zyxwvutsr
P Magnetic Resonance Spectroscopy of the Liver:
Correlation With Standardized Serum, Clinical, and
Histological Changes in Diffuse Liver Disease
31
HESTERN. VAN
zyxw
HALL,' JEROEN
VAN DER GROND,'JAN VAN HATTUM,'CAROLEKOOIJMAN,3
TJAARD
u. HOOGENRLUD,~AND WILLEM P. TH. M. MALI'
WASSENAER-VAN
Many 31PMRS liver studies are assigned to determine TI values and absolute concentrations of hepatic
metabolites or to test the diagnostic value in monitoring hepatic cirrhosis or hepatitis. Although most studies agree on relaxation time values, because these are
independent of the method used, it is still not clear
which hepatic metabolic processes are responsible for
the observed spectroscopic changes in the liver of patients with hepatitis or fibrosis/cirrhosis. Meyerhoff et
a12showed that the phosphomonoester (PME) concentration is increased in viral hepatitis but not in alcoholic hepatitis, whereas Angus et a16 showed that the
PME concentration in alcoholic hepatitis was also increased and even was correlated with the severity of
alcoholic hepatitis. Also, the description of spectroscopic changes in the liver of patients with liver fibrosis
or cirrhosis is unclear: Angus et a16 suggested that hepatic PME concentration was not correlated to the severity of cirrhosis. This suggestion is supported by Rajanayagam et all and Meyerhoff et al,2 who both
31Pmagnetic resonance spectroscopy (MRS)has been showed that hepatic PME is not significantly increased
shown t o be a useful method for studying metabolism in alcoholic cirrhosis. However, other studies"25demonin normal liver and in the liver with diffuse liver abnor- strated that the hepatic PME was increased in patients
malities.'.' ' However, relatively little is known about with liver cirrhosis.
These contradicting results in the potential of MRS
the potential of MRS for diagnosing patients with diffuse liver disease into clinical and histological catego- in detecting hepatitis or cirrhosis may be caused by the
ries. Because experimental conditions (field strength, fact that in a limited number of patients histological
repetition time, acquisition technique, and quantifica- information was present. Because in these studies no,
tion routine used) are often different between studies or only limited, information is provided about the reand patient inclusion criteria are not the same or are sults of biopsy or serum analysis, it cannot be excluded
described poorly, it is difficult to compare individual that other hepatic abnormalities, or more specific hepatic abnormalities, may influence the PME concentrastudies with each other.
tion.
The goal of this study was to find possible correlations between MR spectroscopic results and standardAbbreviations: MRS, magnetic resonance spectroscopy: PME, phosphomonoester; AST, aspartate transaminase: ALT, alanine transaminase; PI, inorized serum, clinical, and biopsy values, to detect heganic phosphorus; VOI, volume of interest: PCr, phosphocreatine.
patic abnormalities in patients with diffuse liver
From the 'Department of Radiodiagnosis, 'Department of Gastroenterology,
disease.
,"Department of Pathology, and 'Department of Neurology of the University
The goal of this study was to analyze the possibilities
of 31PMR spectroscopyto detect abnormal hepatic histological changes in patients with diffuse liver disease. 31P
MR spectroscopy was performed, on a 1.5 T whole-body
spectrometer using an image guided localization technique (ISIS), on 38 patients with various diffuse liver
diseases, who all underwent histological and serum
analysis, and 22 healthy volunteers. Phosphornonoester
expressed as a fraction of total phosphorus (PMEP)
showed a correlation with abnormal serum aspartate
transaminase (AST),
histological intralobular degeneratiodfocal necrosis, portal inflammation, and piecemeal
necrosis. We found a lower correlation for PME/P with
fibrosis. It was not possible to differentiate between fibrosis and cirrhosis. In summary, 31P
MR spectroscopy
is a technique to detect intralobular degeneration, inflammation and necrosis and to a less extent fibrosis. No
diagnostic value was found with respect to steatosis and
cholangitis. Furthermore,31PMR spectroscopy is a poor
method for classifying patients into diagnostic categories. (HEPATOLOGY
1995;21:443-449.)
zyxwvutsr
zyx
zyx
zyxwvutsrqpo
Hospital Iltrecht, the Netherlands.
Received January 10, 1994; accepted September 14, 1994.
Address reprint requests to: Jeroen van der Grond, MD, Academic Hospital
Utrecht, Department of Radiology, Heidelberglaan 100, 3584 CX Utrecht, the
Netherlands.
