Osteoarthritis and Cartilage 18 (2010) 1557e1563

The effects of acute loading on T1rho and T2 relaxation times of tibiofemoral

articular cartilage
R.B. Souza yz *, C. Stehling x, B.T. Wyman k, M.-P. Hellio Le Graverand k, X. Li y, T.M. Link y, S. Majumdar y
y Department of Radiology and Biomedical Imaging, University of California, San Francisco, CA 94107, USA
z Department of Physical Therapy and Rehabilitation Science, University of California, San Francisco, CA 94107, USA
x Department of Clinical Radiology, University of Muenster, Germany
k Pfizer Inc, Groton, CT, USA

a r t i c l e i n f o s u m m a r y

Article history: Objective: To evaluate the effect of acute loading on healthy and osteoarthritic knee cartilage T1r and T2
Received 1 April 2010 relaxation times.
Accepted 4 October 2010 Design: Twenty subjects with radiographic evidence of osteoarthritis (OA) and 10 age-matched controls
were enrolled. Magnetic resonance imaging (MRI) acquisition, including T1r and T2 map sequences were
Keywords: performed unloaded and loaded at 50% body mass. Cartilage masks were segmented semi-automatically
Magnetic resonance imaging
on registered high-resolution spoiled gradient-echo (SPGR) images for each compartment (medial and
T1rho
lateral). Cartilage lesions were identified using a modified Whole Organ Magnetic Resonance Imaging
T2
Acute loading
Score (WORMS) score. Statistical differences were explored using separate two-way (group  loading
Osteoarthritis condition) Analysis of Variance (ANOVA) using age as a covariate to evaluate the effects of loading on T1r
Cartilage and T2 relaxation times.
Results: A significant decrease in T1r (44.5  3.8 vs 40.2  4.8 ms for unloaded and loaded, respectively;
P < 0.001) and T2 (31.8  3.8 vs 30.5  4.8 ms for unloaded and loaded, respectively; P < 0.001) relaxa-
tion times was observed in the medial compartment with loading while no differences were observed in
the lateral compartment. This behavior occurred independent of WORMS score. Cartilage compartments
with small focal lesions experienced greater T1r change scores with loading when compared to cartilage
without lesions or cartilage with larger defects (P ¼ 0.05).
Conclusions: Acute loading resulted in a significant decrease in T1r and T2 relaxation times of the medial
compartment, with greater change scores observed in cartilage regions with small focal lesions. These
data suggest that changes of T1r values with loading may be related to cartilage biomechanical properties
(i.e., tissue elasticity) and may be a valuable tool for the scientist and clinician at identifying early
cartilage disease.
Ó 2010 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Introduction unlikely to be observed on radiographs. Given as such, considerable

Diagnosis of knee osteoarthritis (OA) is typically performed
though identification of bone changes and joint space narrowing on
radiographs using the KellgreneLawrence (KL) scoring system1. This
indirect measure of cartilage disease is likely preferred due to its
relatively low cost and minimal requirements in analysis time.
However, the earliest changes in cartilage degeneration are alter-
ations in the biochemistry of the extracellular matrix and are
* Address correspondence and reprint requests to: Richard B. Souza, Department
of Radiology and Biomedical Imaging, University of California, 185 Berry St., Suite
350, San Francisco, CA 94107, USA.
E-mail address: richard.souza@ucsf.edu (R.B. Souza).
efforts have recently been focused on investigation of cartilage
biochemistry through recent advancements in magnetic resonance
imaging (MRI) technology2e8. More specifically, T1r and T2 relaxa-
tion time mapping have been developed as methods to quantify
cartilage composition in vivo. T1r relaxation time, also known as
spin-lattice relaxation in the rotating frame, probes the interaction
between motion-restricted water molecules and the macromolec-
ular environment. It has been demonstrated that T1r relaxation
times are negatively correlated with proteoglycan content and
tissue hydration of articular cartilage3,9. T2 relaxation times reflect
the ability of energy exchange between free water proton molecules
and are related to matrix structure and water content10e12.
These quantitative MR techniques allow for insight into the
biochemistry and function of the cartilage. There has been some

