ORIGINAL ARTICLE

Serum/Ascites Albumin Gradient in

Differential Diagnosis of Ascites
M Beg*, S Husain**, N Ahmad***, N Akhtar*

Abstract
Aim
The classification of ascites into ‘transudative’ and ‘exudative’ has recently been challenged. The present study was
aimed to differentiate ascites on the basis of serum/ascites albumin gradient, a proposed biochemical criteria for
differential diagnosis of ascites and also to compare its diagnostic accuracy with the traditional marker : ascitic fluid
total proteins, classifying ascitic fluid into transudate and exudate.

Material & method

Paired ascitic fluid and serum samples from 100 patients were examined with an established method for the diagnosis
of cause of ascites. The present study included 76 patients having ascites related to portal hypertension (cirrhosis - 54,
cardiac - 10, secondary bacterial peritonitis - 6, liver metastasis - 6), and 24 patients of tubercular ascites not related
to portal hypertension.

Results
The diagnostic accuracy of SAAG and AFTP were 96% and 68% respectively.

Conclusion
Differential diagnosis of ascites should be based on the serum/ascites albumin gradient which is a better distinguishing
marker for separating ascites related to portal hypertension from all other causes of ascitic fluid collection, irrespective
of infection.

Key words
Ascites, Serum/ascites albumin gradient, Ascitic fluid total protein.

Introduction clinical conditions especially in cirrhotic patients

Ascites is one of the most common amongst the
various clinical problems confronting a
physician, and ascitic fluid analysis is the most
effective way to diagnose it. The traditional
classification of ascites into ‘exudative’ and
‘transudative’ involves estimation of ascitic fluid
total protein (AFTP), which is high ≥ 2.5 gm/dL
in exudate and < 2.5 gm/dL in transudate 1.
This classification, however, is unable to
correctly identify the aetiological factors
responsible for its causation 2,3 and has been
challenged on various occasions in different
* Lecturer
** Resident
*** Senior Resident
Department of Medicine,
JN Medical College,
Aligarh Muslim University, Aligarh-202 002.
on prolonged diuretic theray4, cardiac ascites5,
1/3rd patients of malignant ascites 6 ,
spontaneous bacterial peritonitis 7 , and
sometimes even in normal ascitic fluid 8 .
Moreover, it offers little insight to the
pathophysiology of ascitic fluid formation9.
Further, these drawbacks led to development of a
new approach to classify ascites, based on
albumin gradient between plasma and ascites. In
presence of portal hypertension, oncotic pressure
gradient between plasma and ascitic fluid has to
be raised, to counter-balance the high hydrostatic
pressure driving the fluid to the intraperitoneal
cavity10. Albumin being the single most important
factor of oncotic pressure generation, the
difference between the serum and ascitic albumin
concentration (serum/ascites albumin gradient -
SAAG) was used to differentiate ascitic fluid into
categories : gradient ≥ 1.1 g/dl in cases with portal
hypertension and < 1.1 g/dl in ascites unrelated
to portal hypertension9,10. Various studies have
demonstrated superiority of SAAG in classifying
ascites compared to transudate-exudate concept
but with conflicting observations11,12,13. There have
also been reports arguing against superiority of
SAAG compared with other markers used for
differentiation of ascites into transudate and
exudate especially in non-alcoholic liver disease14.
In view of the above, the present study was
undertaken to evaluate the value of SAAG in the
differential diagnosis of ascites and also to
compare its sensitivity and diagnostic accuracy with
that of AFTP.
Material and method
The present prospective study included 100
patients of ascites. In 76 patients the cause of
ascites was related to portal hypertension. Out of
54 patients, 29 (53.7%) had post-hepatitic, 9
(16.6%) alcoholic, and 16 (29.6%) cryptogenic
cirrhosis. Six patients of cardiomyopathy, 2 each
of corpulmonale, and valvular heart diseases
contributed to cardiac ascites. While contribution
of SBP and malignancy to group I of study was six
each. Twenty four patients of tubercular ascites
were unrelated to portal hypertension and
comprised group II of the study.
Table I : Comparison of AFTP and SAAG in
study groups.
Groups AFTP (g/dl) SAAG (g/dl)
< 2 . 5 ≥ 2.5 < 1 . 1 ≥ 1.1
Group I
Cirrhotics 42 12 4 50
Cardiac failure 0 10 0 10
SBP 6 0 0 6
Liver metastases 4 2 0 6
Group II
Tubercular 0 24 24 0 which is considered exudate according to
Ascitic fluid was collected in all patients by
paracentesis done under sterile condition using
21 gauge needle. The routine testing of ascitic
52 Journal, Indian Academy of Clinical Medicine
fluid included total protein, albumin and cell count.
The same tests were done on blood sample drawn
at the time of abdominal paracentesis. The specific
investigations like ascitic fluid culture, liver biopsy,
upper gastrointestinal endoscopy, lipid profile,
echocardiography, etc., were performed as
required. The statistical analysis was done using
student’s ‘t’ test. The diagnostic accuracy was
calculated as the sum of the true positive plus true
negative results divided by the total number of
cases.
Results
Ascitic fluid total protein were significantly lower
in group I patients (table I and II) compared to
the patients of group II. Albumin gradient was
≥ 1.1 gm/dL in 72 patients of group I, while
AFTP was < 2.5 gm/dL in 52 patients of this
group (table II). All the patients of group II had
serum gradient of < 1.1 gm/dL (Table I).
Therefore it was observed that the serum
albumin ascitic gradient had a diagnostic
sensitivity of 94.73% and 96% accuracy
compared to AFTP, which is 65.62% and 68%
respectively (Table I, II).
Discussion
The results of present study show that the serum/
ascitic fluid albumin is a useful marker for the
diagnosis of ascites, as it has diagnostic accuracy
of 96%. Similar observations have been also
reported by other studies11,12,13. If the gradient is
> 1.1 gm/dl, the underlying cause is almost
always related to portal hypertension. The
application of albumin gradient disregards the
concept of transudate versus exudate as it provides
a more rational approach, separating ascitic fluid
into two categories on the basis of the presence
or absence of portal hypertension11. The albumin
gradient retains its ability even in infected ascites,
traditional concept, although it usually develops
in patients of cirrhosis, which, owing to the low
ascitic fluid total protein concentration, is
traditionally labelled as transudative ascites. In