Copyright h 1995 by the American Association for the Study of Liver
Diseases.
0270-9139/95/2102-0027$3.00/0
SUBJECTS AND METHODS
Subjects. 31PMR spectra of the liver were obtained from
22 healthy control subjects (21 to 65 years of age) and 38
patients (17 to 64 years) with diffuse liver disease (Table 1).
Only patients of whom biopsies were performed within 1year
before the spectroscopic examination were included in this
443
zyxwvut
zyxwvutsrq
zyxwvutsrqponm
zyxwvutsr
zyxwvuts
zyxwvutsrqponmlkjih
TABLE1. Individual Histological, Clinical, Serum, and Spectroscopic Patient Data
Disease
Normal
Range
AICAH
B
B
B
B
B
B
B
B
C
C
C
C
C
C
C
C
C
CCH
GVH
PBC
PBC
PBC
PBC
PSC
PSC
PSC
W
W
W
W
W
W
W
W
W
W
W
id
f
0-4
0
pi
0-4
0
3
1
1
0
0
0
1
0
0
0
0
0
0
0
3
1
0
1
1
0
0
0
0
1
0
1
0
0
1
0
0
0
0
0
1
0
0
0
4
3
1
3
4
1
3
3
1
3
0
1
3
4
4
3
1
3
4
1
4
3
4
4
4
4
0
0
3
0
0
1
0
1
3
3
0
1
pbn
0-6
0
0-4
0
0
1
1
5
1
2
1
0
1
0
1
3
1
4
1
1
3
4
0
1
1
1
3
3
6
0
0
1
0
0
0
0
0
1
0
0
0
CP
A-C
aP
ggt
AST
ALT
pmol/L
t100
IUL
t40
pmol/L
<30
pmollL
a1
g L
0
bi
pmol/L
3-17
<30
t35
-
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
0
1
0
1
1
1
0
1
0
0
0
0
0
0
0
0
0
7
14
10
6
11
11
12
5
7
9
5
9
10
8
8
14
8
7
34
69
7
11
208
54
11
20
7
12
6
14
10
8
11
7
16
5
6
6
50
73
46
110
86
48
108
60
45
65
60
103
28
67
61
114
53
130
128
787
200
409
1,642
510
66
430
122
85
90
65
102
81
49
343
134
370
126
90
19
39
24
70
73
21
54
50
18
15
24
32
18
32
62
161
10
62
90
2,460
90
61
526
285
157
160
712
15
63
15
26
57
34
19
72
34
21
55
15
59
63
47
34
27
77
68
25
31
15
70
38
84
48
96
41
79
554
66
35
53
223
98
23
125
61
12
18
10
25
35
19
26
62
24
16
15
11
174
110
130
32
58
88
43
23
58
4
80
55
180
69
250
55
70
41
3 15
55
106
284
106
26
150
226
13
17
10
46
106
27
55
90
46
21
7
41
42
38
35
41
41
32
47
38
37
43
37
38
40
36
40
33
35
39
32
35
33
27
30
29
35
32
40
32
15
39
32
37
36
41
40
40
40
A
A
A
ch
0-2
0
st
0-2
0
0
1
1
0
4
1
3
0
1
0
0
4
4
1
3
4
0
4
1
0
0
1
1
3
4
3
0
1
4
0
3
0
0
1
4
3
1
0
0
2
0
0
0
0
0
0
0
0
0
0
1
1
2
1
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
1
1
0
0
0
1
0
B
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
A
A
B
B
A
A
A
A
A
A
A
A
A
A
B
A
A
A
NTPP
PMEP
0.07-0.19
Pim
0.21-0.29
PDEP
0.33-0.49
0.10-0.22
PH
7.08-7.34
0.14
0.18
0.15
0.16
0.15
0.10
0.22
0.13
0.13
0.18
0.16
0.15
0.15
0.17
0.22
0.20
0.12
0.18
0.09
0.08
0.15
0.08
0.10
0.28
0.15
0.18
0.11
0.14
0.17
0.10
0.09
0.23
0.11
0.14
0.25
0.18
0.11
0.10
0.19
0.13
0.27
0.27
0.21
0.23
0.22
0.18
0.25
0.24
0.20
0.21
0.17
0.23
0.22
0.21
0.23
0.22
0.22
0.22
0.23
0.18
0.14
0.28
0.25
0.16
0.23
0.30
0.22
0.21
0.28
0.06
0.26
0.25
0.26
0.22
0.21
0.20
0.45
0.54
0.40
0.40
0.50
0.47
0.33
0.48
0.45
0.37
0.46
0.48
0.42
0.37
0.40
0.41
0.43
0.40
0.52
0.49
0.47
0.58
0.64
0.28
0.39
0.49
0.49
0.39
0.43
0.46
0.47
0.51
0.50
0.43
0.29
0.42
0.50
0.50
0.22
0.15
0.19
0.17
0.13
0.19
0.23
0.21
0.17
0.21
0.17
0.15
0.26
0.23
0.16
0.18
0.22
0.20
0.17
0.20
0.15
0.15
0.12
0.17
0.22
0.18
0.18
0.17
0.18
0.23
0.16
0.19
0.13
0.18
0.21
0.18
0.18
0.21
