ACTA FAC. MED. NAISS. 2003; 20 (4): 209-212

Original article

THE VALUE OF SERUM-ASCITES ALBUMIN GRADIENT IN DIFFERENTIAL DIAGNOSIS OF ASCITES AND THE PROPOSAL FOR THE NEW CUT-OFF VALUE
 

Goran Bjelaković, Aleksandar Nagorni , Ivanka Stamenković , Daniela Benedeto-Stojanov, Vesna Brzački, Suzana Raičević, Biljana Radovanović¹, Marija Bjelaković², Vuka Katić, Vesna Živković ³, Dragomir Vučetić⁴

¹Clinic of Gastroenterology and Hepatology, Clinical Centre Niš,
²Institute of Anatomy,
³Institute of Pathology,
⁴Medical Faculty, University of Niš, Clinic of Gynecology and Obstetrics, Clinical Center Niš

BACKGROUND

Ascites is the accumulation of free intraperitoneal fluid, and commonly encountered clinical finding. Over 90% of the cases are caused by one of four conditions: liver cirrhosis, neoplasm, congestive heart failure, or tuberculosis. Traditionally, several biochemical parameters (glucose, cholesterol, total proteins) were determined in the ascitic fluid and used for differential diagnosis of ascites. Pare and co-workers reported that the serum-ascites albumin gradient, SAAG was good biochemical marker of portal hypertension (1). The SAAG, based on oncotic-hydrostatic balance, is calculated by measuring the albumin concentration of ascitic fluid and subtracting it from the serum value obtained at the same time. If the SAAG is greater than 11 g/L, the patent has portal hypertension, and conversely, if SAAG is less than 11 g/L, there is no portal hypertension. It has been shown that it categorizes ascites better than other biochemical parameters (2-5). This study evaluates the value of serum-ascites albumin concentration gradient in differential diagnosis of ascites.

PATIENTS AND METHODS

This prospective study included 171 patients with ascites (130 with cirrhotic and 41 with malignant ascites) hospitalized at Clinic of Gastroenterology and Hepatology, Clinical Centre in Ni{. After clinical examination, the patients underwent laboratory and instrumental methods, selectively applied. Different imaging studies were performed to assess the cause of ascites formation e.g. chronic liver disease, liver metastasis, and neoplasm. The localization of tumor was determined by laparotomy or autopsy, and for patients who did not undergo one of these procedures, an opinion was made based on the imaging modalities. The diagnosis of cirrhosis was established on the basis of the clinical examination, biochemical test and instrumental examination and/or liver biopsy. Patients underwent abdominal paracentesis in the first 24 hours after the admission. The aseptic technique was used, with a 22 gauge needle, in the left lower abdominal quadrant, and the samples of ascitic fluid were immediately conducted to biochemical, cytological and microbiological laboratory to be analyzed. At the same time blood samples were taken for simultaneous ascitic fluid and blood determination of albumin concentration. Ascitic fluid and blood were examined for albumin concentration, applying standard biochemical methods. Smears of ascitic fluid were fixed and stained with Hematoxylin-Eosin and Papanicolaou and microscopically examined for their cellular content. The sediment from 40 ml. of ascitic fluid was cultured for mycobacteria. The Student's t-test was used for statistical analysis of the data. Results are expressed as mean standard deviation. The Receiver Operating Characteristic curve analysis, Analyze-It Software, Ltd was used to calculate the cut-off value for serum-ascites albumin gradient.
 

RESULTS

After the examinations had been carried out, the patients were divided into two groups with cirrhotic ascites and malignant ascites. A total number of 171 patients with ascites, 130 with cirrhotic, mean age 60 years, and 41 with malignant ascites, mean age 63 years were studied. In the group with malignant ascites 6 patients (14.63%) had liver metastasis, but only two of them (4.88%), massive liver metastasis that caused portal hypertension. The mean value of SAAG (- SD) in the group of patients with cirrhotic ascites was 21.89 - 8.34 g/L and was statistically significant, higher than SAAG in group with malignant ascites of 11.17 - 7.13 g/L, (p < 0.001) (Figure 1). The cut-off value for SAAG of 11 g/L had high sensitivity (97.56%) but low specificity (46.34%), so we decide to determine new cut-off value using ROC analysis. With the probability of p < 0.05, the interval of reliability was 11.2-19.7 g/L with a cut-off value of 15.86 g/L (Figure 2). We conclude that the cut-off value for the SAAG has to be corrected to a higher level to achieve the maximal sensitivity and specificity, and thus help in differential diagnosis of ascites.
 