1063-4584/$ e see front matter Ó 2010 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.joca.2010.10.001
1558 R.B. Souza et al. / Osteoarthritis and Cartilage 18 (2010) 1557e1563
evidence that KL grading is correlated with T1r and T2 relaxation
times13,14. However, the indirect nature of the MR assessments may
lead to insensitivity of small differences in relaxation times.
Detection of early changes in cartilage biochemistry is critical to
begin to identify when to intervene on the disease process. Direct
measurement of cartilage lesions, such as with the Whole Organ
Magnetic Resonance Imaging Score (WORMS), has been reported to
be significantly related to T1r values15. These findings are promising
as a semiquantitative clinical grading system done on a standard
clinical MRI scan that can predict subtle changes in matrix struc-
ture, similar to T1r and T2 relaxation times would be valuable in
early detection and diagnosis of OA.
Loading plays a major role in the development of OA. Both
excessive load, as well as inadequate load have been shown to
result in degenerative changes in articular cartilage16,17. Currently
the influence of acute loading on articular cartilage biochemistry
remains uncertain. Two recent studies have evaluated the effects of
acute loading on T2 relaxation time of knee cartilage18,19. While
both studies reported decreases in T2 relaxation times, suggesting
a loss of water content or alteration to matrix structure, the findings
were not uniform throughout the knee cartilage. In addition, these
studies were performed on healthy volunteers, making it unclear if
these findings are also observed in persons with cartilage disease.
To date, no studies have evaluated the effects of acute loading on
T1r relaxation times of cartilage. It is unclear if loading patterns of
diseased cartilage behave differently than healthy cartilage. A
better understanding of the role of loading on T1r and T2 relaxation
times may provide some insight into the relationship between
cartilage disease and mechanical loading in vivo.
Therefore, the purpose of the current project is to expand on the
previous work in this area through a comprehensive evaluation of
acute loading on knee articular cartilage T1r and T2 relaxation times
in subjects with OA and healthy controls. It is hypothesized that
acute loading will result in a decrease in T1r and T2 relaxation times
in all subjects. Furthermore, we hypothesize that subjects with
cartilage lesions will exhibit greater T1r and T2 change scores with
loading when compared to controls without cartilage abnormalities.
Method
Study population
All subjects were recruited from University of California, San
Francisco (UCSF) orthopedic surgery clinics via physician referral
(OA subjects) or through advertising accomplished using study
flyers (controls). Inclusion criteria were female sex, age >40 years,
and a body mass index (BMI) of 25e35 kg/m2. The subjects who
passed phone screening were invited to have anterioreposterior
extended weight-bearing X-rays taken. On these radiographs,
presence or absence as well as grade of OA were documented using
the KL grading system. Inclusion criteria for the OA participants
were: self-reported pain, aching, or stiffness most days of a month
during the past year in addition to a KL 2 (osteophytes and no joint
space narrowing) or 3 score (osteophytes and joint space narrow-
ing) of the study knee (readings performed by C.S. and T.L.) with the
same degree of OA, less severe OA, or no OA in the contra-lateral
knee. Medial JSW had to be larger than 2 mm, and the medial JSW
smaller than the lateral JSW. Inclusion criteria for normal controls
were: self-reported infrequent (or no) knee pain, aching, or stiff-
ness during the past year and no radiographic evidence of OA on
either knee (bilateral KL 0 score). General exclusion criteria were
MRI contraindications including (potential) pregnancy, a history of
knee surgery (including, but not limited to meniscus surgery),
history of knee disease other than OA (i.e., inflammatory, crystal-
line, or infectious), and intra-articular steroid injection of the study
knee. All procedures were explained in detail to each subject and all
subjects signed an informed consent form approved by the
University’s Committee on Human Research.
MR imaging
MRI of the study knee of each subject was acquired using a 3 T