7.12
7.23
7.34
7.29
7.19
7.04
7.34
7.44
7.36
7.28
7.26
7.21
7.09
7.43
7.37
7.34
7.02
7.16
7.19
7.29
7.10
7.23
7.42
7.29
7.14
7.09
7.20
7.16
7.05
7.31
7.32
7.17
7.14
7.26
7.10
7.38
7.36
7.07
Abbreviations: AICAH, auto immune chronic active hepatitis; B, hepatitis B; C, hepatitis C; CCH, cryptogene chronic hepatitis; GVH, graft versus host immune disease; PBC,
primary biliar cirrhosis; PSC, primary sclerosing cholangitis; W, Wilson’s disease; pbn, periportal bridging necrosis; id, intralobular degeneration and focal necrosis; pi, portal
inflammation; f, fibrosis; st, steatosis; ch, cholangitis; bi, bilirubin; ap, alkaline phosphatase; ggt, gamma-glutamyl transpeptidase; AST, aspartate transaminase; ALT, alanine
transaminase; al, albumin; CP, Child-Pugh score; PME, phosphomonoester; Pi, inorganic phosphate; PDE, phosphodiester; NTP, 8-P NTP; P, total phosphorus.
HEPATOLOGY
Vol. 21, No. 2, 1995
zyx
zyxwvut
zyxwvutsr
VAN WASSENAER-VAN HALL ET AL
445
TABLE2. Numerical Scoring of Liver Biopsy Specimens
Periportal and Bridging
Necrosis
0 None
1 Mild piecemeal necrosis
3 Moderate piecemeal necrosis
(less than 50% of the
circumference of most portal
tracts)
4 Marked piecemeal necrosis
(more than 50% of the
circumference of most portal
tracts)
5 Moderate piecemeal necrosis
plus bridging necrosis
6 Marked piecemeal necrosis
plus bridging necrosis
10 Multilobular necrosis
Intralobular Degeneration
and Focal Necrosis
None
Mild (acidophilic bodies,
ballooning degeneration,
or scattered foci of
hepatonuclear necrosis
in <: of lobule or
nodules)
Moderate involvement of
<: of lobules or nodules
Marked involvement of
>: of lobules or nodules
Portal Inflammation
Fibrosis
0 None
1 Mild (sprinkling of
inflammatory cells in
<$of portal tracts)
0 None
1 Fibrous portal expansion
3 Moderate (increased
inflammatory cells in
<: of portal tracts)
3 Bridging fibrosis (portal-portal
or portal-central linkage)
4 Marked dense packing of
4 Cirrhosis
inflammatory cells in
>: of portal tracts
zyxwvutsr
Reprinted with permission."
study. This study was limited to patients having diffuse liver
disease with a stable course. Therefore the kind of disease
and severity obtained from the biopsy specimen is likely to
be representative for the whole liver and also representative
for at least on year. Patients showing abnormalities on MRI
or biopsy specimens indicating malignancy were excluded
from this study. Biopsy specimens were scored for periportal
and bridging necrosis, intralobular degeneration and focal
necrosis, portal inflammation, and fibrosis according to Knodell et a1" (Table 2). Histological steatosis was scored 0, 1,
and 2 for absent, mild, and severe, respectively, and for cholangitis, we scored 0 or 1 (absent or present, respectively).
Liver tests (aspartate transaminase [AST], alanine transaminase [ALT], alkaline phosphatase, gamma-glutamyltranspepsidase, bilirubin, albumin, and lactate dehydrogenase)
were measured in all patients within 3 days of the study.
