Complement and immunoglobulin levels in serum and ascitic fluid of patients with spontaneous bacterial peritonitis, malignant ascites, and tuberculous peritonitis.

ABSTRACT

Background. We determined complement and immunoglobulin levels in ascitic fluid and serum of 47 patients with spontaneous bacterial peritonitis, malignant ascites, or tuberculous ascites.

Methods. Paracentesis was done to confirm the underlying cause of ascites, Biochemical, hematologic, and microbiologic investigations were also done.

Results. The highest serum and ascitic fluid C3 and C4 levels and ascitic fluid IgM, IgA, and IgG levels were found in patients with tuberculosis. Ascitic fluid C3 level was found to be higher in the tuberculous group than in the patients with spontaneous bacterial peritonitis or malignant ascites. Ascitic fluid 04 levels were higher in patients with tuberculosis than in those with spontaneous bacterial peritonitis.

Conclusion. We believe that further studies of the in vivo kinetics of immunoglobulins and complement in ascitic fluid of various causes are necessary for a better understanding of the host defense mechanisms of these fluids.

**********

SPONTANEOUS BACTERIAL PERITONITIS (SBP) is a frequent complication of cirrhotic ascites, whereas it is rare in other types of ascites, especially that of malignant origin. Bacteremia is common among patients with severe acute or chronic liver disease. At the time of spontaneous ascitic fluid infection, blood cultures are found to grow bacteria in 54% of patients. (1-3) Many patients with cirrhosis, especially those with ascites, are complement-deficient. The opsonic activity of the ascitic fluid has been shown to correlate closely with the total protein concentration, and a deficiency of ascitic fluid opsonic activity has been proposed as the factor that results in "susceptible ascites." (1,4)

The complement components have been shown by several investigators to play an important role in the development of immunologically mediated inflammatory reactions.5 Several studies in the English-language literature have determined complement and immunoglobulin levels in patients with cirrhosis and spontaneous bacterial peritonitis, (3,6,9) but few such studies have been done in patients with tuberculosis. In our study, we measured complement and immunoglobulin levels in serum and ascitic fluid of patients with spontaneous bacterial peritonitis, malignant ascites, and tuberculous peritonitis.

PATIENTS AND METHODS

Between January 1999 and February 2000, a total of 47 patients (27 male and 20 female) with ascites had studies of serum and ascitic fluid complement and immunoglobulin levels. Paracentesis was done in all of the patients to confirm the underlying cause of ascites. Biochemical, hematologic, and microbiologic investigations were also done, as well as complete blood count, biochemical examinations, ultrasonography, and cytologic examination of the ascitic fluid.

Spontaneous Bacterial Peritonitis (Group 1)

The 13 patients with SBP (8 male and 5 female) had a mean age of 51.56 [+ or -] 10.45 years and a history of chronic liver disease due to hepatitis B (9 cases) or hepatitis C (4 cases). According to the Child-Pugh classification, 8 of the cases were group C and 5 were group B. The patients with SBP had either a positive microbiologic culture or a neutrophil count of >500/[mm.sup.3] in the ascitic fluid. Culture was positive for Escherichia coli in 7 patients and enterococci in 3. Escherichia coli was detected in 2 patients with a temperature of >38[degrees]C.

Malignancy-Induced Ascites (Group 2)

Of the 21 patients with malignancy, 12 were male and 9 were female, and their mean age was 52.2 [+ or -] 9.15 years. The diagnoses were based on histopathologic examination of the liver, the omentum, or the mass. Malignancies included 5 ovarian, 4 hepatoma, 3 colon, 3 mesothelioma, 2 gastric, 2 breast, 1 cholangiocellular, and 1 pancreatic carcinoma. In 8 of these patients, liver metastasis was detected by ultrasonographic biopsy.

Tuberculous Peritonitis (Group 3)

A total of 13 patients (7 male and 6 female) with a mean age of 49.6 [+ or -] 12.05 years had tuberculous peritonitis, diagnosed by histopathologic examination of the peritoneum after diagnostic laparoscopy. All patients were receiving diuretic therapy, and none had therapeutic paracentesis. Blood and ascitic fluid samples were taken simultaneously, using polypropylene tubes. Blood samples were coagulated completely, and after centrifugation, serum samples were stored deep-frozen. Biochemical parameters and immunoglobulin levels were analyzed by Olympus AU 600 (Olympus, Hamburg, Germany) with Olympus kits. Complement levels were determined by use of commercially available test kits, with normal ranges as follows: IgG, 8.0 to 17.0 g/L; IgA, 1.0 to 4.9 g/L; IgM, 0.5 to 3.2 g/L; C3, 0.5 to 0.9 g/L; and C4, 0.1 to 0.4 g/L.

