Selective Intestinal Decontamination Increases Serum and Ascitic Fluid C3 Levels in Cirrhosis JOSE! SUCH,' CARLOS GUARNER,'GERMAN SORIANO,' MONTSERRAT TEWD6,' JOSE! BARRIOS,~ FRANCISCO TENA,' CARhlEN M~NDEZ,' JAIME ENRfQUEZ,' JOSfi LUISRODRfGUEZ3 AND FRANCISCO VILARDELL' 'Escuela de Patologia Digestiva, 'Unidad de Enfermedades Infecciosas and 3Servicio de Inmunologia, Hospital de la Santa Creu i Sant Pau, 08025 Barcelona, Spain

Selective intestinal decontaminationfor 7 days with norfloxacin was performed in 14 cirrhotic patients with ascites and low ascitic fluid total protein. Variations in serum and ascitic fluid of C3 and C4 and ascitic fluid total protein after therapy were compared with those of a control group of 14 untreated patients with similar characteristics. After oral nodoxacin administration, we saw a significant increase of C3 in serum (p c 0.05) and ascitic fluid (p = 0.01). A significant increase was also observed in ascitic fluid total protein (p c 0.05) but not in serum and ascitic fluid C4. There were no changes in serum C3, ascitic fluid C3, ascitic fluid C4 or in ascitic fluid total protein in group 2. These data demonstrate that selective intestinal decontamination increases serum and ascitic fluid C3 levels and, therefore, might be useful in preventing spontaneous infections in cirrhotic patients at high risk of infection. (HEPATOLOGY 1990;12:1175-1178.)

Spontaneous bacterial peritonitis (SBP) is a severe and frequent complication of cirrhosis with a high mortality rate (1-4). SBP is probably related to several impaired defense mechanisms, such as depressed reticuloendothelial system phagocytic activity (5, 61, leukocyte dysfunction (7,8),reduced serum complement(9, 10) and low bactericidal activity of ascitic fluid (AF) ( 11-13). It is generally assumed that SBP is a consequence of prolonged bacteremia later infecting the A F (14).Infection of this fluid is conditioned by its antimicrobial activity. In cirrhotic patients the bactericidal and opsonic activity of the AF is lower than that observed in noncirrhotic ascites (11, 12) or in normal peritoneal exudate (11, 15). It is directly related to the AF C3 and total protein (TP) concentration (13, 16). Recent investigations have shown that cirrhotic patients with deficient AF opsonic activity (17),low AF C3 (17, 18) and low AF TP (19) (particularly below 15 gm/L) (18), are more predisposed to SBP.

Received May 1, 1989; accepted June 5, 1990. Address reprint requests to: Dr. Carlos Guarner, Escuela de Patologia Digestiva, Hospital de la Santa Creu i Sant Pau, Avgda. Sant Antoni Maria Claret, 167, 08025 Barcelona. Spain. 31/1/24152

The passage of enteric bacteria, the most common causative organisms of bacteremia and SBP (14, 201, into the blood and/or AF may activate the alternative complement pathway. This may result in consumption of components of the alternative complement pathway (16). In fact, a decrease in the AF C3 levels during episodes of SBP (21) and in the serum C3 during bacteremia (22) has been reported. Oral norfloxacin in cirrhotic patients produced selective intestinal decontamination (SID) because it inhibits only the aerobic intestinal flora (23). The aim of this study was to assess the effect of SID with norfloxacin in cirrhotic patients with ascites and low AF T P on the serum C3, AF C3 and AF TP levels. PATIENTS AND METHODS Patients. Twenty-eight consecutively admitted cirrhotic patients with ascites and TP in AF lower than 15 grdL were included in the study. Patients with portal thrombosis, malignancy, gastrointestinal hemorrhage or infection on admission or during the study were excluded. Diagnosis of cirrhosis was established by histological study or clinical history and laboratory findings. Patients were randomly allocated to two groups: group 1 (n = 14)included those patients undergoing SID with norfloxacin (400mg twice daily for 7 days) and group 2 (n = 14)was composed of untreated controls. All patients were restricted in their salt intake (40 mmol/day). Diuretics or antibiotics were not given to any patient during the week before the trial or during the study. Albumin infusion or large-volume paracentesis was not performed in any patient during this period. No clinical or laboratory evidence was suggestive of SBP ( > 500 neutrophils/mm3 and/or positive culture) or other infection in any patient at the beginning or during the study. All patients gave written consent before the study. This study was approved by the ethical committee of our hospital. Methods. Diagnostic paracentesis was performed during the first 24 hr of hospitalization and 7 days later in each patient. The AF volume obtained (30 to 40 ml') was separated into aliquots, one of which was immediately stored at - 70" C after cells and particulate matter were removed by centrifugationat 2,500rpm for 15 min. In other A F samples, measurements of albumin, TP and leukocyte count and cytological study for malignant cells were done. We also performed Gram stain and culture by conventional methods and by inoculation of AF in aerobic and anaerobic blood culture bottles ( S & m S, et al. HEPATOLOGY 1989;9:662,Correspondence).



