GASTROENTEROLOGY 78:1389-1392, 1980
Hepatic Transaminase Activity in
Alcoholic Liver Disease
DANIEL S. MATLOFF, MITCHELL
MARSHALL M. KAPLAN
Gastroenterology Service, Department of Medicine, New England Medical Center
Hospital, Boston, Massachusetts
Glutamic oxaloacetic transaminase (GOT) and glu-
tamic pyruvic transaminase (CPT) activities were
measured in percutaneous needle biopsy specimens
of human liver tissue and compared with trans-
aminase values in serum obtained on the day of
biopsy. Hepatic CPT activity was significantly de-
creased in liver tissue of patients with alcoholic
hepatitis and cirrhosis compared with the activity in
individuals with normal livers (P < 0.05) and indi-
viduals with primary biliary cirrhosis (P < 0.05). The
decreased hepatic CPT activity was not related to
the presence of cirrhosis in biopsy specimens and
was not increased by the addition of saturating
amounts of pyridoxal phosphate to the assay mix-
ture. Hepatic GOT was also slightly but significantly
lowered in individuals with alcoholic liver disease
(P <; 0.05). The GOT/CPT ratio in serum and liver
tissue was increased only in individuals with alco-
holic liver disease, but the increase did not reach
statistical significance. The increased GOT/CPT ra-
tio is · due primarily to the low activity of CPT in
liver 'and serur;n. The less than expected elevation of
CPT in serum of patients with alcoholic hepatitis re-
flects the diminished hepatic CPT activity and lesser
amounts of this enzyme available to leak into serum
from damaged hepatocytes.
Alcoholic liver disease is characterized by distinct
patterns of serum glutamic oxaloacetic transaminase
Received September 14, 1979. Accepted January 20, 1980.
Address requests for reprints to: Marshall M. Kaplan, M.D.,
Gastroenterology Service, New England Medical Center Hospital,
171 Harrison Avenue, Boston, Massachusetts 02111.
This study was presented in part at the Annual Meeting of the
American Gastroenterological Association, May, 1979, in New
Orleans, Louisiana, and published in abstract form in GASTRO-
ENTEROLOGY (76:1195, 1979).
This work was supported in part by Research Grant AM 10571
and Training Grant AM 07024 from the National Institutes of
Health.
© 1980 by the American Gastroenterological Association
0016-5085/80/061389-04$02.25
J. SELINGER, and
(SGOT) and serum glutamic pyruvic transaminase
(SGPT) elevation. 1 ' 2 We have previously demon-
strated that an SGOT /SGPT ratio greater than 2 is
highly suggestive of alcoholic hepatitis or cirrhosis. 1
In addition, SGOT and SGPT values greater than
300 IU are rare in patients with alcoholic liver dis-
ease and correlate poorly with the degree of hepatic
inflammation and liver cell necrosis seen on liver
biopsy! Values may be minimally elevated in the
face of extensive liver cell necrosis. The mecha-
nisms responsible for these clinically useful obser-
vations are not clear. Various factors, including tox-
icity of ethanol to liver mitochondria, damage to
nonhepatic tissue such as skeletal muscle, and vita-
min deficiencies have been proposed as causes but
not proven.
The aim of the present study was to investigate
the mechanism of the characteristic SGOT/SGPT
pattern in patients with alcoholic liver disease. For
this purpose, fragments of percutaneous liver biopsy
material were obtained from patients with various
types of liver disease, including alcoholic liver dis-
ease and the specific activity of glutamic oxaloacetic
transaminase (GOT) and glutamic pyruvic trans-
aminase (GPT) measured in liver tissue.
Materials and Methods
Patients
A 5-10-mg piece of liver, approximately 10-20% of
the total specimen, was obtained prospectively from hos-
pitalized patients undergoing diagnostic percutaneous
liver biopsy. Informed consent was obtained from each
patient. The remaining 80-90% of the specimen was sent
for routine histology. All biopsies were performed using
the 16-gauge, thin-walled Klatskin needle (Becton, Dickin-
son & Co., Orangeburg, N.Y., No. 1403). Within 6 hr of the
biopsy, 5 cm 3 of .serum was obtained from each patient.
All liver biopsies were reviewed by one of us (M. M..
Kaplan) without knowledge of the transaminase values in
1390 MA TLOFF ET AL. GASTROENTEROLOGYVol. 78, No. 6

