Peritonite bacteriana espontânea: uma complicaçäo séria nos cirróticos / Spontaneous bacterial peritonitis: a very serious complication in cirrhotics

É apresentado um caso de Peritonite Bacteriana Espontânea - uma complicaçäo grave que pode desenvolver-se em cirróticos com ascite. O dignóstico precoce e um tratamento imediato diminuíram a mortalidade de episódios da doença. Infelizmente, existe uma alta taxa de recorrência da mesma e o seu aparecimento num cirrótico indica um péssimo prognóstico

Avaliaçäo da textura ecográfica do líquido ascítico e de sua distribuiçäo na cavidade peritoneal / Evaluation of ascitic fluid echographic aspect and its distribution in the peritoneal

Com a finalidade de avaliar a textura ecográfica do líquido ascítico e a sua distribuiçäo na cavidade peritoneal, foram estudados, prospectivamente, 153 pacientes com ascite de diversas etiologias. Em relaçäo à distribuiçäo do líquido na cavidade peritoneal verificou-se que esta näo mantém relaçäo com a etiologia da ascite. O recesso subfrênico direito é o que se encontra preenchido por líquido ascítico no maior número de pacientes, sendo o recesso mais precocemente ocupado nas ascites mínimas; o recesso esplenorrenal acha-se preenchido por líquido ascítico no menor número de pacientes, sendo o espaço mais tardiamente ocupado nas ascites volumosas. Observa-se maior frequência de pacientes com ascite volumosa nos grupos diagnósticos hepatopatia e cardiopatia. Näo se verifica relaçäo entre as características físicas do líquido ascítico e o seu aspecto ecográfico no diagnóstico etiológico da ascite

Ascitic fluid alpha 1-antitrypsin.

Ascitic fluid alpha 1-antitrypsin (AF-AAT) was compared with ascitic fluid total protein (AF-TP) and the serum-ascites albumin gradient (SAAG) in the differential diagnosis of ascites. The study included 82 consecutive patients of which 42 had cirrhosis, 8 hepatoma (with cirrhosis), and 27 malignant ascites (peritoneal 18, liver 9). The concentration of AF-AAT (milligrams per deciliter) was significantly elevated (P less than 0.001) in hepatoma (174 +/- 123), malignant liver disease (232 +/- 119) and peritoneal neoplasms (376 +/- 106) in comparison with cirrhotics (66 +/- 33). In separating ascites caused by cirrhosis or malignancy, AF-AAT (discriminating limit of 120 mg/dl) had a 96% sensitivity, 95% specificity, and 96% diagnostic efficacy, which was superior to the 87% observed for AF-TP and 86% for the SAAG. Similar results were obtained for the A/S AAT ratio but this test was not available in all patients. AF-AAT was particularly useful in patients with malignancy causing portal hypertension as assessed by SAAG (hepatoma, malignant liver disease). We conclude that AF-AAT may be a valuable parameter in the differential diagnosis of ascites.

Elevated peritoneal fluid luteinizing hormone and prolactin concentrations in infertile women with endometriosis.

In this study, we compared (Mann-Whitney U-test) the peritoneal fluid FSH, LH and PRL levels, measured by RIA, at the follicular and luteal phases of the menstrual cycle in women with (n = 43; age 25-44 years) and with no evidence of endometriosis (n = 35; age 25-39 years) who were considered as controls. Both follicular and luteal phase FSH concentrations of women with endometriosis were not statistically different (n = 22 vs 18; 0.32-5.8 vs 0.50-8.2 IU/l, P = 0.247; n = 13 vs 14; 0.6-6.5 vs 0.66-6.7 IU/l, P = 0.604) compared to their respective controls. In contrast to FSH, the concentrations of LH at follicular (n = 19 vs 17; 3.1-34.2 vs 2.3-12.2 IU/l, P = 0.01) and luteal (n = 17 vs 15; 2.1-95.4 vs 1.3-17.9 IU/l, P = 0.02) phases of the test group was significantly elevated at both phases of the cycle. With respect to differences in PRL concentrations at follicular phase no significant change (n = 21 vs 16; 1030-5800 vs 1305-4650 mIU/l; P = 0.255) was observed. The greatest difference in luteal PRL concentrations (P = 0.007) was obtained between the women with endometriosis and controls (n = 17 vs 17; 1895-8600 vs 1041-5000 mIU/l). The results suggest that disordered synchronization of neuroendocrine mechanisms controlling LH and PRL may be the underlying abnormality causing infertility in our group of patients with endometriosis.