S U C H E'I' .41, 150



, / / '

? -












a W









FIG.1. Serum C3 levels in group 1 at the beginning and end of the study. A significant increase was observed Ip < 0.051. TABLE 1. Clinical characteristics of the two groups of patients on admission Characteristiax

Age Sex

Group 1 591



77 77


Pugh class (B/C) Origm (alcoholicMBViun 814 'L known) Prothrombin 578 * 4 5 time (5%) Bilirubin (+moVL) 41 0 = 26 5 Peripheral WBC 6,889 * 3.694 count (per mm')


- 85


7 512"



575 145 46 3 5 2 1 4 5.597 c 3,886



"One case had its origin in PBC. No statistical differences were appreciated between the two groups

A serum sample was obtained simultaneously and stored at -70" C. Serum analysis was carried out in all patients by standard methods. C3 and C4 determinations in serum and AF frozen samples were performed by nephelometry (24 1 using the Nephelometer ICS-I1 (Beckman Instruments Corp., Palo Alto, CAI with anti-C3 and anti-C4 antisera (Beckman Instruments Corp. ) . Body weights were recorded at the beginning and end of the study. Statisticalanalysis was performed with Student's t test and Fisher's exact test for comparison of both groups and with Student's t test for paired data. A p value ~ 0 . 0 5was considered statistically significant. All values were expressed as mean S.D.



FIG.2. AF C3 levels in group 1 at the beginning and end of the study. .4 significant increase was observed Cp = 0.01 I .


95 68



RESULTS Cirrhosis was of alcoholic origin in 15 subjects (eight treated patients and seven controls) and was hepatitis B virus-related in nine (four treated patients and five controls). One patient had PBC and three patients had cirrhosis of unknown origin (one treated patient and two controls). Liver biopsy was performed in three patients from the group receiving norfloxacin and in four from the control group. No differences were observed between the two groups in age, sex, degree of hepatic insufficiency on the Pugh scale, origin of cirrhosis, prothrombin time, bilirubin or peripheral WBC count (Table 1 ) . Table 2 shows body weights and hematocrits and levels of C3, C4, T P and albumin in serum and AF in both groups of patients at the beginning and end of the study. There were no statistical differences in these parameters between groups at the beginning of the study. No significant changes were observed in body weight, hematocrit, serum C4 and albumin or in AF C4 and albumin in any group. A significant increase in serum C3 (p < 0.05), AF C3 (p = 0.01) (Figs. 1 and 2) and AF TP (p < 0.05) levels was observed in group 1. No significant changes in serum C3, AF C3 or AF T P were observed in group 2. In Table 3 the variation in ascitesherum ratios (NS) of C3, C4, T P and albumin in the two groups at the

Vol. 12, No. 5,1990



TABLE 2. Body weight, hematocrit and C3, C4, total protein and albumin in serum and AF Group 2

Group 1



65.4 5 14.3 32.8 4.2 66.2 2 30.5 13.4 2 4.6 64.0 t 8.9 28.3 t 3.3 8.1 t 3.6 1.4 2 0.8 6.6 t 3.2 4.3 rt 2.4

65.0 t 14.5 32.8 2 5.5 71.5 ? 34.3" 12.9 t 5.2 66.3 t 7.9 28.9 t 5.6 10.2 rt 5.7b 1.6 2 1.0 8.7 rt 4.5" 4.6 2 2.1


Body weight (kg) Hematocrit (46) Serum C3 (mg/dl) Serum C4 (mg/dl) Serum TP (gm/L) Serum albumin (gm/L) AF C3 (mg/dl) AF C4 (mg/dl)

AF TP (gm/L) AF Albumin (gm/L)


68.5 ? 35.6 2 69.2 2 13.1 t 67.8 t 29.4 2 8.7 2 1.5 2 7.5 2 4.6 ?