GPT GOT
150

~
500
c:

!'::
'
<D

w
f-
0
0::
100
t 400 -o
::0
0
--i
0.. rn
Figure 1. CPT and GOT activity in human liver tis- 0'
300 z
sue. NL = normal liver histology; FL = '::>
fatty liver with no history of et hanol 50 200
abuse; ALD =alcoholic liver disease, both
alcoholic hepatitis and cirrhosis; PBC = •. 100
primary biliary cirrhosis; CAH = chronic
active hepatitis; and CPH = chronic per-
sistent hepatitis (n =number of patients). NL FL ALD PBC CAH CPH NL FL ALD PBC CAH CPH

the liver tissue. Patients with the following diagnoses were
included in the study: normal liver biopsies (11 patients
whose diagnostic liver biopsies yielded histologically nor-
mal liver tissue), fatty liver with no history of alcohol
abuse (8 patients), alcoholic liver disease (9 patients), pri-
mary biliary cirrhosis (13 patients), chronic active hepa-
titis (9 patients) and chronic persistent hepatitis (4 pa-
tients) . Diagnoses were based on clinical data and
currently accepted histologic criteria.
SGPT were measured in the hospital's Clinical .C hemistry
laboratory by the Karmen method. .
Statistical Analysis
Statistical comparisons were carried out by analy-
sis of variance-using a program written . fora Data Gen-
eral Corporation Nova 830 computer. Valu.~s are expressed
as mean ± 1 SEM.

Assays .
Liver tissue was immediately washed in ice-cold
0.25 M sucrose, weighed, and homogenized in 1.0 ml of
0.25 M sucrose in an all-glass Ten Broeck tissue grinder.
The crude liver homogenate and serum samples were
stored at -70°C. Glutamic oxabacetic transaminase was
measured by the method of Karmen 3 and GPT measured
by the method of Wroblewski and LaDue.• Both assays
were modified to achieve optimal substrate concentra-
tions.5 Final assay volumes were 1.0 ml. In additional stud-
ies, assays were repeated with the addition of pyridoxal
phosphate, final concentration 0.1 mM. 6 Protein was mea-
sured by the method of Lowry et al.' One unit of enzyme
was defined as that amount which converted 1 ~tmol of
substrate to end product/min. Both assays were linear
with respect to time and protein concentration. Enzyme
activity was expressed as units/gram protein. SGOT and
Results
Liver GOT and GPT specific activities are
given in Figure 1. There was no significant differ-
ence in the protein content of any · of the patient
groups with protein expressed as micrograms/milli-
gram wet weight liver. The specific activity of GPT
in normal liver was 133 ± 18 U/g protein. Hepatic
GPT activity in patients with alcoholic liver disease
was 41% of normal (P < 0.05) and also significantly
lower than that in patients with primary biliary cir-
rhosis (P < 0.05). Decreased h epatic GPT activity
was found only in patients with alcoholic hepatitis
and cirrhosis.
The specific activity of GOT in normal liver was
495 ± 40 U/g protein. Tissue GOT activity in alco-

LIVER SERUM ::0

f- 6 2 ~
0.. 0
(.!)
_J (f)
G)
'f-
0
0
(.!) ~
Figure 2. GOT /GPT ratios in liver and serum of pa- _J (f)

~
G)
tients with liver disease . The abbrevi- -~ 3 I -o
ations are given in the legend of Figure 1. 0::
0
mm --i