Ascitic fluid concentrations of fibronectin and cholesterol: comparison of differential diagnostic value with the conventional protein determination.

Ascitic fluid concentrations of fibronectin, cholesterol and protein were determined in 95 patients: 38 with cirrhosis of the liver, 10 with miscellaneous nonmalignant diseases, 43 with peritoneal carcinomatosis and 4 with liver metastases or hepatocellular carcinoma. Fibronectin, cholesterol and protein at discrimination values of 7.5 mg/100 ml, 45 mg/100 ml and 3.0 g/100 ml, respectively, separated patients with peritoneal carcinomatosis from patients with cirrhosis with an efficiency of 94%, 90% and 85%, respectively. Thus, ascitic fluid determinations of fibronectin and cholesterol offer good discrimination of cirrhotic ascites from ascites related to peritoneal carcinomatosis, superior to the conventional protein determination. However, the failure of all parameters to distinguish ascites caused by miscellaneous nonmalignant diseases from malignancy-related ascites underscores the importance of highly specific methods to confirm a suspected diagnosis of malignancy-related ascites.

Use of two-dimensional gas chromatography with electron-capture detection for the measurement of lipopolysaccharides in peritoneal fluid and plasma from rats with induced peritonitis.

The content of 3-hydroxymyristic acid from Escherichia coli lipopolysaccharide in peritoneal fluid and plasma from rats was determined by two-dimensional gas chromatography with electron-capture detection of the 3-O-pentafluorobenzoyl methyl ester derivative. The detection limit of lipopolysaccharide in peritoneal fluid was 3 ng/ml. An experimental model of E. coli peritonitis in the rat was used, with and without coinjection of bile. The concentrations of lipopolysaccharide were highest in both peritoneal fluid and plasma samples from rats injected with E. coli and bile, reaching a maximum 1 h after injection by the gas chromatographic method. Corresponding Limulus assay results for peritoneal samples showed a small increase of lipopolysaccharide concentrations during the first 4 h after injection, followed by a substantial increase. The results indicate that bile salts cause an increased release of lipopolysaccharide from gram-negative bacterial cells in vivo and that this may be responsible for the high mortality caused by peritonitis. In contrast to the Limulus assay, gas chromatography enables the total amount of lipopolysaccharide in a clinical sample to be determined.

Peritoneal sperm recovery can be consistently demonstrated in women with unexplained infertility.

Diligent analysis of PF 10 to 20 hours after midcycle intracervical insemination with husband's semen in couples with unexplained infertility showed that sperm are consistently able to transverse the reproductive tract in this group of patients. However, this finding does not necessarily imply that the sperm were retained at the site of fertilization or that they were competent to achieve oocyte fertilization. Therefore, further experiments obtaining sperm from the tubal isthmus to assess the effects of their sequestration there on their ability to fertilize human oocytes are needed.

Proposed mechanism of the inflammatory attacks in familial Mediterranean fever.

Peritoneal and synovial fluids of patients with familial Mediterranean fever lack a protein that inhibits neutrophil chemotaxis by antagonizing the complement-derived inflammatory mediator C5a. The C5a inhibitor activity was studied with the use of a C5a binding assay where peritoneal fluids were tested for their ability to inhibit recombinant C5a binding to dibutyryl cyclic adenosine monophosphate-induced U937 cells. In contrast to normal peritoneal fluids, those from patients with familial Mediterranean fever contained less than 1% C5a inhibitor activity. Gel filtration and ion exchange chromatography of peritoneal fluids from those patients did not yield any fraction that inhibited C5a binding. We suggest that the serosal tissue of patients with familial Mediterranean fever is devoid of C5a inhibitor activity and that this deficiency may explain in part the local inflammatory episodes characteristic of this disease.

