Biofilm-producing potential of urinary pathogens isolated from chronic and recurrent urinary tract infections and impact of biofilm on gentamicin and colistin in vitro efficacy.

AIM: The presented study was to compare in vitro biofilm production by bacterial strains from chronic/recurrent and from acute non-complicated UTIs. The activity of gentamicin and colistin on biofilm form of these strains has also been detected, with goal to predict the gentamicin and colistin therapeutic efficacy in the antimicrobial treatment of patients with a suspected presence of biofilm in urinary tract. MATERIAL AND METHODS: The group of 40 bacterial strains repeatedly isolated from patients with chronic or recurrent UTIs was compared with the group of 40 strains from acute UTIs. Both groups contained comparable number of strains of Escherichia coli, Klebsiella spp., Proteus mirabilis and Pseudomonas aeruginosa. Biofilm production was assessed by method in polystyrene microtiter plate. The MIC and MBC values of gentamicin and colistin were detected by broth microdilution assay. The minimal biofilm inhibitory (MBIC) and biofilm eradication concentrations (MBEC) were tested by microdilution method. Non-inactivated biofilm-associated bacteria were detected after overnight incubation in broth medium free of antimicrobials. The statistical analysis of results was performed by Fisher's exact test and by Student's t-test. RESULTS: Biofilm was produced by 90% strains from chronic UTIs, but only by 52% of strains from acute UTIs (p = 0.0004). In the biofilm producing strains, the MBIC values of gentamicin reached from four to 256 mg/L, the MBIC levels of colistin from two to 64 mg/L. The minimal biofilm eradicating concentrations were even higher: for gentamicin from eight to > 512 mg/L, and for colistin from 32 to > 512 mg/L. The differences between MIC and MBIC/MBEC levels were statistically highly significant (p < 0.0001). Presumably, the therapeutic success of parenterally applied gentamicin or colistin on biofilm-related urinary tract infections would be, without respect to the high concentration of gentamicin or colistin achievable in urine during parenteral application, rather unpredictable. Local intravesical instillation would allow for achieving higher gentamicin and colistin concentrations; however, there is need for interpretation criteria for MBEC values concerning therapy, as well as for clinical studies allowing for application of those values to predict clinical success of therapy. CONCLUSIONS: Laboratory detection of biofilm production and evaluation of the MBIC/MBEC values of antimicrobials for strains producing biofilm might be a valuable complement to the microbiologic diagnostics of chronic and recurrent UTIs. It might provide valuable information for more reliable individualised therapy and so decrease the risk of emergence and selection of multiresistant strains during repeated and non-eradicating therapy of chronic and recurrent UTIs.

Fungaemia due to Fusarium spp. in cancer patients.

Five cases of fungaemia due to Fusarium spp. in cancer patients are described. Two were breakthrough cases, despite ongoing therapy with amphotericin B. Three were caused by Fusarium solani, one by F. oxysporum and one by F. dimerum. Four patients died, three of them despite therapy with amphotericin B for between 5-37 days. We describe only the second reported case of F. dimerum fungaemia. Since 1972, 93 cases of systemic infection with Fusarium spp. have been described: 43 had positive blood cultures and the overall mortality was 72%.

Bacteremia due to Enterobacter spp. in cancer patients--analysis of 51 episodes.

Fifty one episodes of bacteremia due to Enterobacter spp. appearing within 7 years among 12 301 admissions in a single cancer institution were studied for risk factors, clinical presentation and outcome. Fifteen episodes were due to Enterobacter aerogenes, 23 due to E. cloacae and 13 due to E. agglomerans. The proportion of bacteremia due to Enterobacter spp. among Gram-negative bacteremias was 10.1% and infection associated mortality was 13.8%. The incidence in 1989-1995 varied from 3.7 to 8.7% and was relatively stable. Most common risk factors were: solid tumors as underlying disease, central venous catheter insertion, prior surgery and prior chemotherapy within 48 h. Neutropenia and urinary catheters were not at high risk in either one of the patients subgroups. Comparing two subgroups of 51 bacteremias, monomicrobial and polymicrobial (when Enterobacter spp. was isolated from blood culture with other microorganism), previous chemotherapy, vascular catheter insertion and prior endoscopy were more frequently associated with polymicrobial Enterobacter spp. bacteremia. There was also differences in infection associated mortality: bacteremias due to Enterobacter spp. only had significantly lower mortality in comparison to polymicrobial Enterobacter spp. bacteremias (3.3 vs. 29.3%; P<0.02). Susceptibility of Enterobacter spp. strains isolated from 51 episodes was stable and showed only two episodes due to quinolone-resistant strains, both in 1992 despite of the use of ofloxacin in prophylaxis of neutropenic patients since 1990 in our institute. Ninety-two to 94% of all strains were susceptible to aminoglycosides, 96-98% to ofloxacin and ciprofloxacin, respectively and 94.9% to meropenem but only 75.5% to ceftazidime.

Viridans streptococcal bacteraemia due to penicillin-resistant and penicillin-sensitive streptococci: analysis of risk factors and outcome in 60 patients from a single cancer centre before and after penicillin is used for prophylaxis.

60 patients with 60 viridans streptococcal bacteraemic episodes (42 due to penicillin-sensitive and 18 due to penicillin-resistant viridans streptococci) were analysed in a population of 12,185 admissions and 1,380 bacteraemic episodes during a 7-year period in a National Cancer Institute. The incidence of viridans streptococci among bacteraemias decreased from 11.5% in 1989 to 2.5% in 1995 after penicillin was introduced for prophylaxis of febrile neutropenia in acute leukaemia in 1993. However, the proportion of penicillin-resistant viridans streptococcal bacteraemias increased from 0 in 1989 and 1990 before any prophylaxis was given, to 12.9-16.7% after quinolones were used for prophylaxis in 1991 and 1992, and to 44.4-81.8% in 1993-1995 after penicillin was added to the quinolones. Mortality rate was higher in the subgroup of penicillin-resistant viridans streptococcal bacteraemias (p < 0.05). Statistically significant risk factors in patients with penicillin-resistant (compared with penicillin-sensitive) viridans streptococcal bacteraemia were: acute leukaemia (p < 0.03), high doses of cytarabine (p < 0.05), mucocutaneous lesions (p < 0.004), breakthrough bacteraemia during prophylaxis with ofloxacine plus penicillin (p < 0.001). Multiple logistic regression analysis showed that only acute leukaemia (OR 2.05, CI 0.85-1.85, p < 0.00452) and penicillin-resistance (OR 0.71, CI 0.103-4.887, p < 0.0209) were significant independent predictors of inferior outcome. Breakthrough bacteraemia during empiric therapy with vancomycine occurred in 5 of 116 patients treated with vancomycine, and during therapy with ampicillin plus gentamicin in 6 patients of 18 treated.