral (TF) knee com-
partments in persons with knee OA.
Methods: Data were derived from the FNIH Biomarkers Consortium, a
nested case-control study within the Osteoarthritis Initiative, in which
600 persons with knee OA were retrospectively selected for one index
knee showing progression of either pain (N ¼103), medial TF joint
space narrowing (JSN, N ¼103), or both pain and JSN (N ¼194), versus
subjects without progression (N ¼200) between 2 and 4-year follow-
up. MRIs of the index knee were obtained at 0, 1 and 2-year follow-up
and scored semiquantitatively for OA features in the PF, medial TF and
lateral TF compartments, according to MOAKS. Features that were
included in the current analysis were: cartilage (CART) morphology
(size of any cartilage loss and proportion of full-thickness cartilage loss,
respectively), bone marrow lesion (BML) size, osteophyte (OP) size,
effusion-synovitis and Hoffa synovitis (SYN), medial and lateral
meniscal (MEN) morphology and extrusion, and anterior cruciate liga-
ment tear (ACLT). For each feature, the maximum score per compart-
ment was selected. Associations among these features were studied by
categorical principal component analysis (CATPCA), investigating
potential common underlying domains represented by these OA fea-
tures by grouping them into components (factors). Components can be
visualized as lines within a multidimensional dot plot of all variables in
the model (i.e. OA features), every component catching the variability of
each variable to a different degree.
Results: Table 1 shows the component matrix as obtained from CATPCA
of baseline data. Numbers represent component loadings (i.e. correla-
tion coefficients of variables with each particular component, varying
between 1 and 1). Variables with more positive or more negative
loadings are best represented by that particular component and, as
such, can be grouped. Through comparison of models with different
numbers of components, a model with three components was judged
the most optimal. Component 1 loaded CART and BML features in all
compartments except for the lateral TF compartment, OP features in all
compartments, and medial MEN features. Component 2 loaded CART
and BML features in the PF compartment, but not OP features. Com-
ponent 3 loaded CART and BML features in the lateral TF compartment
and lateral MEN tear, but not OP features. SYN features and lateral MEN
extrusion showed insufficient associations with the dataset as a whole
to be retained in the model, but loaded maximally on component 1 and
3, respectively, if they were. The total explained variance of the model
was only 11.4% and this hardly improved when the number of compo-
nents was increased (data not shown). Results were almost identical at
1 and 2-year follow-up (data not shown).
Conclusions: Cross-sectional associations among the current spectrum
of OA features in the PF and TF compartments of knees with OA were
limited (i.e. low total explained variance of models), but consistent
between time points. CATPCA created components that differed with
respect to the most affected compartment (i.e. medial TF vs. lateral TF
vs. PF compartment) rather than predominating OA features (e.g. CART
vs. OP features). Of note, CART and BML features clustered together per
knee compartment more than OP size did. SYN features showed only
very weak associations with other OA features.
836
DELAYED GADOLINIUM ENHANCED MRI OF MENISCI AND
CARTILAGE (DGEMRIM/DGEMRIC) IN OVERWEIGHT PATIENTS WITH
KNEE OSTEOARTHRITIS eA CROSS SECTIONAL STUDY OF 86
OVERWEIGHT PATIENTS WITH INTRAARTICULAR ADMINISTERED
GADOLINIUM CONTRAST
S. Hangaardy
,
z, H. Gudbergsen y, C.L. Daugaardy
,
z, H. Bliddal y,
J.D. Nybingz, M.T. Nieminen x
,
k, V. Casu la x
,
¶, C.-J. Tiderius #,
M. Boesen z.
y
Parker Inst., Frederiksberg, Denmark;
z
Dept. of Radiology,
Bispebjerg and Frederiksberg Hospital, Denmark;
x
Res. Unit of Med.
Imaging, Physics and Technology, University of Oulu, Finland;
k
Dept. of
Diagnostic Radiology, Oulu University Hospital, Finland;
¶
Med. Res. Ctr.,
University of Oulu and Oulu University Hospital, Finland;
#
Dept. of
orthopedics, Clinical Sci. Lund, Lund University, Sweden
Purpose: Morphologic changes in the menisci are highly associated
with development of knee osteoarthritis (KOA). The aim of this study
was to examine the delayed gadolinium enhanced MRI of menisci
(dGEMRIM) and its relationship to Kellgren-Lawrence grade (KLG) and
to articular cartilage dGEMRIC in overweight patients with KOA
Methods: 86 overweight patients (BMI >27) with KOA from the CAROT-
study (Clin trial id: NCT00655941) with a mean KLG of 3 had an
ultrasound guided intraarticular injection of 0.1 ml negatively charged
contrast agent (4 mmol/l Multihance) in 10 mL Lidocain. Four inversion
times (50, 350, 650, 1410 ms) dGEMRIC and dGEMRIM was performed in
a 1.5T Philips, Intera scanner with a time delay of 90e120 min. T1
relaxation time-values were calculated for posterior weight bearing
femoral cartilage in the lateral knee compartment, and for the posterior
meniscal horn of both lateral and medial menisci. Due to advanced
medial KOA and missing cartilage, dGEMRIC in the medial compartment
could not be measured.
Table 1
Component matrix as obtained from CATPCA with baseline data. See main text for
explanation. Component loadings lower than .400 and higher than .400 are
written in bold. CART ¼cartilage, BML ¼bone marrow lesion, OP ¼osteophyte,
MEN ¼meniscus, ACLT ¼anterior cruciate ligament tear.