Statistical Analysis

Statistical analysis was done with SPSS statistical software. Data are presented as the means [+ or -] standard deviation. Intergroup comparisons were analyzed by the Mann-Whitney test and Kruskal-Wallis one-way ANOVA test. The Spear-man test was used for correlation. P values < .05 were considered to be statistically significant.

RESULTS

Serum and ascitic fluid C3, C4, IgA, IgM, IgG levels were determined in the 3 groups described. Results are summarized in the Table.

Ascitic fluid studies showed no statistically significant difference in mean total leukocyte, neutrophil, and lymphocyte counts between the three groups. Ascitic fluid albumin levels were significantly elevated in group 3 when compared with group 1 (P = .02). No statistically significant difference was found in ascitic fluid globulin, serum albumin, or serum globulin levels between the three groups. Ascitic fluid protein levels were significantly elevated in group 3 when compared with group 1 (P = .027), and group 2 (P = .047). Serum protein levels were significantly higher in group 3 than in group 1 (P = .044). Serum/ascitic fluid albumin gradient was found to be significantly higher in group 2 than in group 3 (P = .014). Serum glucose levels were significantly elevated in group 1 (P = .001) and group 2 (P= .036) when compared with group 3. Ascitic fluid glucose levels were significantly elevated in group 1 (P = .011) and group 2 (P = .003) when compared with group 3.

Serum C3 levels were higher in group 2 than in group 1 (P = .006) and higher in group 1 than in group 3 (P = .011). Ascitic fluid C3 levels were higher in group 3 than in group 1 (P= .006) and group 2 (P= .036).

Serum C4 levels were higher in group 2 (P = .007) and group 3 (P = .002) than in group 1 and higher in group 3 than in group 2 (P = .047). Ascitic fluid C4 level was higher in group 3 than in group 1 (P = .043).

Serum IgM levels were higher in group 3 than in group 2 (P = .041). Ascitic fluid IgM levels did not differ between the groups. Serum and ascitic fluid IgA and IgG levels also did not differ between the groups.

In group 1, there were positive correlations between ascitic fluid C4 and ascitic fluid C3 (r = 0.768, P= .026), between serum C3 and ascitic fluid C3 (r = 0.766, P = .027), between serum C4 and ascitic fluid C4 (r = 0.763, P = .027), and between serum IgG and ascitic fluid IgG (r= 0.761, P = .028).

In group 2, there were positive correlations between ascitic fluid globulin and ascitic fluid IgA (r = 0.880, P = .001), ascitic fluid IgM (r=0.758, P = .001), ascitic fluid IgG (r = 0.904, P = .001), ascitic fluid C3 (r = 0.901, P = .001), and ascitic fluid C4 (r = 0.781, P = .001). Positive correlation was also detected between ascitic fluid albumin and ascitic fluid IgA (r = 0.809, P = .001), 1gM (r = 0.766, P = .001), IgG (r = 0.833, P = .001), C3 (r = 0.863, P = .001), and C4 (r = 0.712, P = .001). The positive correlations were also detected between ascitic fluid protein and ascitic fluid IgA (r = 0.842, P = .001), 1gM (r = 0.773, P = .001), IgG (r = 0.864, P = .001), C3 (r = 0.886, P = .001), and C4 (r = 0.716, P = .001). We have also detected positive correlations between serum C3 and ascitic fluid C3 (r = 0.631, P = .007) and between serum C4 and ascitic fluid C4 (r = 0.522, P = .032). There were negative correlations between serum/ascitic fluid albumin gradient and ascitic fluid C3 (r = -0.800, P = .001), C4 r = -0.709, P = .001), IgA (r = -0.832, P = .001), and IgG (r = -0.734, P = .001).

In group 3, there were positive correlations between serum IgM and ascitic fluid IgM (r = 0.945, P = .001). There were negative correlations between serum C3 and ascitic fluid C3 (r = -0.962, P = .001); between serum C4 and ascitic fluid C4 (r = -0.860, P = .006); between ascitic fluid globulin and ascitic fluid IgA (r = -0.800, P = .017); between ascitic fluid protein and ascitic fluid IgA (r = -0.910, P = .001), IgM (r = -0.800, P = .017), and IgG (r = -0.800, P = .017); and between ascitic fluid albumin and ascitic fluid IgA (r = -0.910, P = .001), IgM (r = -0.800, P = .017), and IgG (r = -0.800, P = .017).

