[Spontaneous bacterial peritonitis].

AIM OF STUDY: Spontaneous bacterial peritonitis (SBP) is the most frequent infectious complication of liver cirrhosis with serious consequences. Initially, SBP is always treated with empirical, not targeted, antibiotic therapy. Since a retrospective study performed in our department showed suboptimal effectiveness (only 40 %) of empirical antibiotic therapy in accordance with the EASL guidelines, a decision was made to change the protocol. The aims of this prospective study were to determine: (1) the incidence and characteristics of SBP in real clinical practice - in a liver unit of a tertiary hospital, (2) the effectiveness of new antibiotic therapy selected based on analysis of the spectrum of pathogens and their resistance to antibiotics as identified in a retrospective cohort study on SBP carried out in our department, (3) mortality, and to compare these findings with the literature data. MATERIAL AND METHODS: A prospective cohort observational pragmatic study. SETTING: Department of Hepatology, Gastroenterology and Liver Transplantation, 2nd Internal Clinic, Slovak Medical University and F. D. Roosevelt Teaching Hospital with Polyclinic in Banska Bystrica. Time interval: November 2012-August 2013. INCLUSION CRITERIA: hospitalization for liver cirrhosis, ascites ≥ grade 2, informed consent. The study was approved by the local ethics committee. EXCLUSION CRITERIA: malignancy, secondary bacterial peritonitis. DIAGNOSIS: SBP was defined by the count of neutrophil leukocytes in ascites ≥ 250/mm3). Positive ascitic fluid culture was not a necessary condition for the diagnosis. From each patient, 10 mL of ascitic fluid were sampled into two blood culture bottles, anaerobic and aerobic. Therapeutic response: defined as a decrease in NeA to 25 % of the baseline value after 48-72 hours, in accordance with the EASL guidelines. The absence of response was indication for change of the antibiotic therapy strategy. Empirical antibiotic therapy: The drug of choice was piperacillin/tazobactam 4 g/0.5 g i.v. every 8 hours for 5 days. Additionally, 20% human albumin at doses of 1.5 g/kg of patient weight on day 1 and 1.0 g/kg of patient weight on day 3 from the diagnosis was administered. If there was no response, (a) second choice antibiotic therapy according to analysis of the spectrum of pathogens and their resistance as identified in the former retrospective study on SBP, that is, ertapenem 1g i.v. every 24 hours for 5 days, or (b) targeted antibiotic therapy according to analysis of ascitic fluid culture performed in the meantime was initiated. RESULTS: The inclusion criteria were met by 65 patients (99 episodes); the incidence of SBP was 9 out of 99 episodes (9.1 %); 5 out of the 9 cases had positive bacterial culture (56 %), with most of bacteria being Gram-positive (4 out of 5 cases, 89 %). Therapeutic response was documented in 7 out of the 9 cases (78 %). The in-hospital mortality of patients with SBP was 11 %. CONCLUSIONS: SBP was detected in one out of ten patients with cirrhotic ascites. The selection of empirical therapy in accordance with the principles of antibiotic stewardship led to an increase in therapeutic response to more than 75 %. Effective treatment of SBP is a prerequisite for reduction of mortality.

Aetiology, antifungal susceptibility, risk factors and outcome in 201 fungaemic children: data from a 12-year prospective national study from Slovakia.

A total of 201 cases of fungaemia in children in a 12-year national survey from seven University Paediatric Clinics in Slovakia in 1990-2001 was assessed to determine risk factors, therapy and outcome, and to compare those cases with fungaemia in 130 adult cancer patients studied in a similar survey. Four univariate analyses were performed to assess differences in aetiology, antifungal susceptibility and outcome between fungaemia in neonates and paediatric intensive care unit (ICU) patients as well as between paediatric and adult cancer patients with fungaemia. There was a significant difference in aetiology and antifungal susceptibility between the subgroups of children with fungaemia: 83.3% of neonates versus 40.2% in children with cancer were due to Candida albicans. None of the non-albicans Candida spp. (NAC) in neonates but 23.5% of NAC isolates from children with cancer were resistant to fluconazole. C. albicans caused 144 (71.1%) episodes and NAC 48 (23.7%) episodes. Trichosporon beigelii, Blastoschizomyces (Trichosporon) capitatus, Rhodotorula rubra and Cryptococcus laurentii were found less frequently in neonates than in children with cancer (18.8%). There were not many differences in risk factors between paediatric fungaemia and adult cancer fungaemia except C. albicans aetiology, corticosteroid use in therapy, breakthrough fungaemia after ketoconazole prophylaxis and meningitis as a complication, which were observed significantly more frequently among children than in adults, both with cancer and fungaemia. Thirty-three of the paediatric fungaemias were breakthrough cases and appeared frequently in children with cancer. Fifty-one (25.1%) children died with fungaemia (attributable mortality) and 25 (12.7%) due to underlying disease with fungaemia; overall mortality was 37.8% and there was no significant difference in death rates between the subgroups of paediatric patients (neonates, children in ICUs and children with cancer).

[Multiresistant nosocomial bacterial strains and their "in vitro" susceptibility to chloramphenicol and colistin].

OBJECTIVE: A multicenter study was conducted to obtain "in vitro" chloramphenicol and colistin susceptibility data on multiresistant hospital bacterial pathogens in Slovak Republic. MATERIAL AND METHODS: During the period of April-June 2001, 628 clinical bacterial multiresistant isolates from patients with serious infections were selected in 10 hospitals and tested to a large scale of antibiotics by means of a microdilution method. The strains expressed either a significant resistance phenotype (ESBL, MRSA, CoNMRS, MLSB/c, efflux in Ps. aeruginosa), or were resistant to one or more preparations in at least half of reliable unrelated antibiotic groups (beta-lactams, aminoglycosides, quinolons, macrolides). RESULTS: Both chloramphenicol and colistin retained significant "in vitro" activity against many multiresistant hospital bacterial pathogens. The highest activity of chloramphenicol was documented for isolates of Stenotrophomonas maltophilia (76,5 % susceptible, MIC50 = 4 mg/L, MIC90 = 16 mg/L) and of Staphylococcus aureus (76,2 % susceptible, MIC50 = 8 mg/L, MIC90 = 16 mg/L). In tested Pseudomonas aeruginosa (82,5 % susceptible, MIC50 = 2 mg/L, MIC90 = 16 mg/L) and Stenotrophomonas maltophilia (88,2 % susceptible, MIC50 = 1 mg/L, MIC90 = 8 mg/L) isolates colistin represented the most "in vitro" effective antibiotic. Colistin was the only "in vitro" effective antimicrobial in four of 120 multiresistant Pseudomonas aeruginosa isolates tested in our study. CONCLUSIONS: The study confirmed a good "in vitro" susceptibility of many multiresistant hospital bacterial pathogens to chloramphenicol and colistin in Slovak Republic. The clinical application of chloramphenicol and colistin might be reconsidered in infections caused by extremely resistant bacteria with prooved susceptibility to these antibiotics. It is important to consider, that the infection danger has to exceed the risk of antibiotic toxicity.