GE Excite Signa MR scanner (General Electric, Milwaukee, WI) with
a transmit/receive eight-channel phased-array knee coil (Invivo,
Orlando, FL). For control subjects the dominant leg was imaged, and
for patients with OA the leg with more severe OA was imaged. The
imaging protocol included sagittal two-dimensional (2D)
T2-weighted fat-saturated fast spin-echo (FSE) sequences (TR/TE:
4000/48 ms, field of view (FOV): 14 cm, matrix: 384  192, slice
thickness: 2.0 mm, echo train length: 9, bandwidth: 23.44 kHz,
number of excitations (NEX): 2), coronal 2D T2-weighted fat-satu-
rated FSE sequences (TR/TE: 3000/10 ms, FOV: 14 cm, matrix:
384  192, slice thickness: 2.0 mm, echo train length: 10, band-
width: 23.44 kHz, NEX: 2), coronal three-dimensional (3D) water
excitation high-resolution spoiled gradient-echo (SPGR) sequences
(TR/TE: 22/7.0 ms, flip angle 18, FOV: 14 cm, matrix: 512  512, slice
thickness: 1.5 mm, bandwidth: 31.25 kHz, NEX: 1), coronal 3D T1r-
weighted images based on SPGR acquisition that was previously
developed in our lab (Magnetization-prepared Angle-modulated
Partitioned-k-space Spoiled Gradient-Echo Snapshots,[MAPSS]20;
TR/TE: 9.3/3.7 ms; FOV: 14 cm, matrix: 256  128, slice thickness:
3 mm, BW: 31.25 kHz, views per segment: 64, recovery time: 1.5 s,
time of spin lock: 0, 10, 40, 80 ms, frequency of spin lock: 500 Hz),
and coronal 3D T2 imaging sequence based on SPGR acquisition (all
parameters were the same as T1r except for TE: 4.1, 14.5, 25,
45.9 ms). T1r and T2 maps were quantified on a pixel-by-pixel basis.
In-plane spatial resolution for both map sequences was
0.5  0.5 mm.
Procedures
All imaging procedures were performed at the Department of
Radiology and Biomedical Imaging at the University of California,
San Francisco. The study consisted of two phases: unloaded
imaging, and loaded imaging at 50% body weight. All imaging was
performed in the early morning to avoid the influence of weight-
bearing throughout the day as a confounder. Subjects arrived at the
imaging center 30 min prior to their appointment time and were
kept in an unloaded position (wheelchair) until their scan time.
Subjects were positioned in the MR scanner in a supine position in
a MR-compatible axial loading device (Fig. 1). The study knee was
positioned in 10 degrees of external rotation with the knee flexed to
20 degrees, and stabilized with padding to prevent movement
during scanning. At this time sagittal and coronal FSE images,
coronal SPGR images, and T1r and T2 maps were acquired as
described above. Next, a load of 50% of the subjects’ body weight
was applied to the loading device, which transmitted the force
through a pulley system using a loading footplate to the test lower
extremity. The result was a 50% body weight load on the tested
lower extremity, intended to mimic static standing conditions.
Once loaded, all image sequences were repeated. The time between
load application and T1r and T2 map sequence acquisition was
approximately 20 min.
MR image analysis
All MR images of the knee obtained during unloading were
evaluated and scored independently on picture archiving commu-
nication system (PACS) workstations (Agfa, Ridgefield Park, NJ,
USA) by two musculoskeletal radiologists separately with 20 and 4
 R.B. Souza et al. / Osteoarthritis and Cartilage 18 (2010) 1557e1563
Fig. 1. Schematic (left) and photo (right) of loading device.
years of experience in musculoskeletal imaging. In instances of
conflicting scoring, consensus readings by both radiologists were
performed. During the reading session ambient light was reduced
and no time constraints were used. Both evaluators were blinded to
KL status and relaxation time data.
WORMS was used to semiquantitatively evaluate the cartilage in
all subjects21e23. Since only a relatively small number of lesions
were expected in these subjects with mild, moderate and no OA,
the number of anatomical compartments was reduced from 15 to 4
compartments and included: medial and lateral femur, and medial
and lateral tibia. Cartilage abnormalities were scored using a stan-
dard WORMS eight-point scale: 0 ¼ normal thickness and signal;
1 ¼ normal thickness but abnormal signal on fluid sensitive
sequences; 2.0 ¼ partial-thickness focal defect <1 cm in greatest
width; 2.5 ¼ full-thickness focal defect <1 cm in greatest width;
3 ¼ multiple areas of partial-thickness (Grade 2.0) defects inter-