DISCUSSION

Spontaneous bacterial peritonitis is a severe and frequent complication of cirrhosis with a high mortality rate. (10,11) Spontaneous bacterial peritonitis is probably related to several impaired defense mechanisms, such as depressed reticuloendothelial system phagocytic activity, leukocyte dysfunction, reduced serum complement, and low bacterial activity of ascitic fluid. Infection of ascitic fluid is related to its antimicrobial activity. In cirrhotic patients, the bactericidal and opsonic activity of the ascitic fluid is lower than that observed in noncirrhotic ascites or in normal peritoneal exudate (3,12,13) Recent investigations have shown that cirrhotic patients with low ascitic fluid C3 and low ascitic fluid total protein are more predisposed to SBP. (6-9) In fact, a decrease in the ascitic fluid C3 levels during episodes of SBP and in the serum C3 during bacteremia has also been reported. (14,15) A reduction in serum complement levels and a disturbed opsonization (endogenous antimicrobial activity), e ither related or not to the hypocomplementemia, has been described also, albeit not uniformly in patients with chronic liver disease.(16,17)

The increase in serum C3 and ascitic fluid C3 levels may be caused by several mechanisms, such as inhibition of serum and ascitic fluid complement-factor consumption and/or improved hepatic complement synthesis. (18) The activation of the alternative complement pathway by bacteria or their products in blood or ascitic fluid accounts for this consumption.

The increase of serum and ascitic fluid C3 levels observed by Such et al (8) in patients having selective intestinal decontamination also seems to support this hypothesis. Changes in hepatic synthesis capacity may be another mechanism by which serum and ascitic fluid C3 concentration could change. (19) However, no significant variations in serum and ascitic fluid C4 and albumin levels were observed in any of the groups, though C4 and albumin are also synthesized in the liver. (20)

Akalin et al (3) reported complement and immunoglobulin concentrations in cirrhotic ascites to be significantly lower than noncirrhotic ascites and normal (nonascitic) peritoneal fluid. Concentrations of complement were similar in normal peritoneal fluid and in ascitic fluid in noncirrhotic patients. When cirrhotic ascitic fluids were compared with normal peritoneal fluid values, only IgM was found to be significantly low. It has been postulated that the massive ascites that often occurs in cirrhotic patients dilutes these components. However, if dilution is an important factor, Akalin et al (3) should also have found low complement and immunoglobulin levels in noncirrhotic ascites. Instead, they found higher concentrations of these components, all of these patients had massive ascites.

The presumed mechanism for the decrease of C3 but not C4 in the ascitic fluid is a combination of dilution, low hepatic synthesis, and greater consumption of C3 due to an activation of the alternative complement pathway. (6-9,17-20) The other possibility is acceleratered catabolism or consumption of complement in the ascitic fluid. Local consumption of complement in infected pleural effusions has been shown. (21) If there is an increased consumption of complement in ascites, it should be in cirrhotic ascites, because values of complement were similar in fluids obtained from patients with ascites due to causes other than cirrhosis and in normal peritoneal fluid. Runyon (22) reported that cirrhotic patients with an ascitic fluid total protein value less than 1 g/dL are at greater risk for SBP and actually had SBP during hospitalization 10 times more frequently than patients with higher protein levels in the ascitic fluid. In contrast, almost half the patients with low ascitic fluid C3 levels became infected. Ot hers have reported, however, that reduced serum C3 and C4 levels were found more frequently in cirrhotic patients with SBP than in cirrhotic patients without SBP.

Tuberculous peritonitis is one of the common causes of ascites in Third World countries. (23) To diagnose this condition, a simple screening test with high sensitivity and specificity would be most desirable. The English-language literature contains limited data concerning the complement and immunoglobulin levels in patients with tuberculous peritonitis. In our study, we detected the highest serum and ascitic fluid C3 and C4 levels and the highest ascitic fluid IgM, IgA, and IgG levels in patients with tuberculous peritonitis.

