Information Technology Reference
In-Depth Information
6 Measurements of Bone Marrow Fat Content
The bone marrow is the non-mineralized component of bone. The interaction of the
mineralized and non-mineralized component plays an important role in the path-
ophysiology of age-related bone loss [
92
]. The bone marrow is more metabolically
active and responsive than the mineralized component of bone. Therefore, it has
been hypothesized that the quanti
cation of bone marrow fat content could be used
as a new diagnostic and therapeutic approach for osteoporosis. Bredella et al. [
93
]
demonstrated that vertebral bone marrow fat is positively associated with visceral
fat. Furthermore, it has been reported that visceral adiposity and the metabolic
syndrome have potential detrimental effects on bone health [
94
,
95
]. On the con-
trary, obese women are at decreased risk for developing osteoporosis [
96
].
In the lights of these con
icting results, considerable research effort has been
undertaken recently to gain more insights into the bone marrow metabolism and its
relationship with BMD, bone strength, and other body fat depots [
97
]. Proton single-
voxel magnetic resonance spectroscopy (
1
H-MRS) is the most widely used method
to non-invasively quantify bone marrow fat in-vivo [
98
]. Other non-spectroscopic
MR methods include T
1
T1-weighted imaging and chemical shift-based water-fat
separation techniques, e.g. the Dixon methods and the Iterative Decomposition of
water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) method
[
99
fl
101
]. All methods lack ionizing radiation.
1
H-MRS is considered as gold
standard [
102
,
103
]. It requires to prescribe a volume of interest in the exact desired
anatomical location which can be technically demanding. Figure
12
shows a
representative
-
1
H-MRS-based spectrum of L4: the spectrum shows the methyl
(
-
(CH
2
)
n
-
CH
3
) peak at 0.9 ppm (peak 1), the superposition of the methylene
(
-
(CH
2
)
n
-
) peak at 1.30 ppm and the
ʲ
-carboxyl (
-
CO
-
CH
2
-
CH
2
-
) peak at
ʱ
1.59 ppm (peak 2), the superposition of the
-ole
nic (
-
CH
2
-
CH=CH
-
CH
2
-
) peak
ʱ
at 2.00 ppm and the
-carboxyl (
-
CO
-
CH
2
-
CH
2
-
) peak at 2.25 ppm (peak 3),
the water peak at 4.7 ppm (peak 4), and the ole
) peak at 5.3 ppm
(peak 5).
1
H-MRS-based bone marrow fat fraction (also named fat content) has been
previously de
nic (
CH=CH
-
-
) fat peak at 1.30 ppm and the water
peak at 4.7 ppm as the relative fat signal intensity amplitude in terms of a percentage
of total signal intensity amplitude (fat and water) [
102
,
103
]. Li et al. [
104
] per-
formed
1
H-MRS in six subjects twice at four vertebral body levels (L1
ned by the methylene (
(CH
2
)
n
-
-
L4) on the
same day with repositioning between the two scans. The averaged coef
-
cient of
variation (CV) of vertebral bone marrow fat content was 1.7 %, suggesting good in-
vivo reproducibility. T
1
T1-weighted MR imaging is a conventional practice which is
not technically demanding in terms of acquisition. A threshold is usually applied on
the T
1
T1-weighted images to de
ne voxels as bone marrow voxels (i.e. bone marrow
adipose tissue volume) in the segmented vertebra [
105
]. The bone marrow fat
fraction is de
ned as the number of bone marrow voxels divided by the total number
of voxels in the segmented vertebra. The applied threshold is often set at the same
level as subcutaneous adipose tissue on the grey scale [
101
,
105
,
106
]. The intra- and
interobserver reproducibility for the assessment of bone marrow adipose tissue in