We correlated biopsy scores, liver tests, and the ChildPugh13 classification with phosphomonoester expressed as a
fraction of total phosphorus (PMEP), inorganic phosphate
expressed as a fraction of total phosphorus (PUP), phosphodiester expressed as a fraction of total phosphorus (PDEP),
and nucleoside triphosphate (NTP) expressed as a fraction of
total phosphorus (p-P NTPP). The hepatic NTP signal
mainly (280%) consists of adenosine triphosphate (ATP).
Spectroscopic ratios are expressed as percentage metabolite
of total phosphorus signal, in which total phosphorus was
defined as the signal of PME + Pi + PDE + p-P NTP.
Patients and volunteers participated in this study after
their informed consent was obtained. All subjects fasted for
at least 6 hours before MR spectroscopy examination.
31PMR Spectroscopy. MRS was performed on a 1.5 Tesla
Philips whole-body system. A switchable ('H-31P) surface coil
with a diameter of 15 cm was placed to the right side of
the liver. The coil position was identified by coronal proton
images, made with the body coil. The volume of interest (VOI)
was selected on the basis of the anatomic information of the
coronal as well as axial images. The VOI varied from 200 to
500 mL according to the size of the particular liver. Field
homogeneity was optimized by shimming on the proton signal. Volume-selective "P spectra were recorded using the
ISIS technique with a repetition time of 1,500 msec and 256
measurements. The total duration of the examination was
30 minutes: 7 to 8 minutes patient preparation and positioning of the surface coil, 7 to 8 minutes MR imaging, 7 minutes
shimming, tuning/matching and RF optimization, and 7
minutes MR spectroscopy acquisition time. The averaged free
induction decays were zero filled to 4,096 data points and
processed with a convolution difference procedure (150 Hz)
to remove broad signals from less mobile phospholipids. After
exponential multiplication (4 Hz) and Fourier transformation, a linear phase correction was applied. Subsequent t o
baseline correction, quantification of the metabolites was
achieved by integrating peaks of interest.
Phosphocreatine (PCr) was seen in almost all of the MR
spectra. Although PCr is not present in the liver,3 it can
sometimes be observed because of muscle contamination in
the VOI in the ISIS experiment. pH was derived from the
chemical shift of inorganic phosphate measured relative to
the chemical shift of a-P NTP (referenced a t -7.50 ppm).14
Statistical Analysis. All spectroscopic data are presented
a s mean ? SD. To analyze the differences in mean PMEP, Pi/
P PDEP, and p-P N T P P among groups (grouped according to
serum value, biopsy score or Child-Pugh score), analysis of
variance (ANOVA) was used. If the result of ANOVA was
significant for PMEP, Pi/P PDE/P, or 0-P NTPF, we analyzed this metabolite ratio within the group using the unpaired Student's t-test, which was corrected for repeated
measures. Furthermore, these ratios were compared with
control values using the same test.
We performed ANOVA in which patient groups were defined as follows: For the biopsy scores we used the numerical
scoring of liver biopsy specimens (Table 2),12for steatosis we
used 0, 1, and 2 for absent, mild, and marked, respectively,
for cholangitis, we scored 0 or 1 (absent or present, respec-
zyxwvu
zyxwvutsrqpo
446 VAN WASSENAER-VAN HALL ET AL
HEPATOLOGY
February 1995
zyxwvutsr
z
zyxwvu
zyxwvuts
tively). The liver tests bilirubin and aIbumin we scored two
categories; below or above the upper limit of the reference
range for undiseased controls, for alkaline phosphatase,
gamma-glutamyl transpepsidase, AST, and ALT we scored
three categories; below the upper limit of the reference range
for undiseased controls, below 3 times the upper limit of the
reference range for undiseased controls, and above 3 times
the upper limit of the reference range for undiseased controls.