Chen et al (24) reported that the ascitic fluid and serum C3, C4, and protein levels were significantly lower in patients with SBP than in those without SBP. Wang et al (25) found that the ascitic fluid level of C3 or C4 was significantly higher in patients with peritoneal carcinomatosis than in patients with sterile cirrhotic ascites, SBP, or hepatocellular carcinoma.

The source of immunoglobulins and complements in normal peritoneal fluid and in cirrhotic or noncirrhotic ascites is not clear. It has been shown that high concentrations of specific immunoglobulins can be produced in experimental ascites. (26) If there is local production of immunoglobulins in humans, that may also be impaired in cirrhotic patients. Serum immunoglobulin levels were high in both groups but were similar in the patients with and without SBP, suggesting that the serum level of most immunoglobulin classes has little or no impact on the susceptibility for SBP in individuals with advanced liver disease. (7) Mal et al (6) reported elevated IgG, C3, and C4 concentrations that were significantly higher in patients with malignant ascites than in cirrhotic patients. Valicenti et al (27) reported that levels of immunoglobulins, especially IgG and IgA, are usually high in neoplastic effusions.

We believe further studies of the in vivo kinetics of immunoglobulins and complement in ascitic fluid of various causes are necessary for a better understanding of the host defense mechanisms of these fluids.

TABLE

Serum and Ascitic Fluid Findings in Patients With Spontaneous Bacterial
Peritonitis (Group 1), Malignant Ascites (Group 2), and Tuberculous
Peritonitis (Group 3).

                                                     Group 1

Parameter                                            (n = 13)
Age (yr)                                        51.56 [+ or -] 10.45
Serum/ascitic fluid albumin gradient             1.44 [+ or -] 0.46
Serum albumin (mg/dL)                            2.25 [+ or -] 0.21
Serum globulin (mg/dL)                           3.64 [+ or -] 0.39
Serum protein (mg/dL)                            5.89 [+ or -] 0.54
Serum glucose (mg/dL)                            96.0 [+ or -] 9.65
Serum C4 (g/L)                                   0.17 [+ or -] 0.05
Serum C3 (g/L)                                   0.95 [+ or -] 0.21
Serum IgA (g/L)                                  5.17 [+ or -] 0.68
Serum IgM (g/L)                                  1.97 [+ or -] 0.34
Serum IgG (g/L)                                 19.24 [+ or -] 2.16
Ascitic fluid albumin (mg/dL)                    0.81 [+ or -] 0.26
Ascitic fluid globulin (mg/dL)                   1.55 [+ or -] 0.55
Ascitic fluid protein (mg/dL)                    1.86 [+ or -] 0.43
Ascitic fluid glucose (mg/dL)                   91.38 [+ or -] 6.82
Ascitic fluid leukocyte count (/[mm.sup.3])   1,137.5 [+ or -] 363.5
Ascitic fluid neutrophil count (/[mm.sup.3])    545.0 [+ or -] 214.7
Ascitic fluid lymphocyte count (/[mm.sup.3])    267.0 [+ or -] 185.0
Ascitic fluid C4 (g/L)                           0.08 [+ or -] 0.01
Ascitic fluid C3 (g/L)                           0.33 [+ or -] 0.09
Ascitic fluid IgA (g/L)                          1.53 [+ or -] 0.25
Ascitic fluid IgM (g/L)                          0.47 [+ or -] 0.08
Ascitic fluid IgG (g/L)                          7.45 [+ or -] 1.24

                                                    Group 2

Parameter                                           (n = 21)
Age (yr)                                       52.2 [+ or -] 9.15
Serum/ascitic fluid albumin gradient           1.46 [+ or -] 0.15
Serum albumin (mg/dL)                          2.71 [+ or -] 0.19
Serum globulin (mg/dL)                         3.68 [+ or -] 0.17
Serum protein (mg/dL)                          6.38 [+ or -] 0.24
Serum glucose (mg/dL)                         85.47 [+ or -] 4.2
Serum C4 (g/L)                                 0.38 [+ or -] 0.04
Serum C3 (g/L)                                 1.71 [+ or -] 0.15
Serum IgA (g/L)                                3.83 [+ or -] 0.29
Serum IgM (g/L)                                1.52 [+ or -] 0.12
Serum IgG (g/L)                               15.61 [+ or -] 1.12
Ascitic fluid albumin (mg/dL)                  1.25 [+ or -] 0.25
Ascitic fluid globulin (mg/dL)                 1.41 [+ or -] 0.21
Ascitic fluid protein (mg/dL)                  2.66 [+ or -] 0.44
Ascitic fluid glucose (mg/dL)                 92.35 [+ or -] 4.64
Ascitic fluid leukocyte count (/[mm.sup.3])   867.9 [+ or -] 292.8
Ascitic fluid neutrophil count (/[mm.sup.3])  498.0 [+ or -] 198.8
Ascitic fluid lymphocyte count (/[mm.sup.3])  226.7 [+ or -] 161.5
Ascitic fluid C4 (g/L)                          0.1 [+ or -] 0.01
Ascitic fluid C3 (g/L)                         0.52 [+ or -] 0.07
Ascitic fluid IgA (g/L)                         1.2 [+ or -] 0.14
Ascitic fluid IgM (g/L)                        0.37 [+ or -] 0.07
Ascitic fluid IgG (g/L)                        7.43 [+ or -] 0.93