mixed with areas of normal thickness, or a Grade 2.0 defect wider
than 1 cm but <75% of the region; 4 ¼ diffuse (75% of the region)
partial-thickness loss; 5 ¼ multiple areas of full-thickness loss
(grade 2.5) or a grade 2.5 lesion wider than 1 cm but <75% of the
region; 6 ¼ diffuse (75% of the region) full-thickness loss.
Cartilage masks were segmented semi-automatically on high-
resolution SPGR images that were registered to the map sequences
using in-house developed software based on the VTK CISG Regis-
tration Toolkit. Slice selection was determined based on cartilage-
on-cartilage contact during the loaded condition. The same number
of slices were segmented in the unloaded and loaded conditions for
each subject. Due to the difficulty in determining borders between
femoral and tibial cartilage during loaded imaging, each compart-
ment (medial and lateral) was assessed as a unit (i.e., medial femoral
cartilage þ medial tibial cartilage ¼ medial compartment, etc).
The segmented 3D masks were then applied to the T1r and T2
maps for quantification. Average T1r and T2 relaxation times of
cartilage in the medial and lateral compartments were compared
between conditions (unloaded vs loaded), and between groups
(diseased vs controls). Data were stratified based on WORMS
scoring: WORMS scores of 0 were defined as “W_controls”.
Compartments with signal abnormalities or focal cartilage defects
less than 1 cm (WORMS 1, 2 or 2.5) were defined as “W_mild_OA”.
Finally, compartments with multiple cartilage defects and defects
greater than 1 cm (WORMS score of 3 or greater) were defined as
“W_severe_OA”.
Statistical analysis
Statistical differences between loading conditions were
explored using separate two-way (group  loading condition)
Analysis of Variance (ANOVA) with repeated measures design for
the dependent variables T1r and T2 relaxation times, using age as
a covariate. All statistically significant interactions will be reported
1559
and post hoc analyses will be performed using paired (loading
analysis) or independent (group analysis) samples t-tests. If no
interactions are observed, individual main effects will be reported.
This analysis was repeated for medial and lateral compartments. All
statistical analyses were performed using SPSS statistical software
(SPSS Inc., Chicago, IL) with an alpha level of P < 0.05.
Results
Study group
Twenty females with radiographic evidence of OA [mean  SD
(standard deviation): age ¼ 57.1  4.7 years; BMI ¼ 27.9  2.6 kg/
m2] and 10 healthy female controls (mean  SD: age ¼ 52.9  6.5
years; BMI ¼ 28.0  2.0 kg/m2) participated. Of the 20 females with
radiographic evidence of OA, there was a combination of KL 2
(n ¼ 10) and KL 3 (n ¼ 10). All control subjects had bilateral radio-
graphs that received KL 0 scores.
WORMS analysis
Absence of cartilage lesions (W_controls) was observed in 15
subjects for the medial compartment and 21 subjects for the lateral
compartment. Small focal lesions (W_mild_OA) were observed in
four subjects for the medial compartment and six subjects for the
lateral compartment. Finally, larger lesions (W_severe_OA) were
observed in 11 subjects for the medial compartment and three
subjects for the lateral compartment. No statistically significant
relationships were observed between WORMS score and weight
(r2 ¼ 0.014) or WORMS score and BMI (r2 ¼ 0.001) in this cohort.
T1r and T2 change scores with loading
When combining all 30 subjects, an 8% statistically significant
reduction in T1r relaxation times of the medial compartment was
observed during the loading condition (44.2  3.4 ms vs
40.8  4.1 ms for unloaded and loaded, respectively; P < 0.001). A
smaller, yet statistically significant 3% reduction was observed in T2
relaxation times for the medial compartment with acute loading
(31.6  2.8 ms vs 30.7  3.0 ms for unloaded and loaded, respec-
tively; P < 0.001). No statistically significant differences were
observed in the lateral compartment for either T1r relaxation times
(41.9  4.6 ms vs 41.8  4.9 ms for unloaded and loaded, respec-
tively; P ¼ 0.396) or T2 relaxation times (31.5  3.2 ms vs
31.1  3.6 ms for unloaded and loaded, respectively; P ¼ 0.191).
WORMS vs T1r and T2
This behavior of decreased relaxation times in the medial
compartment with no change in lateral compartment occurred
1560 R.B. Souza et al. / Osteoarthritis and Cartilage 18 (2010) 1557e1563