DISCUSSION

Diagnostic paracentesis became increasingly important over the last two decades as the key initial test in the assessment of the ascitic patient(7). Biochemical analyses of ascitic fluid is of great importance for etiological diagnosis of ascites. The distinction between cirrhotic and malignant ascites has important therapeutic and prognostic implications. This study was designed to determine the diagnostic value of serum-ascites albumin gradient (SAAG) in patients with ascites. Hoefs first reported a significant correlation between SAAG and portal venous pressure in patients with chronic liver disease (3). Similarly, Rector reported a highly significant correlation between SAAG as well as the portal pressure in patients with alcoholic liver disease (2). Contrary to this observation, Kajani et al. (6) reported no significant correlation between measured portal venous pressure and SAAG, and the three of their patients had a SAAG <1.1 g/dL (3), similar to the study of Al-Knawy et al. in which five of liver disease patients had a SAAG <1.1 g/dL, despite the presence of esophageal varices by gastroscopy, which indicates the presence of portal hypertension (4). In our study, two patients with malignant ascites (4.88%) had SAAG higher than 11 g/L. Moreover, the patients had massive liver metastasis. This is a rare occasion when malignant ascites has values greater than 11 g/L due to the portal hypertension caused by an extensive replacement of normal liver tissue by tumor. Our earlier study (8) confirmed that SAAG has no provision for patients with two causes for ascites formation (i.e. "mixed" ascites). Usually, these patients have cirrhosis including one additional cause such as peritoneal carcinomatosis, peritoneal tuberculosis or hepatocellular carcinoma, and SAAG higher than 11 g/L. Chen et al. found that while the serum ascites albumin gradient offered the best diagnostic accuracy and specificity in the diagnosis of malignancy-related ascites its sensitivity was not good enough. They indicated that serum ascites albumin gradient and tumor markers are not sensitive parameters in the diagnosis of malignancy-related ascites (9).
We concluded that SAAG was a useful diagnostic parameter for the differential diagnosis of ascites, which could be used to separate ascites caused by portal hypertension from other causes of ascites without portal hypertension. We proposed the new cut-off value for discriminating the patients with and without portal hypertension of 16 g/L. This value seems to have higher sensitivity and specificity, than the previous value of 11 g/L.
 

REFERENCES

1. Pare P, Talbot J, Hoefs JC. Serum-ascites albumin concentration gradient: a physiologic approach to the differential diagnosis of ascites. Gastroenterology 1983;85:240-244.
2. Rector WG Jr., Reynolds TB. Superiority of the serum-ascites albumin difference over ascites total protein concentration in separation of "transudative" and "exudative" ascites. Am J Med 1984;47:83-85.
3. Hoefs JC, Serum protein concentration and portal pressure determine the ascitic fluid protein concentration in patients with chronic liver disease. J Lab Clin Med 1983;102:260-273.
4. Al-Knawy BA. Etiology of ascites and the diagnostic value of serum-ascites albumin gradient in non-alcohol liver disease. Ann Saudi Med 1997;17:26-28.
5. Runyon BA, Montano AA, Acaridiasis EA, et al. The serum-ascites albumin gradient is superior to exudate-transudate concept in differential diagnosis of ascites. Ann Intern Med 1992;117:215-220.
6. Kajani MA, Young YK, Alexander JA, et al. Serum-ascites albumin gradients in nonalcoholic liver disease. Dig Dis Sci 1990;35:33-37.
7. Hoefs, JC. Diagnostic paracentesis A potent clinical tool Gastroenterology 1990;98:230-236.
8. Bjelaković G, Tasić T, Stamenković I et al. Biochemical, cytological and microbiological characteristics of the cirrhotic, malignant and "mixed" ascites. Arch Oncol 2001; 9:95-101.
9. Chen SJ, Wang SS, Lu CW et al. Clinical value of tumeur markers and serum-ascites albumin gradient in the diagnosis of malignancy-related ascites. Journal of Gastroenterology & Hepatology 1994;9: 396-400.