All normal values used are indicated in Table 1. The ChildPugh classification was scored in categories A, B, or C.13
RESULTS
Diagnostic Categories
Typical MR spectra of a healthy volunteer and of a
patient showing increased P M E P ratio are shown in
Fig. 1A and lB, respectively. The results of the biopsies
and the serum analyses and corresponding metabolite
ratios obtained with MR spectroscopy are summarized
in Table 1. In this table, the spectroscopic results of
the control subjects are expressed as a range extending
2 SD below and above the mean. Figure 2 shows the
distribution of the MR data grouped to diagnostic categories. The shaded area represents the mean metabolite ratios of the control subjects -+ 2 x SD. There was
no significant difference in metabolic ratios or pH between any of the diagnostic categories with control subjects or other diagnostic groups.
the distribution of the P M E P ratio for all diagnostic
categories grouped according to the biopsy score for
periportal and bridging necrosis (Table 2). For periportal and bridging necrosis, we did not find a statistically
significant difference in P M E P between patients with
a biopsy score of 0 and 1(0.13 -+ 0.04, n = 18, and 0.15
t 0.04, n = 14, respectively). We found a significant
difference in the P M E P ratio between patients with a
biopsy score 0 and 3 and up (0.13 -+ 0.04 versus 0.18
-+ 0.05, n = 9, P < .01).
Intralobular Degeneration and Focal Necrosis. Figure 3B shows the distribution of the P M E P ratio for all
diagnostic categories grouped according to the biopsy
score for intralobular degeneration and focal necrosis
(Table 2). For intralobular degeneration and focal necrosis, we found a significant difference in the hepatic
P M E P ratio between patients with a biopsy score of 0
and 1 (P < .0005, P M E P = 0.13 -+ 0.04, n = 26, and
0.19 ? 0.04, n = 10, respectively. Because of the low
number of patients with a biopsy score of 2 (n = 2), we
did not perform statistical tests with the spectroscopic
data from this biopsy score ( P M E P = 0.18 ? 0.04).
Portal Inflammation. Figure 3C shows the distribution of the P M E P ratio for all diagnostic categories
grouped according to the biopsy score for portal inflammation (Table 2). For portal inflammation we did
not find a statistically significant difference in the
P M E P ratio between patients with a biopsy score of 0
and 1 in P M E P (0.11 ? 0.02, n = 7, and 0.13 t 0.04,
n = 9, respectively), whereas the mean P M E P ratio for
patients with biopsy scores 3 and 4 were significantly
increased compared to biopsy score 0 (0.17 -+ 0.04, n
= 12, P < .005, and 0.16 t- 0.05, n = 10, P < .05,
respectively).
Fibrosis. Figure 3D shows the distribution of the
P M E P ratio for all diagnostic categories grouped according to the biopsy score for fibrosis (Table 2). For
fibrosis we did not find a statistically significant difference in the hepatic P M E P ratio between patients with
a biopsy score of 0 and 1(0.13 t 0.04, n = 13, and 0.12
? 0.03, n = 11,respectively). Between patients with a
biopsy score 0 and 3 we found a significant difference
in the P M E P ratio (0.13 -+ 0.04 versus 0.18 +- 0.06, n
= 6, P < ,051. We also found a significant difference in
P M E P between patients with a biopsy score 0 and 4
(cirrhosis) (0.13 t 0.04 versus 0.18 -+ 0.06, n = 8, P <
.05).
zyxwvuts
zyxwvu
zyxwvutsrqp
Biopsies and Serum Values
We found a statistically significant difference (ANOVA) ( P < .05) among the means of the subcategories
of periportal bridging necrosis, intralobular degeneration, portal inflammation, fibrosis, and AST with the
P M E P ratio. This difference was not statistically significant for PUP, PDEP, or 0-P NTPP.
No statistically significant difference (ANOVA) was
found in any of the metabolite ratios among the means
of the subcategories of steatosis, cholangitis, bilirubin,
alkaline phosphatase, gamma-glutamyl transpepsidase, alanine transferase, albumin, and the ChildPugh score.
Periportal and Bridging Necrosis. Figure 3A shows
y-P m
AST
FIG. 1. 31P spectrum of the liver of a healthy volunteer (left)
and a patient showing a n increased hepatic P M E P ratio. Peaks are
assigned to phosphomonoester (PME, 6.5 ppm), inorganic phosphate
(Pi, 4.9 pprn), phosphodiester (PDE, 2.6 ppm), phosphocreatine (PCr,
0 ppm), y-P NTP (-2.6 ppm), LY-P
NTP (-8.0 ppm), and 0-P NTP
(-16.5 ppm).