Table I
T1r Relaxation times with loading

WORMS

Medial compartment Lateral compartment

W_controls (n ¼ 15) W_mild_OA (n ¼ 4) W_severe_OA (n ¼ 11) W_controls (n ¼ 21) W_mild_OA (n ¼ 6) W_severe_OA (n ¼ 3)

Unloaded (ms) mean  SD 42.67  1.61 44.76  0.65 46.16  4.73 40.35  3.90 44.45  4.44 47.87  2.39
Loaded (ms) mean  SD 38.92  1.67 39.35  3.21 43.79  5.11 39.91  3.52 45.53  5.85 47.59  2.57
Change (%) 8.8 12.1* 5.1y 1.1 2.4 0.6
Significance P < 0.001 P ¼ 0.014 P ¼ 0.004 P ¼ 0.237 P ¼ 0.103 P ¼ 0.346

* Indicates W_mild_OA change score is significantly different from W_controls (P ¼ 0.05) and W_severe (P ¼ 0.02).
y
Indicates W_severe_OA is significantly different from W_controls (P ¼ 0.01).

consistently in all groups, regardless of WORMS score (Tables I and
II, Fig. 2). A statistically significant difference in average T1r and T2
relaxation times was observed between the groups (load  group
interaction: F ¼ 5.061, P ¼ 0.014). Both the W_mild_OA group and
the W_severe_OA group had significantly higher T1r and T2 relax-
ation times than the W_controls (P < 0.05). Interestingly, not all
WORMS groups responded to the same magnitude with loading. A
significantly greater change of T1r relaxation times in the medial
compartment was observed in the W_mild_OA group when
compared to the other WORMS groups. The change of T1r relaxa-
tion times of W_mild_OA group (12.1%) was found to be signifi-
cantly greater than either the W_control group (8.8%; P ¼ 0.05) or
the W_severe_OA group (5.2%; P ¼ 0.02). In addition, the
W_severe_OA group T1r relaxation change scores were significantly
smaller than the W_control group (P ¼ 0.01). No other statistically
significant group differences were observed with regard to T1r or T2
relaxation time change scores. Finally, no other statistically signif-
icant load  group interactions were observed.
Discussion
Consistent with our hypothesis, reductions in T1r and T2 relax-
ation times were observed with acute loading. However, contrary
to our hypothesis, not all compartments showed this behavior
(Fig. 3). Across all subjects, significant reductions were observed in
the medial compartment with no change in the lateral compart-
ment. While controlling for age, a significantly larger reduction in
T1r relaxation times of the medial compartment was observed in
subjects with small focal cartilage lesions when compared to either
control subjects or subjects with more extensive cartilage lesions.
Our interpretation of this data is that acute loading results in
deformation of the collagen network and expulsion of water
molecules within the cartilage matrix resulting in a small reduction
in T2 relaxation times. The reduction in T1r relaxation times, which
has previously been shown to be sensitive to changes in proteo-
glycan content with an inverse relationship3, is not suggesting an
increase in the number on proteoglycan molecules, but rather an
overall increase in proteoglycan concentration as the cartilage
becomes compressed under load.
Supporting data of this theory comes from Liess and
colleagues24, who demonstrated a small but significant 3% reduc-
tion in T2 relaxation times following a bout of exercise. Similarly,
Mosher and colleagues25 showed an approximately 5% reduction
following 30 min of running. Additionally, several studies by Eck-
stein and colleagues26e28 have shown reductions in knee cartilage
thickness and volume following bouts of exercise. Interestingly,
these authors reported that the tibial cartilage was most responsive
to high-impact exercise26. It is possible that larger changes in T1r
and T2 relaxation times might have been observed if we used
a combination of high-impact loading and static loading in the
present study. However, we feel that the influence of simulated
static standing is important to understand the behavior of cartilage
under various activities of daily living and thus the current study
compliments the work done by Eckstein and colleagues and others
by furthering our understanding of the influence of loading on
cartilage behavior. However, it should be clearly noted that the
findings of Eckstein and colleagues were measures of cartilage
morphology and may or may not correspond with changes in T1r
and T2 relaxation times as measured in the current study.
One unexpected finding of the current study was that reduc-
tions in T1r and T2 relaxation times were limited to the medial
compartment. These findings, however, are consistent with Nishii
and colleagues18 who found that reductions in T2 relaxation times
were on average greater than those in the lateral compartment.
These authors also performed a subcompartment analysis, dividing
the tibiofemoral cartilage into six zones and were able to demon-
strate a significant reduction in some regions of the lateral tibial
cartilage18. However, it should be noted that these authors acquired
sagittal images while coronal images were acquired in the current
study. It is unclear if image plane orientation would affect results of
cartilage loading. Using a similar analysis, Nag and colleagues19
reported reductions in both medial and lateral femoral cartilage.
One possible explanation for this discrepancy is that 20 of the 30
subjects in the current study had radiographic evidence of medial
joint disease. When analyzing control subjects response to loading,

Table II
T2 Relaxation times with loading

WORMS

Medial compartment Lateral compartment

W_controls W_mild_OA W_severe_OA W_controls W_mild_OA W_severe_OA

(n ¼ 15) (n ¼ 4) (n ¼ 11) (n ¼ 21) (n ¼ 6) (n ¼ 3)
Unloaded (ms) mean  standard 30.16  1.90 32.32  1.52 33.42  3.14 30.38  2.79 33.99  3.24 34.68  1.41
deviation
Loaded (ms) mean  standard 29.30  1.85 30.78  1.93 32.60  3.70 29.94  3.18 33.43  3.47 35.06  2.13
deviation
Change (%) 2.9 4.8 2.5 1.4 1.6 1.1
Significance P ¼ 0.042 P ¼ 0.021 P ¼ 0.008 P ¼ 0.168 P ¼ 0.364 P ¼ 0.407
 R.B. Souza et al. / Osteoarthritis and Cartilage 18 (2010) 1557e1563 1561

Fig. 2. T1r (A & B) and T2 (C & D) changes in articular cartilage during loading.