f.li
~
The liver and serum GOT /CPT ratio of
each patient was calculated. The ratios
shown in the figure are the mean of the
individual ratios. NL FL ALD PBC CAH CPH NL FL ALD PBC CAH CPH
June 1980
holic hepatitis and cirrhosis was 71% of normal (P <
0.05). Tissue GOT activity was also reduced in pa-
tients with chronic active hepatitis (75% of normal),
but the difference did not reach statistical signifi-
cance (P < 0.2).
Of the 9 patients with alcoholic liver disease, 6
had acute alcoholic hepatitis without cirrhosis, and 3
had unequivocal cirrhosis. Equally low tissue GPT
values were found in the alcoholic hepatits patients
without cirrhosis (48 ± 15 U/g protein) as in those
with cirrhosis (69 ± 22 U/g protein). Hepatic GPT
activity in patients with chronic hepatitis and pri-
mary biliary cirrhosis whose liver biopsies demon-
strated unequivocal cirrhosis were similar to values
in those patients without cirrhosis. Hence, cirrhosis
per se did not account for the lowered hepatic GPT
activity. The addition of saturating amounts of the
GPT and GOT cofactor, pyridoxal phosphate, to the
assay mixtures did not increase GPT or GOT activ-
ity in liver tissue of patients with alcoholic liver dis-
ease and chronic active hepatitis. Thus, the lowered
GPT activity in patients with alcoholic liver disease
was not due to a deficiency of this cofactor in the as-
say mixture or in liver tissue.
Table llists the mean SGOT and SGPT values in
each group of patients. Figure 2 demonstrates the
GOT /GPT ratios in serum and liver tissue of pa-
tients with the different types of liver disease. Serum
and tissue ratios are increased only in patients with
alcoholic liver disease, compared with those with
normal liver histology and other types of liver dis-
ease, but the difference is statistically significant
only in serum (P < 0.025).
Discussion
This study demonstrates that liver GPT activ-
ity is selectively decreased in patients with alcoholic
liver disease compared with individuals with normal
livers and other types of liver disease. The decreased
liver GPT activity is specific for alcoholic liver dis-
ease and not due to chronicity of liver disease or cir-
rhosis per se. Hepatic GOT activity is also decreased
in alcoholic liver disease but to a lesser extent than
GPT. In addition, the diminished hepatic GOT activ-
ity is not as specific, because it is also decreased in
patients with chronic hepatitis. Although there was
no statistical difference between the normal and
chronic active hepatitis group, a true difference may
have been missed because of the small sample size
and wide variation in individual values.
These data help explain the characteristics serum
transaminase pattern in patients with alcoholic
hepatitis and cirrhosis, only modest elevations of
SGOT and SGPT accompanied by an elevated
SGOT/SGPT ratio. In alcoholic hepatitis and cir-
HEPATIC TRANSAMINASES IN ALCOHOLIC HEPATITIS 1391
Table 1. Serum Transaminase Values in Patients with
Liver Disease
No. SCOT SGPT
Normal liver biopsies 11 35 ± 7 55± 18
Fatty liver 8 38±4 40 ± 12
Alcoholic liver disease 9 107 ± 30 41 ±9
Primary biliary cirrhosis 13 58.5 ± 16 81 ± 24
Chronic active hepatitis 9 202 ± 70 281 ± 105
Chronic persistent hepatitis 4 88± 13 91 ± 8
rhosis, the less than expected elevations of SGOT
and SGPT reflect diminished hepatic activity of
these enzymes. There is, accordingly, less enzyme
available for leakage into serum from damaged or
destroyed hepatocytes.
The cause of the lowered hepatic GPT activity in
patient with alcoholic liver disease is not known.
However, available data suggest that it may be re-
lated to chronic pyridoxal phosphate deficiency. Lu-
meng and Li have shown that pyridoxal phosphate
deficiency is common in patients with chronic alco-
hol abuse." They measured serum pyridoxal phos-
phate levels in 66 alcoholic subjects without evi-
dence of liver or hematologic disease and found
lowered levels in 35. Their data suggested that the
lowered levels were due to increased catabolism of
pyridoxal phosphate mediated not by alcohol, but
by acetaldehyde, a product of alcohol metabolism. 8