Characterization of a panel of monoclonal antibodies to human coagulation factor XI and detection of factor XI in Hep G2 cell conditioned medium.

We have produced a panel of ten monoclonal antibodies specific to coagulation factor XI. Western blot analysis demonstrates that 9 of these antibodies react with the heavy chain of factor XI and one with the light chain. Seven of these antibodies inhibit factor XI and factor XIa activity. We have used immobilised monoclonal antibody for the production of factor XI deficient plasma and to purify factor XI to homogeneity with high yield in a simple two-step procedure. These monoclonal antibodies were used to develop highly sensitive immunoassays capable of detecting less than 0.01 mu factor XI antigen ml-1. A strong correlation was found between antigen and activity levels in 11 patients with hereditary deficiency indicating that none was cross-reacting material positive. Cultured Hep G2 cells were found to synthesize small amounts of factor XI antigen and this could also be detected by functional assay and by western blot analysis.

Relation of intraperitoneal and intravascular coagulation and fibrinolysis related antigens in peritoneal dialysis.

Patients received 2,000 ml of dialysate intraperitoneally with five exchanges per day during continuous peritoneal dialysis (CAPD) for the treatment of terminal renal insufficiency. During a dwell time of 4 h the dialysate reached a total protein concentration up to 100 mg/dl by mass transfer of intravascular proteins. The composition is dependent on the molecular weight of the proteins. This results in an intraperitoneal hemostatic system of low concentration and different composition. We found an intraperitoneal fibrinogen cleavage and thrombin-antithrombin III-complex formation leading to increased levels of fibrinopeptide A (FPA: 33.3 +/- 7.0 ng/ml) and thrombin-antithrombin III-complex (TAT: 4.7 +/- 0.4 ng/ml) in plasma by mass transfer from dialysate to plasma. t-PA (tissue plasminogen activator) and PAI-1 (plasminogen activator inhibitor type 1) concentrations in plasma were within the normal range. The dialysate concentrations indicated a low local secretion. The fibrinolytic fibrin fragment D-dimer and the fibrinogen degradation product concentrations in plasma were greater than in dialysate. But the relations of the proteins between plasma and dialysate refer to a local intraperitoneal production as well. The results show that intraperitoneal coagulation predominates over fibrinolysis which is accompanied by an intravascular fibrinolysis in patients undergoing CAPD. Neoantigens produced in dialysate and diffused to plasma are comparable to changes seen in disseminated intravascular coagulation.

Comparative analysis of protein content in rat mesenteric tissue, peritoneal fluid, and plasma.

Albumin, transferrin, and total protein concentrations were measured in the mesenteric tissue, peritoneal fluid, and plasma of 12 ketamine-Nembutal-anesthetized Sprague-Dawley rats. Tissue samples were obtained with an 8-mm trephine; tissue water content was determined by a microgravimetric method to be 5.2 +/- 0.3 microgram water/microgram dry wt. Peritoneal fluid was collected by capillary action in hematocrit tubes, and blood samples were taken from a femoral artery catheter. Total protein concentrations of plasma (5.8 +/- 0.3 g/dl) and peritoneal fluid (2.6 +/- 0.1 g/dl) were determined by Lowry assay. Ratios of peritoneal fluid and tissue densitogram areas to plasma area were used to calculate total protein content of peritoneal fluid (2.5 +/- 0.1 g/dl) and tissue (1.8 +/- 0.2 g/dl). Albumin concentrations were 1.1 +/- 0.1 g/dl for tissue, 1.4 +/- 0.1 g/dl for peritoneal fluid, and 2.8 +/- 0.1 g/dl for plasma. Transferrin concentrations were 0.09 +/- 0.01 g/dl for tissue, 0.13 +/- 0.01 g/dl for peritoneal fluid, and 0.28 +/- 0.01 g/dl for plasma. Peritoneal fluid protein concentrations were similar to values found for lymph in previous studies. Protein concentration in the tissue buttons was significantly less than that of peritoneal fluid. This contradicts the widely held assumption that the protein concentration of fluid outside the matrix is representative of a well-mixed interstitial matrix fluid protein concentration.