10.4 4.5 18.1 5.0 8.2 3.6 2.5 0.7 3.6 2.1

67.9 t 10.3 35.7 rt 2.8 67.4 rt 17.5 12.8 t 3.1 68.4 rt 8.5 30.1 rt 4.6 9.3 rt 3.3 1.6 ? 0.8 8.1 2 3.5 5.0 t 2.6

For paired data: "p < 0.05 and b p = 0.01 respective to the initial value.

TABLE 3. Variation in A/S of C3, C4, TP and albumin during the study Group 1

Als c3 c4 TP Albumin


0.128 0.103 0.104 0.150

2 rt 2 2

0.041 0.009 0.046 0.073

Group 2




0.153 t 0.057' 0.123 2 0.018 0.129 5 0.064 0.161 2 0.077

0.130 t 0.040 0.117 t 0.062 0.110 t 0.045 0.157 2 0.066

0.140 5 0.041 0.123 t 0.052 0.117 2 0.045 0.165 2 0.078

"p c 0.05 with respect to the initial value.

beginning and the end of the study are shown. There were no statistical differences in initial values between the two groups. There was a significant increase in A/S C3 in group 1 (p < 0.05),whereas group 2 did not show significant changes. We did not find statistical differences in the NS C4, NS TP or A/S albumin between the beginning and the end of the study in either group. DISCUSSION

This study demonstrates that oral norfloxacin increases C3 in serum and AF and also the A F TP levels in cirrhotic patients with low A F TP. The increase in serum C3 and AF C3 levels may be caused by several mechanisms, such as inhibition of serum and AF complement-factor consumption and/or improved hepatic complement synthesis (9). The activation of the alternative complement pathway by bacteria or their products in blood or AF accounts for this consumption (16). It is generally assumed that the source of the bacteria or their products is the intestine because most bacteremia and SBP in cirrhotic patients are caused by enteric bacteria (14, 20). SID with norfloxacin in volunteers (25) and granulocytopenic (26) and cirrhotic patients (23) resulted in decreases in the aerobic intestinal flora. A lower incidence of infections caused by enterobacteria has been reported as well in granulocytopenic patients receiving norfloxacin (26,27) and in cirrhotic patients with gastrointestinal hemorrhage treated with oral nonabsorbable antibiotics (28). Therefore SID should decrease the passage of enterobacteria to the blood and AF and, consequently, the consumption of complement factors. The increase of serum and AF C3 levels observed in our study in patients undergoing SID seems to support this hypothesis.

Changes in hepatic synthesis capacity may be another mechanism by which serum and AF C3 concentration could change (29). However, no significant variations in serum and AF C4 and albumin levels were observed in any of the groups, although C4 and albumin are also synthesized in the liver (30).No differences were found in liver test data between the two groups. Therefore, although we have not performed liver function tests, it is improbable that the serum and AF C3 increase in the group undergoing SID could be attributed to improvement in hepatic synthesis. Because a direct correlation between serum and AF C3 has been reported (18),the increase of AF C3 in patients undergoing SID could also be due to that of serum C3 levels. However, the increase in the A/S C3 in the treated group seems to support the idea that factors other than the increase in serum C3 should account for the increase in AF' C3 in this group. The increase of C3 and TP in AF in treated patients could also be a consequence of concentration of AF by spontaneous or induced diuresis. It has been previously reported that AF TP, C3 and opsonic activity increase during diuretic treatment in cirrhotic patients (30, 31). In this study no patient received diuretic treatment, but spontaneous diuresis may occur in cirrhotic patients subjected to bed rest and reduced sodium intake (32). Nevertheless, during the study body weight decreased similarly, but not significantly, in both groups. Furthermore, no significant differences in AF C4 levels were found in either group, whereas this factor increases after drug-induced diuresis (31). Hence we can assume that a similar spontaneous diuresis may have occurred in both groups. According to our results, changes in AF concentration alone did not account for