we observed a slightly larger change score in the medial
compartment than in the lateral compartment (W_controls ¼ 3% vs
1.5% for change scores in the medial and lateral compartment,
respectively), but these differences were far less than were seen in
the combined OA groups. Finally, it should be noted that OA of the
medial compartment is more common than lateral compartment
OA and that medial cartilage thinning is more commonly observed
in healthy adults29,30. Taken together with our data, this might
suggest that acute loading occurs to a greater degree in the medial
compartment than the lateral compartment. However, further
exploration of this concept, including contact area and cartilage
deformation studies are warranted.
This study is the first to evaluate the effects of acute loading on T1r
relaxation times. Similar, to the T2 relaxation time data, a larger
change score was observed in the medial compartment. A significant
greater change score was observed in subjects in the W_mild_OA
group when compared to subjects in the W_control group or
W_severe_OA group. Stahl and colleagues evaluated T2 relaxation
times and WORMS analysis in persons with OA15. However, these
authors stratified subjects based on KL score making direct compar-
ison to our data impossible. While it remains highly speculative to
interpret this discrepancy in group behavior, one possible explanation
for these findings is that the greater change scores represent a deficit
in cartilage mechanical properties leading to greater changes in tissue
relaxation times with loading. Using finite element modeling data,
Guettler and colleagues reported that changes from osteochodral
lesions have significant load distribution effects to adjacent cartilage
with smaller lesions and that these effects were not significantly
greater with larger lesions31. While this may explain our observation
of greater change scores in the subjects with small cartilage lesions, it
does not explain why we saw statistically smaller change scores in
subjects with extensive cartilage lesions.
One possible explanation for this discrepancy is a transition of
articular cartilage composition toward fibrocartilage in subjects