Vol. 2, No. 1 and 2
January-June 2001
fact, ascites with low protein concentration is more
prone to develop infection7. The results of the
present study reinforce the conclusions of the
reports which showed that albumin gradient is
superior to the transudate-exudate concept in
classifying ascitic fluid collections of varied
aetiology11,13. The utility of albumin gradient in
non-alcoholic liver disease has been debated14.
However, in the present study the test was found
to have significant diagnostic accuracy in ascites
caused by both alcoholic and non-alcoholic liver
disease. The high albumin gradients in cardiac
failure patients is also a manifestation of an
elevated portal pressure due to increased inferior
vena caval pressure5 and also in malignant ascites
it is the elevated portal pressure due to metastasis
in liver and peritoneum which is responsible for
increased albumin gradients6. This is explained
on the basis of the equilibrium of starting forces,
when ascites is related to portal hypertension, this
increments of portal pressure should be counter
balanced by an increased difference of osmotic
forces (and thus albumin concentration)
between serum and ascites. Since serum
albumin is already low in decompensated liver
disease; this lead to the well known very low
albumin ascitic fluid concentrations in patients
with cirrhosis. High serum/ascites albumin
gradients values indicate higher levels of portal
hypertension3.
Table II : Diagnostic sensitivity and accuracy of SAAG compared to AFTP in study groups.
Group I Group II
AFTP 1.80 + 1.05 3.8 + 0.93
(gm/dl)
SAAG 1.41 + 0.65 0.71 + 0.27
(gm/dl)
Conclusion
The results of the present study show that the
serum/ascites albumin gradient is a test with
significant diagnostic accuracy in separating
Journal, Indian Academy of Clinical Medicine
Vol. 2, No. 1 and 2
ascites related to portal hypertension from the
forms of ascitic fluid collection caused by
mechanisms unrelated to portal hypertension. It
does not provide exact cause of ascites. The
presence of high albumin gradient only means,
the presence of portal hypertension. It is superior
to previously proposed transudate-exudate
concept, not only because of its higher diagnostic
accuracy but also because it provides a better
approach to pathogenesis of ascitic fluid collection.
The transudative-exudative ascites should be
replaced with the ascites related to portal
hypertension (high gradient) and ascites not
related to portal hypertension (low gradient)
respectively.
References
1. Rovelstad RA, Bartholomew LG, Cain JC et al. The value
of examination of ascitic fluid and blood for lipids and
for proteins by electrophoresis. Gastroenterology 1958;
34: 436-50.
2. Sampliner RE, Iber FL. High protein ascites in patients
with uncomplicaed hepatic cirrhosis. Am J Med Sci 1974;
267: 275-9.
3. Rector WG, Reynolds TB. Superiority of the serum-ascites
alubumin difference over the ascites total protein
concentration in separation of ‘transudative’ and
‘exudative’ ascites. Am J Med 1984; 77: 83-5.
4. Hoef JC. Increase in ascitic WBC and protein
concentration during diuresis in patients with chronic liver
disease. Hepatology 1981; 1: 249-54.
5. Runyon BA. Cardiac ascites : A characterization. J Clin
Diagnostic
p value
Sensitivity Accuracy
< 0.001 65.62% 68%
< 0.001 94.73% 96%
Gastroenterol 1988; 10: 410-2.
6. Runyon BA, Hoefs JC, Morgan TR. Ascitic fluid analysis
in malignancy-related ascites. Hepatology 1988; 8:
1104-09.
7. Runyon BA. Low protein concentration ascitic fluid is