The impact of peritoneal trauma on intra-abdominal fibrinolytic activity, adhesion formation and early embryonic development in a rabbit longitudinal model.

Intra-abdominal adhesions may interfere with fertility following gynaecological surgery and injury to the peritoneum plays a central role in the pathogenesis. Tissue plasminogen activator and its antagonists play a pivotal role in the intra-abdominal balance between fibrinolysis and adhesion formation. This process may be cycle-dependent in women. In order to establish the impact of the fibrinolytic activity on adhesion formation after a standardized trauma, a rabbit longitudinal model was developed, which allowed the study of possible differences between the periods before and after ovulation. The influence of extra-genital adhesions on early embryonic development was investigated. No cycle-dependent changes in fibrinolytic activity of the peritoneal fluid (PF) or of the serum could be demonstrated. No correlation was found between post-operative adhesion formation and the fibrinolytic activity during surgery. Three weeks after surgery, a significant increase in fibrinolytic activity of the PF was observed. The rank order of sampling is suggested to account for these differences. Extra-genital adhesions did not markedly influence ovulation, ovum pick-up and fertilization in this hormonally controlled rabbit model.

Partial characterization of a C5a-inhibitor in peritoneal fluid.

We have recently described a 40-kDa protein in peritoneal fluid that neutralized the chemotactic activity of the C fraction C5a. It was deficient in peritoneal fluids of patients suffering from familial Mediterranean fever. Further characterization of the inhibitor with the use of 125I-rC5a binding to dibutyryl cAMP-induced U937 cells revealed dependence on the peritoneal fluid concentration, on the time of incubation and on temperature and pH. Fractionation of 125I-C5a on Sephadex G-50 column, before and after incubation with peritoneal fluid, revealed similar fractionation patterns despite loss of biologic activity of the treated C5a (but not its binding to polyclonal anti-C5a antibody). Analysis of rC5a by SDS-PAGE before and after treatment with partially purified C5a inhibitor, revealed slight modification of the inhibitor-treated C5a. Using various protease inhibitors (i.e., PMSF) suggested that the C5a inhibitor is a serine protease. It neutralized C5a by means of limited proteolysis which did not change C5a immunologic properties and changed only slightly its m.w. but abolished its receptor binding and chemotactic functions. It is suggested that the C5a inhibitor plays a role in the regulation of inflammation in serosal tissues and that its deficiency in familial Mediterranean fever may explain the attacks of sterile inflammation characteristic of this disease.

Role of plasmin in the degradation of the stroma-derived fibrin in human ovarian carcinoma.

The aim of this study was to evaluate the type of enzymes involved in tumor-associated fibrinolysis of the stroma component fibrin in ovarian cancer patients. For this purpose, the high-molecular-mass fibrin degradation products (HMM-XDP) were isolated from malignant ascitic fluid by protamine sulfate precipitation and further purified by gel filtration and acid precipitation. After reduction with 2-mercaptoethanol, the peptide chain components were separated by reverse-phase high-performance liquid chromatography (RP-HPLC). The nature of these components was elucidated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and N-terminal amino acid sequence analysis and compared with fibrin-derived fragments formed in vitro. The results indicate that plasmin is the essential protease involved in the degradation of the stroma-derived fibrin portion found in ovarian cancer ascites.

Bacterial peritonitis following esophageal injection sclerotherapy for variceal hemorrhage.

Patients presenting with ascites and upper gastrointestinal hemorrhage were studied prospectively. Five patients presenting with acute variceal hemorrhage were found not to have pre-existing spontaneous bacterial peritonitis on initial paracentesis. However, three of these five developed findings compatible with bacterial peritonitis after sclerotherapy. Although the number of cases is small, our results support the monitoring of ascitic fluid after sclerotherapy.