For AST we found a significant difference in P M E P
between category 1(n = 13) and 2 (n = 20) (1= normal
value and 2 = 1 to 3 times normal value, 0.13 t 0.03
and 0.16 -+ 0.05, P < .02, respectively). Between category 1 and 3 (n = 5) (more than three times normal
value P M E P = 0.17 t 0.081, we did not find a statistically significant difference. We only found a significant
difference in the hepatic P M E P ratio ( P < .02) between
control subjects and category 2. Figure 4 shows the
distribution of the P M E P ratio for all diagnostic categories grouped according to the AST concentration. In
zyxwvutsrqpo
zy
zyxwvuts
zyxwvutsrqp
HEPATOLOGY
Vol. 21, NO.2, 1995
VAN WASSENAER-VAN HALL ET AL
447
7.40
7.20
7.00
0.30
0.20
0.10
0.00
0.80
0.80
0.40
0.20
0.40
0.30
0.20
0.10
0.00
0.30
0.20
zyxwvutsrq
FIG. 2. The individual patient metabolite ratios grouped
according to disease. The shaded
area represents the control
range ? 2 x SD (n = 22).
0.10
I
I
0.00
this figure we excluded 2 AST values as outlyer who
were far outside the normal range. The correlation coefficient between the PMEP ratio and serum AST was
0.45 ( P < .005).
DISCUSSION
Clinical. The most important finding of this report
is that hepatic 31PMRS can detect pathologic processes
such as beginning necrosis and moderate inflammation
but to a less extent for detecting fibrosis. On the contrary, 31PMRS is a poor method for classifying patients
into diagnostic categories.
The increase in PMEP, which is in most literature
attributed to hepatitis alone, is in our study for the
most part associated with histological intralobular degeneratiodfocal necrosis, periportal and bridging necrosis, moderate portal inflammation, and abnormal
serum AST. The finding of a correlation of PMEP with
a combination of these three hepatic histological abnormalities may indicate a correlation with what they
have in common, i.e., tissue damage. Increased serum
levels of AST, a mitochondria1 and cytosolic enzyme
distributed throughout the liver, can imply disturbed
integrity of hepatocytic membranes (leakage) or cellular necrosis. Although generally ALT, a cytosolic enzyme predominantly in the periportal zone, is a better
indicator of hepatitis than AST, we did not find a significant correlation between ALT and the PMEP ratio.
However, in acute hepatocellular necrosis due to chemical or vascular injury, the elevation of AST is often
greater than the elevation of ALT.15 We expected a
correlation between AST and ALT, both serum markers
for all forms of acute and chronic hepatitis, with histological portal inflammation. The absence of this correlation can be explained by the fact that biopsy and
serum analysis was not performed at the same time.
Furthermore, although we limited our study to diffuse
liver disease and did not include focal liver disease, it
should be realized that conclusions based on biopsy
specimens only should be interpreted with care, because of the well-known possibility of sampling errors.
Fig. 3A through 3D shows that the 31PMRS is a good
method to detect pathological processes as periportal
and bridging necrosis, intralobular degeneration and
zyxw
zyxwv
z
zyxwvutsrqpo
zyx
HEPATOLOGY
February 1995
448 VAN WASSENAER-VAN HALL ET AL
B
0.30
0.24
PME/P
1
+
I
0.1 8
0.12
0.06
0.00
0.1 8
zyxwvutsrqpo
zy
* I
0.12
1
0.06
**
0
1
2
***
0.00
3
4
5
0
6
Periportal Bridging Necrosis
C
0
A
0
1
2
3
4
Intralobular Degeneration
D
0.30
I
PME/P
0.24
A
0.24
0
Q
0
0.1 8
0.1 8
0.1 2
0.1 2
0.06
0.00
1
I'
+
* *
0.06
** *
0.00
0
1
2
3
4
FIG. 3. The distribution of
the P M E P ratios for all diagnostic categories grouped according
to the biopsy score for periportal
and bridging necrosis (A), intralobular degeneration and focal
necrosis (B), portal inflammation (C), and fibrosis (D). The two
horizontal lines represent the
control range for the P M E P ratio 5 2 x SD (n = 22). A = autoimmune chronic active hepatitis,
A = hepatitis B, 0 = hepatitis
C, = cryptogene chronic hepatitis, 0 = graft versus host immune disease, = primary biliary cirrhosis, 0 = primary
= Wilsclerosing cholangitis,
son's disease. Statistical indices
represent the significance between each histological subgroup and the normal control
values; * = P < .05, ** = P <
,005, *** = P < ,0005. In A, the
biopsy scores from category 2-6
were treated as one group.