                                                    Group 3

Parameter                                           (n = 13)
Age (yr)                                       49.4 [+ or -] 12.05
Serum/ascitic fluid albumin gradient            .67 [+ or -] 0.21
Serum albumin (mg/dL)                          2.55 [+ or -] 0.1
Serum globulin (mg/dL)                         4.11 [+ or -] 0.05
Serum protein (mg/dL)                          6.69 [+ or -] 0.06
Serum glucose (mg/dL)                          70.5 [+ or -] 0.50
Serum C4 (g/L)                                 0.52 [+ or -] 0.05
Serum C3 (g/L)                                 1.88 [+ or -] 0.11
Serum IgA (g/L)                                3.68 [+ or -] 0.84
Serum IgM (g/L)                                1.97 [+ or -] 0.13
Serum IgG (g/L)                               13.62 [+ or -] 1.64
Ascitic fluid albumin (mg/dL)                  1.88 [+ or -] 0.25
Ascitic fluid globulin (mg/dL)                 2.18 [+ or -] 0.35
Ascitic fluid protein (mg/dL)                  4.05 [+ or -] 0.56
Ascitic fluid glucose (mg/dL)                  65.5 [+ or -] 3.47
Ascitic fluid leukocyte count (/[mm.sup.3])   540.0 [+ or -] 109.4
Ascitic fluid neutrophil count (/[mm.sup.3])  368.0 [+ or -] 175.5
Ascitic fluid lymphocyte count (/[mm.sup.3])  179.5 [+ or -] 105.6
Ascitic fluid C4 (g/L)                         0.16 [+ or -] 0.04
Ascitic fluid C3 (g/L)                         0.87 [+ or -] 0.15
Ascitic fluid IgA (g/L)                        1.55 [+ or -] 0.31
Ascitic fluid IgM (g/L)                         0.7 [+ or -] 0.14
Ascitic fluid IgG (g/L)                        7.57 [+ or -] 1.46

References

(1.) Hoefs JC, Runyon BA: Spontaneous bacterial peritonitis. Dis Mon 1985; 31:1-48

(2.) Graudal N, Milman N, Kirkegaard E, et al: Bacteremia in cirrhosis of the liver. Liver 1986; 6:297-301

(3.) Akalin HE, Laleli Y, Telatar H: Bactericidal and opsonic activity of ascitic fluid from cirrhotic and noncirrhotic patients. J Infect Dis 1983; 147:1011-1017

(4.) Runyon BA: Patients with deficient ascitic fluid opsonic activity are predisposed to spontaneous bacterial peritonitis. Hepatology 1988; 8:632-635

(5.) Jose P: Complement-derived peptide mediators of inflammation. Br Med Bull 1987; 43:336-349

(6.) Mal F, Huu TP, Bendahou M, et al: Chemoattractant and opsonic activity in ascitic fluid. a study in 47 patients with cirrhosis or malignant peritonitis. J Hepatol 1991; 12:45-49

(7.) Rabinovitz M, Gavaler JS, Kumar S, et al: Role of serum complement, immunoglobulins, and cell-mediated immune system in the pathogenesis of spontaneous bacterial peritonitis. Dig Dis Sci 1989; 34:1547-1552

(8.) Such J, Guarner C, Soriano G, et al: Selective intestinal decontamination increases serum and ascitic fluid C3 levels in cirrhosis. Hepatology 1990; 12:1175-1178

(9.) Runyon BA, Morrissey RL, Hoefs JC, et al: Opsonic activity of human ascitic fluid: a potentially important protective mechanism against spontaneous bacterial peritonitis. Hepatology 1985; 5:634-637