Component
12 3
CART defect size medial patellofemoral ,401 ,297 -,404
CART defect size medial tibia and central femur ,434 -,529 -,233
CARTdefect size medioposterior femur ,445 -,495 -.088
CARTdefect size lateral patellofemoral ,533 ,589 -,100
CART defect size lateral tibia and central femur ,352 ,164 ,391
CART defect size lateroposterior femur ,180 ,064 ,623
CART defect severity medial patellofemoral ,280 ,372 -,355
CART defect severity medial tibia and central femur ,422 -,470 -.143
CART defect severity medioposterior femur ,277 -,401 -,023
CART defect severity lateral patellofemoral ,491 ,618 -,081
CART defect severity lateral tibia and central femur ,219 ,053 ,358
CART defect severity lateroposterior femur ,067 ,052 ,558
BML size medial patellofemoral ,131 ,340 -,256
BML size medial tibia and central femur ,422 -,511 -,158
BML size medioposterior femur ,154 -,429 ,007
BML size lateral patellofemoral ,397 ,605 -,062
BML size lateral tibia and central femur ,230 -.040 ,342
BML size lateroposterior femur ,103 ,003 ,375
OP size medial patellofemoral ,726 ,132 -.122
OP size medial tibia and central femur ,738 -,228 -,007
OP size medioposterior femur ,740 -,010 ,085
OP size lateral patellofemoral ,728 ,299 ,033
OP size lateral tibia and central femur ,785 ,152 ,142
OP size lateroposterior femur ,698 ,088 ,078
Medial MEN tear ,326 -,633 ,033
Medial MEN extrusion ,469 -,454 -,111
Lateral MEN tear ,107 ,033 ,440
ACLT ,236 -,267 ,157
Explained variance (%) 5,7 3,7 2,0
Crohnbach's alpa ,855 ,756 ,519
Abstracts / Osteoarthritis and Cartilage 26 (2018) S60eS474 S463
Results: For posterior femoral cartilage (N ¼86) the mean T1 relaxation
time was 441 ms (confidence limits of the mean (CLM) 427e455 ms). The
mean T1 relaxation time in the lateral menisci (N ¼85) was 498 ms (CLM
478e518ms) and for the medial menisci (N ¼62) mean T1 relaxationtime
was 484 (CLM 458e510 ms). A positive correlation was found between
medial and lateral menisci with R ¼0.62 (P<0.0001) and a similar trend
was observed between lateral cartilage and lateral with R ¼0.26
(P¼0.02). Comparing meniscus T1 relaxation time-values from the most
affected knee compartment to KLG showed trends toward increasing T1-
values for KLG 1e3 and a decreasing T1-value for Menisci KLG 4
Conclusions: The positive correlation between lateral and medial
menisci may indicate a parallel degradative processes occurring in both
knee compartments. The correlation between menisci and cartilage T1-
values suggests concomitant, but different, degeneration in the two
tissues in OA. The interpretation of the numerical, and insignificantly,
inverse u-shaped relation between meniscal T1-values and KLG is in
accordance with previous findings comparing various degrees of
meniscal degeneration with both dGEMRIM and proteoglycans content
from biopsies.
837
REPEATABILITY AND DISCRIMINATION VALIDITY OF CARTILAGE
IMAGING BIOMARKERS FOR EXPERIMENTAL MEDICINE STUDIES OF
KNEE OSTEOARTHRITIS
J.W. MacKay y, T.D. Turmezei z, J. Kaggie y, A.R. Morgan x, R.L. Janiczek x,
W. Khan y, S. McDonnell y, M.J. Graves y, G.M. Treece y, A.W. McCaskie y,
F.J. GIlbert y.
y
Univ. of Cambridge, Cambridge, United Kingdom;
z
Norfolk
&Norwich Univ. Hosp., Norwich, United Kingdom;
x
GlaxoSmithKline,
Stevenage, United Kingdom
Purpose: The utility of magnetic resonance imaging (MRI) quantitative
imaging biomarkers (QIBs) of cartilage in experimental medicine
studies of potential disease modifying osteoarthritis treatments
(DMOATs) is uncertain due to the small sample sizes and short follow-
up involved. To be relevant to such studies, candidate QIBs must dem-
onstrate acceptable repeatability, discriminative ability and respon-
siveness. The purpose of this study was to evaluate test-retest
repeatability and discriminative ability of 4 cartilage QIBs at the knee
(thickness, T1rho, T2 and delayed gadolinium enhanced MRI of cartila ge
[dGEMRIC]) in order to assess their utility in experimental medicine
studies.
Methods: We imaged 9 participants with mild-moderate knee osteo-
arthritis (OA), characterised by radiographs with medial tibiofemoral
predominant disease and Kellgren-Lawrence grades 2e3, and 4 healthy
volunteers (HVs) matched for age, sex and body mass index. Partic-
ipants were imaged at baseline and 1 month.
MR studies were performed on a 3T system (MR 750, GE Healthcare)
and were split into 2 sessions. The first MR session included a 3-
dimensional spoiled gradient echo (3D SPGR) sequence and T1rho and
T2 mapping sequences. At the end of this session, we administered an
intravenous gadolinium based contrast agent (Dotarem, Guerbet LLC) at
double dose (0.2 mmol/kg). Participants then performed 10 min of
exercise on a stationary cycle to facilitate contrast penetration into the
joint for dGEMRIC. The second MR session began 90 min following
injection of contrast agent and consisted of T1 mapping using a variable
flip angle technique. MR pulse sequence details are provided in Table 1.
We performed surface based analysis of cartilage parameters using
Stradwin software (http://mi.eng.cam.ac.uk/~rwp/stradwin). In brief,
this consists of semi-automated thickness measurement on 3D
SPGR images, with generation of accurate inner and outer cartilage
Abstracts / Osteoarthritis and Cartilage 26 (2018) S60eS474S464