0
Portal Inflammation
focal necrosis, and portal inflammation, but less for
detecting fibrosis. These findings are in agreement with
earlier
that suggested that an increased
PME was mainly correlated to viral hepatitis and not
to fibrosis. Cox et aI4 demonstrated that there was no
change in the PME/ATP ratio in patients with liver
cirrhosis only. However, when cirrhosis was accompanied by hepatitis, a significant increase in PMEfATP
was observed. Nevertheless, other studies show that
both hepatitis and cirrhosis contribute to an increased
PME.3,5The differences for MR spectroscopy to diagnose liver abnormalities in these studies may be caused
by the very broad patient inclusion criteria. Although
we also found a statistical difference (P < .005) in PME/
P between hepatitis patients and control subjects (0.16
2 0.03 and 0.13 2 0.04, respectively), our study shows
that hepatic 31PMRS is also a technique to detect differences between mild, moderate or severe inflammation, and also between differences between mild and
severe fibrosis. Therefore, patient selection that is only
based on the presence or absence of inflammation (hepatitis) may introduce a lot of patient bias.
1
2
Fibrosis
3
4
zy
Technical. The PME peak is composed mainly from
resonances that arise from phosphorylcholine (PC) and
phosphorylethanolame (PE),which are both precursors
in phospholipid biosynthesis. In small amounts, sugar
phosphates and adenosine monophosphate contribute
to the PME peak.16 Increased levels of phospholipid
precursors in general point at an increased phospholipid biosynthesis, and the increase in PMEP in hepatic inflammation probably reflects hepatocyte regeneration, which is likely to occur in inflammation and
necrosis.
Although we realize that presentation of absolute
metabolite concentrations is preferred, in this study
metabolite ratios rather than absolute metabolic concentrations are measured. Measurements involving the
calculation of absolute concentrations require individual TI measurements. To calculate saturation factors,
the MRS experiments need to be repeated with at least
one different repetition time. Unfortunately, because
of the physical and mental condition of the patients,
they would not stand such a long-lasting effort. However, especially because the TI values of liver metabo-
zyxwvuts
zy
z
z
zyxwvutsrqp
HEPATOLOGY
Vol. 21, No. 2, 1995
VAN WASSENAER-VAN HALL ET AL
REFERENCES
0.30
PME/P
0.00
449
1. Rajanayagam V, Lee RR, Ackerman Z, Bradley WG, Ross BD.
Quantitative P-31 spectroscopy of the liver in alcoholic cirrhosis.
J Magn Reson Imag 1992;2:183-190.
2. Meyerhoff DJ, Boska MD, Thomas AM, Weiner MW. Alcoholic
liver disease: quantitative image-guided P-31 MR spectroscopy.
Radiology 1989; 173:393-400.
3. Oberhaensli R, Rajagopalan B, Galloway GJ, Taylor DJ, Radda
GK. Study of human liver disease with P-31 magnetic resonance
spectroscopy. Gut 1990;31:463-467.
4. Cox IJ, Menon DK, Sargentoni J , Bryant DJ, Collins AG, Coutts
GA, Iles RA, et al. P-31 magnetic resonance spectroscopy of the
human liver using chemicalshift imaging techniques. J Hepatol
1992; 141265-275.
5. Munakata T, Griffiths RD, Martin PA, Jenkins SA, Shields R,
Edwards RHT. An in uivo 31P MRS study of patients with liver
cirrhosis: progress towards a non-invasive assessment of disease
severity. NMR Biomed 1993;6:168-172.
6. Angus WA, Dixon RM, Rajagopalan B, Ryley NG, Simpson KJ,
Peters TJ, Jewel1 DP, et al. A study of patients with alcoholic
liver disease by 31-P nuclear magnetic resonance spectroscopy.
Clin Sci 1990;78:33-38.
7. Cox IJ, Bryant DJ, Collins AG, George P, Harman RR, Hall AS,
Hodgson HJF, et al. Four dimensional chemical shift MR imaging of phosphorus metabolites of normal and diseased human
liver. J Comput Assist Tomogr 1988; 12(3):369-376.
8. Brinkmann G, Melchert UH. A study of T1-weighted 31P MR
spectroscopy from patients with focal and diffuse liver disease.
Magn Reson Imag 1992; 10:949-956.
9. Meyerhoff DJ, Karczmar GS, Matson GB, Boska MD, Weiner
MW. Non invasive quantitation of human liver metabolites using
image guided 31Pmagnetic resonance spectroscopy. NMR Biomeb1%0;3:17-22. 10. Cox IJ. Coutts. Gadian DG. Ghosh P. Sareentoni J. Youne IR.