January-June 2001 53
 predisposed to spontaneous bacterial peritonitis.
Gastroenterology 1986; 91: 1343-6.
8. Bouckaert PX, Evers JLH, Doesberg WH et al. Patterns of
change in proteins in the peritoneal fluid of women
during the periovulatory phase of the menstrual cycle. J
Reprod Fertile 1986; 77: 329-36.
9. Rector WG. An improved diagnostic approach to ascites
(editorial). Arch Intern Med 1987; 147: 215.
10. Pare P, Talbot J, Hoefs JC. Serum-ascites albumin
concentration gradient : A physiologic approach to the
differential diagnosis of ascites. Gastroenterology 1983;
85: 240-4.
11. Runyon BA, Montano AA, Akriviadis EA et al. The serum-
ascites albumin gradient is superior to the exudate-
transudate concept in the differential diagnosis of ascites.
With Best Compliments from
ALEMBIC
Presents Comprehensive Diabetic Range
GLZ 80/40
(Gliclazide Tab)
ALA-100
(Alpha Lipoic Acid)
&
Sionara 100/200 (Celecoxib 100/200mg Cap)
GOLD STANDARD IN ARTHRITIS MANAGEMENT
54 Journal, Indian Academy of Clinical Medicine
Ann Intern Med 1992; 117: 215-20.
12. Goyal AK, Goyal SK, Pokharna DS, Sharma SK.
Differential diagnosis of ascitic fluid : Evaluation and
comparison of various biochemical criteria with a special
reference to serum ascites albumin concentration
gradient and its relation to portal pressure. Tropical
Gastroenterology 1989; 10 (1): 512-55.
13. Gupta R, Misra SP, Dwivedi M et al. Diagnostic ascites :
Value of total protein albumin, cholesterol their ratios
serum ascites albumin and cholesterol gradient. J
Gastroenterology Hepatology 1995; 10 (3): 295-9.
14. Kajani MA, Yoo YK, Alexander JA et al. Serum-ascites
albumin gradients in nonalcoholic liver disease. Dig Dis
Sci 1990; 35: 33-7.
GLZ Plus
(Gliclazide + Metformin)
80mg 500mg
Forminal 500/850
(Metformin)

Vol. 2, No. 1 and 2
January-June 2001