Decreased levels of pyridoxal phosphate have also
been found in both serum and liver tissue of patients
with alcoholic fatty liver and cirrhosis. 9 ' 10 Deficient
dietary intake in alcoholics may also contribute to
the decreased serum pyridoxal phosphate concen-
tration in this population, but there is little data
bearing on this.
Although there are no studies directly relating
serum or hepatic pyridoxal phosphate levels to he-
patic GPT activity in humans, data in rats do suggest
such a relationship. Ludwig and Kaplowitz have re-
cently reported that hepatic GPT activity was low-
ered by 47% in rats maintained for 8 wk on a vitamin
Be deficient diet. 11 Vitamin Be deficiency appeared to
inhibit GPT synthesis since addition of pyridoxal
phosphate to assay mixtures did not increase GPT
activity. We obtained similar results in our patients.
Addition of excess amounts of pyridoxal phosphate
to GPT assays in human liver homogenates did not
increase GPT activity. That pyridoxal phosphate
deficiency in rats can be induced by chronic alcohol
feeding is suggested by the studies of Henley et al. 12
They noted that ethanol, given as the sole source of
fluid in rats on an otherwise adequate diet, lowered
hepatic GPT activity significantly compared with
pair fed rats, P < 0.001. After 24 wk on such a diet,
the hepatic GOT /GPT ratio was significantly in-
1392 MATLOFF ET AL.
creased in the alcohol fed rats, due entirely to the
decrease in hepatic GPT activity. These investiga-
tors did not measure pyridoxal phosphate metabo-
lism. However, extrapolation of data derived from
alcoholic patients suggests that there may be a rela-
tionship between chronic alcohol abuse, pyridoxal
phosphate deficiency, and lowered levels of hepatic
GPT activity. 8 Unfortunately, our data do not pro-
vide a definite answer. We did not measure serum or
hepatic vitamin B6 content in our patients and did
not perform liver biopsies in alcoholic patients with-
out liver disease. It should be possible, however, to
test this hypothesis in prospective studies and deter-
mine if large amounts of supplemental vitamin B6
will restore the hepatic and serum transaminase pat-
tern to normal.
References
1. Cohen JA, Kaplan MM: The SCOT/SGPT ratio-An indicator
of alcoholic liver disease. Am J Dig Dis 24:835-838, 1979
2. DeRitis, F, Coltorti M, Giusti G: Serum transaminase activities
in liver disease. Lancet 1:685, 1972
3. Karmen A: A note on the spectrophotometric assay of glu-
tamic-oxaloacetric transaminase in human blood serum. J
Clin Invest 34:131, 1955
GASTROENTEROLOGY Vol. 78, No.6
4. Wroblewski F, LaDue JS: Serum glutamic pyruvic trans-
aminase in cardiac and hepatic disease. Proc Soc Exp Bioi
Med 9:569, 1956
5. Bergmeyer HA, Bernt E: Glutamate-oxaloacetate trans-
aminase and glutamate-pyruvate transaminase. In: Methods
of Enzymatic Analysis. Edited by HU Bergmeyer. New York
and London, Academic Press, 1965, p 837-853
6. H¢rder M, Moore RE, Bowers GN Jr: Aspartate amino-
transferase activity in human serum. Factors to be considered
in supplementation with pyridoxal 5' -phospate in vitro. Clin
Chern 22 :1876, 1976
7. Lowry OH, Rosebrough NJ, Farr AI, et al: Protein measure-
ment with the Falin phenol reagent. J Bioi Chern 193:265, 1951
8. Lumeng L, Li TK: Vitamin B6 metabolism in chronic alcohol
abuse. Pyridoxal phosphate synthesis and degradation in hu-
man erythrocytes. J Clin Invest 53:693, 1974
9. Hines JD, Love DS: Determination of serum and blood pyri-
doxal phosphage concentrations with purified rabbit skeletal
muscle apophosphorylase b. J Lab Clin Med 73:343, 1969
10. Frank 0, Luisada-Opper A, Sorrell MF, et al: Vitamin deficits
in severe alcoholic fatty liver of man calculated from multiple
reference units. Exp Mol Pathol 15:191, 1971
11. Ludwig S, Kaplowitz N: Effect of pyridoxine deficiency (B 6 -D)
on serum and liver transaminase (T) in experimental liver in-
jury. Gastroenterology 76:1290, 1979
12. Henley KS, Wiggins HS, Hirschowitz GI, et al: The effect of
oral ethanol on glutamic pyruvic and glutamic oxaloacetic
transaminase activity in the rat liver. QJ Stud Alcohol 19:54,
1958