Fig. 3. Representative T1r color map during unloaded (A) and loaded (B) conditions.
1562 R.B. Souza et al. / Osteoarthritis and Cartilage 18 (2010) 1557e1563
with advanced disease. Support of this hypothesis comes from the
observation of fairly high variability in T1r and T2 relaxation times
for W_severe_OA group in response to loading. When this group
was separated into two subgroups e those receiving a WORMS
score of 3 and those receiving a score of 5 e it was observed that
subjects with severe and extensive cartilage lesions as would be the
case when receiving a WORMS score of 5, often had lower T1r and
T2 relaxation times than either W_mild_OA group or the subjects
receiving a WORMS score of 3 in the W_severe_OA group (mean T1r
times for WORMS 5: 42.8 ms vs 47.4 ms and 44.8 for the WORMS 3
group and W_mild_OA group, respectively). This data is consistent
with other emerging data from our laboratory that suggests that T1r
and T2 relaxation times are most useful in evaluating early cartilage
degeneration and less effective at evaluating cartilage in later
stages of pathology. Additionally, segmentation difficulty in subject
with large cartilage lesions may be a significant source of error and
should be evaluated carefully. Taken together, this finding of greater
change scores in subjects with early stage of OA is exciting as the
response in T1r relaxation times to acute loading may provide non-
invasive insight into the biomechanical properties on cartilage.
When comparing the two classification systems evaluated in the
current study, T1r and T2 relaxation times were consistently
different between subjects with diseased cartilage and healthy
controls when stratified based on WORMS. Although average T1r
and T2 relaxation times were greater in the KL diseased groups, this
failed to reach statistical significance. These findings are as expec-
ted since WORMS analysis is a direct measure of cartilage pathology
rather than an indirect measure through joint space narrowing as is
done in KL grading.
Our finding of increased sensitivity of WORMS classification
compared to KL scoring is consistent with a previous investigation
by Li and colleagues13, who found that a significant increase in T1r
and T2 values was observed when subjects were separated based on
cartilage thinning grading from T2-weighted FSE and T1-weighted
SPGR images. An additional advantage of the WORMS score is that
it is compartment specific whereas a single KL score is assigned to
an entire knee joint. Together these data suggest that WORMS
scoring may be a better assessment tool for identifying early
changes in cartilage biochemistry when T1r and/or T2 analyses are
unavailable.
Several limitations need to be highlighted with regard to the
current investigation. First, all subjects with radiographic OA had
medial joint space narrowing which limits the generalizability to
subjects with lateral OA or bilateral compartment OA. Second,
loading in the current study was performed in a static fashion at
50% body mass for a limited period of time. Data by Eckstein and
colleagues26 suggests that higher impact loading may result in
greater changes with loading. Furthermore, our study only inves-
tigated the effects of acute loading on cartilage relaxation time
response. Analysis of the cartilage response after the removal of
loads may provide equally valuable information and should be
considered in future studies. Due to fatigability issues, load was
only applied to the lower extremity for approximately 20 min prior
to data acquisition and greater changes may have been observed
with longer periods of loading. Third, cartilage compartments were
evaluated as a unit rather than separating medial and lateral
femoral cartilage from medial and lateral tibial cartilage. While this
was necessary given the challenges in cartilage plate segmentation
during loaded scans, it cannot be concluded from this study if
changes in relaxation times are more predominant in either the
femoral or tibial cartilage plates. Fourth, it needs to be noted that
our cohort included a small sample of individuals with lesions and
a larger database of loading and unloading images would be valu-
able in furthering our understanding of this phenomenon. Finally,
we classified subjects with KL scoring of 3 as severe OA, but it
should be pointed out that the term “severe OA” has been used in
the past to describe patients with KL scores of 3 and/or 4. Therefore,
we have to limit our findings of the severe OA group to those
receiving a score of 3 as subjects receiving a score of 4 may or may
not behave similarly.
In conclusion, acute loading resulted in a significant decrease in
T1r and T2 relaxation times of the medial compartment, with
greater change scores observed in cartilage regions with small focal
lesions. No significant changes were observed in the lateral
compartment with loading. These data suggest that changes of T1r
values with loading may be related to cartilage biomechanical
properties (i.e., tissue elasticity) and may be a valuable tool for the
scientist and clinician at identifying early cartilage disease. Later
stage disease presents challenges in tissue segmentation due to loss
of the superficial zone, thinning and reactive changes of the carti-
lage composition including fibrosis, which may make the use of
techniques such as T1r and T2 relaxation changes with loading less
suitable.
Author contributions
All of the listed authors played key roles in the conception and
design, data acquisition, analyses and interpretation, manuscript
revision and final approval of the manuscript. Richard B. Souza
takes responsibility for the integrity of the work as a whole.
Conflict of interest
There are no competing interests to declare.
Acknowledgements
Source of funding: Pfizer, Inc., Groton, CT.
The authors would like to thank Eric Han from GE Healthcare for
his help with pulse sequence development and Choongsoo Shin for
his assistance with data acquisition.
References