SUCH El’ .%I,

the increase in AF C3 and AF TP seen in patients undergoing SID. The increase in AF T P observed in patients treated with norfloxacin could be due to several mechanisms. AF TP concentration is determined by serum T P concentration and portal pressure (33).Changes in these two parameters could induce a redistribution of proteins with subsequent variation in their levels. In our study we did not find statistical variations in serum TP. albumin and C4 between the begmning and the end of the study in any of the two groups. Moreover, the increase in the NS TP, AJS albumin and AJS C4 did not reach statistical significance in the treated group or in the control group. Consequently, we think redistribution of proteins alone does not explain the increase in AF TP and AF C3 in the treated group. However, we cannot exclude changes in portal pressure resulting from SID as a factor that could cause AF TP to vary because this has not been measured. As we have previously commented, norfloxacin produces SID and, consequently, enteric bacteria do not reach AF. Therefore not only C3 but also other proteins that play some role in the bactericidal and opsonic activity of AF - such as fibronectin, immunoglobulins and other complement factors of the alternative pathway-would increase because of a decline in their consumption. This could explain, at least in part, the increase in AF T P observed in the treated group. We conclude that treatment with norfloxacin in cirrhotic patients with low AF TP increases serum C3 and AF C3 and T P concentration and, therefore, AF bactericidal activity. Because the incidence of SBP depends on the AF TP, C3 and opsonic activity (17-19), our results suggest that prophylactic therapy with norfloxacin in cirrhotic patients with ascites at high risk of SBP may decrease the incidence of this complication.

10 11

12 13


15 16


18 19 20

21 22 23


Acknowledgments: We are grateful to our colleagues and the nursing staff of our department for their collaboration.




1 . Curry N, McCallum RW. Guth PH. Spontaneous bacterial peritonitis in cirrhotic ascites: a decade of experience. Am J Dig Dis

1974;19:685-692. 2. Correia JP, Conn HO. Spontaneous bacterial peritonitis in hepatic cirrhosis: endemic or epidemic? Med Clin North Am 1975;59:963981. 3. Hoefs JC, Canawati HN, Sapico FL. Hopluns R. Weiner J. Mont1982; gomerie JZ. Spontaneous bacterial peritonitis. HF.PATOI.OW 2:399-407. 4. Pinzello G, Simonetti RG, Craxi A, Di Piazza S, Spano C, Pagliaro L. Spontaneous bacterial peritonitis: a prospective investigation in predominantly nonalcoholic cirrhotic patients. HEPATOLOGY 1983; 3:545-549. 5. Lahnborg G, Friman L, Berghem L. Keticuloendothelial function in patients with alcoholic liver cirrhosis. Scand J Gastroenterol 1981;16:481-489. 6 . Rimola A, Soto R, Bory F, Arroyo V, Piera C, Rodes J. Reticulaendothelid system phagocytic activity in cirrhosis and its relation 1984;4:53-58. to bacterial infections and prognosis. HEPATOI.OW 7. Feliu E, Gougerot MA, Hakin J, Cramer E. Auclair C, Ruett B, Boivin P. Blood polymorphonuclear dysfunction in patients with alcoholic cirrhosis. E u r J Clin Invest 1977;7:571-577