Saturation effects in P-31 'magnetic resonance spectra ofy the
human liver. Magn Reson Med 1991; 17:53-61.
11. Buchthal SD, Thoma WJ, Taylor JS, Nelson S J , Brown TR. In
viuo T1 values of phosphorus metabolites in human liver and
muscle determined a t 1.5 T by chemical shift imaging. NMR
Biomed 1989;5/6:298-304.
12. Knodell RG, Ishak KG, Black WC, Chen TS, Craig R, Kaplowitz
N, Kiernan TW, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. HEPATOLOGY
1981;5:431-435.
13. Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams
R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646-649.
14. Taylor DJ, Bore PJ, Styles P, Gadian DG, Radda GK. Bioenergetics of intact human muscle: a 31P nuclear magnetic resonance
study. Mol Biol Med 1983; 1:77-94.
15. Presig R, Tygstrup N, Price C. Assessment of liver function
search tests. In Millward-Sadler GH, Wright R, Arthur MJP,
eds. Wright's liver and biliary disease. Vol 1, Ed 3. London: Saunders, 1992.
16. Meyerhoff DJ, Karczmar GS, Weiner MW. Abnormalities of the
liver evaluated by
magnetic resonance spectroscopy. Invest
Radio1 1989;24:980-984.
17. Evelhoch JL, Ewy CS, Siegfried BA, Ackerman JJH, Rice DW,
Briggs RW. 31Pspin-lattice relaxation times and resonance linewidths of r a t tissue in vivo: dependence upon the static magnetic field strength. Magn Reson Med 1985;2:410-417.
18. Stark D, Mosely ME, Bacon BR, Moss AA, Goldberg HI, Bass
NM, James TL. Magnetic resonance imaging and spectroscopy
of hepatic iron overload. Radiology 1985;154:137-142.
19. Roth K, Hubesch B, Meyerhoff DJ, Naruse S, Gober JB, Lawry
TJ, Boska MD, et al. Non-invasive quantitation of phosphorus
metabolites in human tissue by NMR spectroscopy. J Magn Reson 1989;81:299-310.
20. Iles RA, Cox IJ, Bell JD, Dubowitz LMS, Cowan F, Bryant DJ.
3 1-P Magnetic resonance spectroscopy of the human paediatric
liver. NMR Biomed 1990;3:90-94.
zyxw
zyxwvutsrqpo
zyxwvutsrqp
'
0
I
100
Aspartate Transferase (microM/L)
FIG.4. Cartesian plot showing the correlation between the PME/
P ratio and serum AST. The correlation coefficient between the PMEI
P ratio and serum AST was 0.45 (P < ,005). A = autoimmune chronic
active hepatitis, A = hepatitis B, 0 = hepatitis C, 0 = cryptogene
chronic hepatitis, 0 = graft versus host immune disease, W = primary biliary cirrhosis, 0 = primary sclerosing cholangitis,
=
Wilson's disease.
+
lites are expected to be influenced by deposition of paramagnetic Fe or Cu,17-19 individual T1 relaxation
measurements may be very useful. Munakata et a15
determined that there was a trend of a prolongation of
T1 for all metabolites in cirrhotic patients compared
with normal subjects. A prolonged T1 value, in general,
causes a reduced signal intensity using the same repetition time. Using the data of Munukata e t al also a
reduced P M E P ratio would exist in the cirrhotic liver
compared with normal liver, based on the reduced T1
only. Therefore, the observed increase in P M E P with
increased periportal and bridging necrosis, intralobular degeneration and focal necrosis, portal inflammation, fibrosis and AST is probably caused by an increased concentration of PME components, because a
prolonged Tl only would decrease the P M E P ratio.
Nevertheless, even if saturation factors are known,
comparison between different studies would still be difficult. Differences in field strength, technical methods
(Spectroscopic Imaging or ISIS), and quantification
methods are of great influence on the final ratios. Furthermore, age-related changes in the P M E P ratio cannot be excluded,20although our adult control group did
not show an age-related correlation with spectral data.
CONCLUSION
This study shows that 31PMRS of the liver is a poor
method for classifying patients into diagnostic categories, but might be useful to detect markers of liver
tissue damage, being abnormal serum AST, intralobular degeneration, piecemeal necrosis, and portal inflammation.
__
zyxw