1. Kellgren JH, Lawrence JS. Radiological assessment of osteo-
arthrosis. Ann Rheum Dis 1957;16:494e502.
2. Mosher TJ, Dardzinski BJ, Smith MB. Human articular cartilage:
influence of aging and early symptomatic degeneration on the
spatial variation of T2epreliminary findings at 3 T. Radiology
2000;214:259e66.
3. Akella SV, Regatte RR, Gougoutas AJ, Borthakur A, Shapiro EM,
Kneeland JB, et al. Proteoglycan-induced changes in T1rho-
relaxation of articular cartilage at 4T. Magn Reson Med
2001;46:419e23.
4. Maier CF, Tan SG, Hariharan H, Potter HG. T2 quantitation of
articular cartilage at 1.5 T. J Magn Reson Imaging 2003;17:
358e64.
5. Menezes NM, Gray ML, Hartke JR, Burstein D. T2 and T1rho
MRI in articular cartilage systems. Magn Reson Med 2004;51:
503e9.
6. David-Vaudey E, Ghosh S, Ries M, Majumdar S. T2 relaxation
time measurements in osteoarthritis. Magn Reson Imaging
2004;22:673e82.
7. Wheaton AJ, Dodge GR, Borthakur A, Kneeland JB,
Schumacher HR, Reddy R. Detection of changes in articular
cartilage proteoglycan by T(1rho) magnetic resonance
imaging. J Orthop Res 2005;23:102e8.
8. Regatte RR, Akella SV, Lonner JH, Kneeland JB, Reddy R. T1rho
relaxation mapping in human osteoarthritis (OA) cartilage:
comparison of T1rho with T2. J Magn Reson Imaging 2006;23:
547e53.
 R.B. Souza et al. / Osteoarthritis and Cartilage 18 (2010) 1557e1563
9. Wheaton AJ, Dodge GR, Elliott DM, Nicoll SB, Reddy R. Quan-
tification of cartilage biomechanical and biochemical proper-
ties via T1rho magnetic resonance imaging. Magn Reson Med
2005;54:1087e93.
10. Xia Y. Relaxation anisotropy in cartilage by NMR microscopy
(muMRI) at 14-microm resolution. Magn Reson Med 1998;39:
941e9.
11. Mosher TJ, Smith H, Dardzinski BJ, Schmithorst VJ, Smith MB.
MR imaging and T2 mapping of femoral cartilage: in vivo
determination of the magic angle effect. AJR Am J Roentgenol
2001;177:665e9.
12. Lusse S, Claassen H, Gehrke T, Hassenpflug J, Schunke M,
Heller M, et al. Evaluation of water content by spatially
resolved transverse relaxation times of human articular
cartilage. Magn Reson Imaging 2000;18:423e30.
13. Li X, Benjamin Ma C, Link TM, Castillo DD, Blumenkrantz G,
Lozano J, et al. In vivo T(1rho) and T(2) mapping of articular
cartilage in osteoarthritis of the knee using 3 T MRI. Osteoar-
thritis Cartilage 2007;15:789e97.
14. Dunn TC, Lu Y, Jin H, Ries MD, Majumdar S. T2 relaxation time
of cartilage at MR imaging: comparison with severity of knee
osteoarthritis. Radiology 2004;232:592e8.
15. Stahl R, Luke A, Li X, Carballido-Gamio J, Ma CB, Majumdar S,
et al. T1rho, T2 and focal knee cartilage abnormalities in physi-
cally active and sedentary healthy subjects versus early OA
patientsea 3.0-Tesla MRI study. Eur Radiol 2009;19:132e43.
16. Neidhart M, Muller-Ladner U, Frey W, Bosserhoff AK,
Colombani PC, Frey-Rindova P, et al. Increased serum levels of
non-collagenous matrix proteins (cartilage oligomeric matrix
protein and melanoma inhibitory activity) in marathon
runners. Osteoarthritis Cartilage 2000;8:222e9.
17. Hagiwara Y, Ando A, Chimoto E, Saijo Y, Ohmori-Matsuda K,
Itoi E. Changes of articular cartilage after immobilization in
a rat knee contracture model. J Orthop Res 2009;27:236e42.
18. Nishii T, Kuroda K, Matsuoka Y, Sahara T, Yoshikawa H. Change
in knee cartilage T2 in response to mechanical loading. J Magn
Reson Imaging 2008;28:175e80.
19. Nag D, Liney GP, Gillespie P, Sherman KP. Quantification of T(2)
relaxation changes in articular cartilage with in situ mechan-
ical loading of the knee. J Magn Reson Imaging 2004;19:
317e22.
20. Li X, Han ET, Busse RF, Majumdar S. In vivo T(1rho) mapping in
cartilage using 3D magnetization-prepared angle-modulated
1563
partitioned k-space spoiled gradient echo snapshots (3D
MAPSS). Magn Reson Med 2008;59:298e307.
21. Peterfy CG, Gold G, Eckstein F, Cicuttini F, Dardzinski B,
Stevens R. MRI protocols for whole-organ assessment of the
knee in osteoarthritis. Osteoarthritis Cartilage 2006;14(Suppl
A):A95e111.
22. Stahl R, Luke A, Ma CB, Krug R, Steinbach L, Majumdar S, et al.
Prevalence of pathologic findings in asymptomatic knees of
marathon runners before and after a competition in compar-
ison with physically active subjects-a 3.0 T magnetic reso-
nance imaging study. Skeletal Radiol 2008;37:627e38.
23. Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D,
et al. Whole-Organ Magnetic Resonance Imaging Score
(WORMS) of the knee in osteoarthritis. Osteoarthritis Cartilage
2004;12:177e90.
24. Liess C, Lusse S, Karger N, Heller M, Gluer CC. Detection of
changes in cartilage water content using MRI T2-mapping in
vivo. Osteoarthritis Cartilage 2002;10:907e13.
25. Mosher TJ, Liu Y, Torok CM. Functional cartilage MRI T2 mapping:
evaluating the effect of age and training on knee cartilage
response to running. Osteoarthritis Cartilage 2010;18:358e64.
26. Eckstein F, Lemberger B, Gratzke C, Hudelmaier M, Glaser C,
Englmeier KH, et al. In vivo cartilage deformation after
different types of activity and its dependence on physical
training status. Ann Rheum Dis 2005;64:291e5.
27. Eckstein F, Tieschky M, Faber SC, Haubner M, Kolem H,
Englmeier KH, et al. Effect of physical exercise on cartilage
volume and thickness in vivo: MR imaging study. Radiology
1998;207:243e8.
28. Eckstein F, Tieschky M, Faber S, Englmeier KH, Reiser M.
Functional analysis of articular cartilage deformation,
recovery, and fluid flow following dynamic exercise in vivo.
Anat Embryol (Berl) 1999;200:419e24.
29. Ledingham J, Regan M, Jones A, Doherty M. Radiographic
patterns and associations of osteoarthritis of the knee in
patients referred to hospital. Ann Rheum Dis 1993;52:520e6.
30. Cicuttini FM, Wluka AE, Wang Y, Davis SR, Hankin J, Ebeling P.
Compartment differences in knee cartilage volume in healthy
adults. J Rheumatol 2002;29:554e6.
31. Guettler JH, Demetropoulos CK, Yang KH, Jurist KA. Osteo-
chondral defects in the human knee: influence of defect size
on cartilage rim stress and load redistribution to surrounding
cartilage. Am J Sports Med 2004;32:1451e8.