LA,Williams R. Abnormalities of neutrophil phagocytosis, intracellular lulling and metabolic activity in alcoholic cirrhosis and hepatitis. HEPATOI.(K;Y 1986;6:252-262. Fox RA, Dudley FJ, Sherlock S. The serum concentration of the third component of complement in liver disease. Gut 1971;12:574578. Potter BJ.Trueman AM, Jones EA. Serum complement in chronic liver disease. Gut 1973;14:451-456. 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. Akalin HE, Laleli Y,Telata H. Bactericidal and opsonic activity of ascitic fluids from cirrhotic and non-cirrhotic patients. J Infect Dis 1983;147:101 1- 10 17. Runyon BA, Morrissey RL, Hoefs JC, Wyle F. Opsonic activity of human ascitic fluid: a potentially important protective mechanism against spontaneous bacterial peritonitis. HEPATOLOGY 1985;5:634637. Hoefs JC. Runyon BA. Spontaneous bacterial peritonitis. Dis Mon 1985;31:1-48. Bercovici B, Michel J. Miller J, Sacks G. Antimicrobial activity of human peritoneal fluid. Surg Gynecol Obstet 1975;141:885-887. Fromkes JJ. Thomas FB, Mekhjian HS, Evans M. Antimicrobial activity of human ascitic fluid. Gastroenterology 1977;73:668-672. Runyon BA. Patients with deficient ascitic fluid opsonic activity are predisposed to spontaneous bacterial peritonitis. HEPATOLOGY 1988;8:632-635. Such J, Guarner C, Enriquez J, Rodriguez JL, Seres I, Vilardell F. Low C3 in cirrhotic ascites predisposes to spontaneous bacterial peritonitis. J Hepatol 1988;6:80-84. Runyon BA. Low-protein-concentration ascitic fluid is pred~sposed to spontaneous bacterial peritonitis. Gastroenterology 1986;91: 1343- 1346. Graudal N, Milman N, Kirkegaard E, Korner B, Thompsen AC. Bacteremia in cirrhosis of the liver. Liver 1986;6:297-301. 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. Fearon DT. Ruddy S, Schur PH, McCabe WR. Activation of the properdin pathway of complement in patients with Gram negative bacteremia. N Engl J Med 1975;292:937-940. Gines P, Rimola A, Marco F, Almela M, Marques X, Rodamilans M, Jimenez de Ants MT, et al. Oral norfloxacin produces a selective bowel decontamination in cirrhotic patients [Abstract I. J Hepatol 1988;7:S136. Buffone GJ. Immunonephelometric and turbidimetric measurement of specific plasma proteins. In: Rose NR, Frieldman H. eds. Manual of clinical immunology. 2nd ed. Washington, D.C.: American Society for Microbiology, 1980, 23-28. Nord CE, Edlund C, Bergan T, Josefsson K, Solberg R. Impact of norfloxacin on human oropharyngeal and colonic microflora [Abstract]. Rev Infect Dis 1988;10:S96. Karp J. Merz WG, Hendricksen C, Laughon B, Redden T, Bamberger BJ, Bartlett JG, et al. Oral norfloxacin for prevention of Gram-negative bacterial infections in patients with acute leukemia and granulocytopenia. Ann Intern Med 1987;106:1-7. Winston DJ, Winston GH, Champlin RE, Karp J, Bartlett J , Finley RS, Joshi J H , et al. Norfloxacin for prevention of bacterial infections in granulocytopenic patients. Am J Med 1987;82:40-46. Rimola A, Bory F, Teres J, Perez-Ayuso RM, Arroyo V, Rod& J. Oral, nonabsorbable antibiotics prevent infection in cirrhotics with gastrointestinal hemorrhage. HEPATOLOGY 1985;5:463-467. Colter HR, Alper CA. Rosen FS. Current concepts in immunology: genetics and biosynthesis of complement proteins. N Engl J Med 198 1;304:653-656. Hoefs JC. Increase in ascites white blood cell and protein concentrations during diuresis in patients with chronic liver disease. HEPATOLOGY 1981;1:249-254. Runyon BA, Van Epps DE. Diuresis of cirrhotic ascites increases i t s opsonic activity and may help prevent spontaneous bacterial 1986;6:396-399. peritonitis. HEPATOLOGY Arroyo V, RcdCs J . A rational approach to the treatment of ascites. Postgrad Med J 1975;51:558-562. 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.

8 Kajkovic


27 28

29 30 31 32 33


Selective intestinal decontamination increases serum and ascitic fluid C3 levels in cirrhosis.

Selective intestinal decontamination for 7 days with norfloxacin was performed in 14 cirrhotic patients with ascites and low ascitic fluid total prote...
433KB Sizes 0 Downloads 0 Views