(10.) Hoefs JC, Canawati HN, Sapico FL, et al: Spontaneous bacterial peritonitis. Hepatology 1982; 2:399-407

(11.) Pinzello G, Simonetti RG, Craxi A, et al: Spontaneous bacterial peritonitis: a prospective investigation in predominantly nonalcoholic cirrhotic patients. Hepatology 1983; 3:545-549

(12.) Simberkoff MS, Moldover NH, Weiss G: Bactericidal and opsonic activity of cirrhotic ascites and nonascitic peritoneal fluid. J Lab Clin Med 1978; 91:831-839

(13.) Bercovici B, Michel J, Miller J, et al: Antimicrobial activity of human peritoneal fluid. Surg Gynecol Obstet 1975; 141:885-887

(14.) Fromkes JJ, Thomas FB, Kraut EH: Activation of the alternative complement pathway in ascitic fluid during spontaneous bacterial peritonitis. J Clin Gastroenterol 1982; 4:347-350

(15.) Fearon DT, Ruddy S, Schur PH, et al: Activation of the properdin pathway of complement in patients with gram negative bacteremia. N Engl J Med 1975; 292:937-940

(16.) Potter BJ, Trueman AM, Jones EA: Serum complement in chronic liver disease. Gut 1973; 14:451-456

(17.) Fox RA, Dudley FS, Sherlock S: The serum concentrations of the third component of the complement in liver disease. Gut 1971; 12:574-578

(18.) Fromkes JJ, Thomas FB, Mekhjian HS, et al: Antimicrobial activity of human ascitic fluid. Gastroenterology 1977; 73:668-672

(19.) Colter HR, Alper CA, Rosen FS: Current concepts in immunology: genetics and biosynthesis of complement proteins. N Engl J Med 1981; 304:653-656

(20.) Hoefs JC: Increase in ascites white blood cell and protein concentrations during diuresis in patients with chronic liver disease. Hepatology 1981; 1:249-254

(21.) Lew PD, Zubler R, Vaudaux P, et al: Decreased heat-labile opsonic activity and complement levels associated with evidence of C3 breakdown products in infected pleural effusions. J Clin Invest 1979; 63:326-334

(22.) Runyon BA: Low-protein concentration ascitic fluid is predisposed to spontaneous bacterial peritonitis. Gastroenterology 1986; 91:1343-1346

(23.) Dwivedi M, Misra SP, Misra V, et al: Value of adenosine deaminase estimation in the diagnosis of tuberculous ascites. Am J Gastroenterol 1990; 85:1123-1125

(24.) Chen SM, Lo GH, Lai KH, et al: Serum and ascitic concentration of C3, C4 and protein in cirrhotic patients with spontaneous bacterial peritonitis. Chung Hua I Hsueh Tsa Chih (Taipei) 1994; 54:87-92

(25.) Wang SS, Lee FY, Chao Y, et al: Clinical significance of complements in ascitic diseases: elevated complement levels disapproving the liver disease origin. Proc Nail Sci Counc Repub China B 1996; 20:51-57

(26.) Munoz J: Production in mice of large volumes of ascites fluid containing antibodies. Proc Soc Exp Biol Med 1957; 95:757-759

(27.) Valicenti PA, Torino ML, Osatinsky R, et al: Hemopexin and immunoglobulins in benign and malignant ascites. experience in 66 patients. Ada Gastroenterol Latinoam 1986; 16:81-92

RELATED ARTICLE: KEY POINTS

* The highest serum and ascitic fluid C3 and C4 levels and ascitic fluid IgM, IgA, and IgG levels were found in patients with tuberculosis.

*Ascitic fluid C3 level was found to be higher in the tuberculous group than in the patients with spontaneous bacterial peritonitis or malignant ascites.

*Ascitic fluid C4 levels were higher in patients with tuberculosis than in those with spontaneous bacterial peritonitis.

*An elevated ascitic fluid IgM level is a marker for active infection in spontaneous bacterial peritonitis and tuberculosis.

From the Departments of Internal Medicine and Gastroenterology, Inonu University School of Medicine and Turgut Ozal Medical Center, Malatya, Turkey.

Reprint requests to Ramazan Sari, MD, Akdeniz University School of Medicine, Department of Internal Medicine, Division of Endocrinology and Metabolism, Antalya TR-07070, Turkey.