Occurrence and detection of AmpC beta-lactamases among Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis isolates at a veterans medical center.

AmpC beta-lactamases are cephalosporinases that confer resistance to a wide variety of beta-lactam drugs and that may thereby create serious therapeutic problems. Although reported with increasing frequency, the true rate of occurrence of AmpC beta-lactamases in Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis remains unknown. We tested a total of 1,286 consecutive, nonrepeat isolates of these three species and found that, overall, 45 (3.5%) yielded a cefoxitin zone diameter less than 18 mm (screen positive) and that 16 (1.2%) demonstrated AmpC bands by isoelectric focusing. Based on the species, of 683 E. coli, 371 K. pneumoniae, and 232 P. mirabilis isolates tested, 13 (1.9%), 28 (7.6%), and 4 (1.7%), respectively, demonstrated decreased zone diameters and 11 (1.6%), 4 (1.1%), and 1 (0.4%), respectively, demonstrated AmpC bands. Cefoxitin resistance was transferred for all but 8 (E. coli) of the 16 AmpC producers. We also describe a three-dimensional extract test, which was used to detect phenotypically isolates that harbor AmpC beta-lactamase. Of the 45 cefoxitin-resistant isolates, the three-dimensional extract test accurately identified all 16 AmpC producers and 28 of 29 (97%) isolates as non-AmpC producers. Interestingly, most (86%) isolates in the latter group were K. pneumoniae isolates. These data confirm that, at our institution, E. coli, K. pneumoniae, and P. mirabilis harbor plasmid-mediated AmpC enzymes.

Bismuth-mediated disruption of the glycocalyx-cell wall of Helicobacter pylori: ultrastructural evidence for a mechanism of action for bismuth salts.

The mechanism of bismuth's bactericidal activity against Helicobacter pylori was investigated using transmission electron microscopy (TEM) and analytical electron microscopy (AEM); time-kill kinetic methods evaluated the effect of excess divalent cations. TEM analysis of untreated H. pylori revealed a normal morphology. In contrast, H. pylori exposed to bismuth salts had swollen, distorted cells with membrane-cell wall blebbing and a cytoplasm containing electron-dense, sometimes crystalline aggregates. By AEM, swollen cells contained bismuth at the cell periphery, whereas bacillary forms contained cytoplasmic bismuth localizations. Time-kill studies showed that the bactericidal activity of bismuth could be prevented by pretreatment with divalent cations. The effects of bismuth salts on the glycocalyces-cell walls of H. pylori with reversal of bactericidal activity by divalent cations are identical to those produced by other polycationic agents on various gram-negative bacilli. We conclude that disruption of the glycocalyces-cell walls of H. pylori is one mechanism of action for bismuth salts.

In-vitro evaluation of nitrofurantoin as an alternative agent for metronidazole in combination antimicrobial therapy against Helicobacter pylori.

Increasing metronidazole resistance suggests the need for alternative antibiotics for combination therapy of Helicobacter pylori infections. We evaluated a metronidazole-resistant and a clarithromycin-resistant strain of H. pylori under stationary growth phase conditions that favoured physiological conditions in order to determine if nitrofurantoin might be a suitable alternative for metronidazole in combination therapy. The results demonstrated that the triple combination of bismuth, tetracycline and nitrofurantoin achieved greater bactericidal activity against these two strains than did the combination of bismuth, tetracycline and metronidazole. These results suggest that further evaluation is warranted.

The bactericidal activity of clarithromycin versus ampicillin alone and in combination with omeprazole and/or bismuth against clarithromycin-susceptible and clarithromycin-resistant strains of Helicobacter pylori.

We evaluated the in vitro bactericidal effect of clarithromycin versus ampicillin alone and in combination against clarithromycin-sensitive and clarithromycin-resistant strains of Helicobacter pylori. No combination containing clarithromycin achieved complete bactericidal effect against clarithromycin-resistant strains. Complete bactericidal effect was achieved for all strains by triple-agent combinations that contained bismuth, omeprazole, and relatively high concentrations of ampicillin. These in vitro results demonstrate the additive bactericidal activity provided by a combination of agents for the eradication of H. pylori. Bismuth may play a particularly important role in the bactericidal effect.

Hospital-wide restriction of clindamycin: effect on the incidence of Clostridium difficile-associated diarrhea and cost.

BACKGROUND: Widespread antibiotic use has been associated with increases in both bacterial resistance and nosocomial infection. OBJECTIVE: To characterize the impact of hospital-wide clindamycin restriction on the incidence of Clostridium difficile-associated diarrhea and on antimicrobial prescribing practices. DESIGN: Prospective, observational cohort study. SETTING: University-affiliated Veterans Affairs Medical Center. PATIENTS: Hospitalized patients with symptomatic diarrhea. MEASUREMENTS: Clinical data on individual patients and data on antibiotic use were obtained from hospital pharmacy records. Hospital-wide use of antimicrobial agents was monitored. Isolates of C. difficile underwent antimicrobial susceptibility testing and molecular typing. RESULTS: An outbreak of C. difficile-associated diarrhea was caused by a clonal isolate of clindamycin-resistant C. difficile and was associated with increased use of clindamycin. Hospital-wide requirement of approval by an infectious disease consultant of clindamycin use led to an overall reduction in clindamycin use, a sustained reduction in the mean number of cases of C. difficile-associated diarrhea (11.5 cases/month compared with 3.33 cases/month; P < 0.001), and an increase in clindamycin susceptibility among C. difficile isolates (9% compared with 61%; P < 0.001). A parallel increase was noted in the use of and costs associated with other antibiotics with antianaerobic activity, including cefotetan, ticarcillin-clavulanate, and imipenem-cilastin. The hospital realized overall cost savings as a result of the decreased incidence of C. difficile-associated diarrhea. CONCLUSIONS: Hospital formulary restriction of clindamycin is an effective way to decrease the number of infections due to C. difficile. It can also lead to a return in clindamycin susceptibility among isolates and can effect cost savings to the hospital.

Plasmid-mediated resistance to expanded-spectrum cephalosporins among Enterobacter aerogenes strains.

Resistance to expanded-spectrum cephalosporins commonly develops in Enterobacter aerogenes during therapy due to selection of mutants producing high levels of the chromosomal Bush group 1 beta-lactamase. Recently, resistant strains producing plasmid-mediated extended-spectrum beta-lactamases (ESBLs) have been isolated as well. A study was designed to investigate ESBL production among 31 clinical isolates of E. aerogenes from Richmond, Va., with decreased susceptibility to expanded-spectrum cephalosporins and a positive double-disk potentiation test. Antibiotic susceptibility was determined by standard disk diffusion and agar dilution procedures. Beta-lactamases were investigated by an isoelectric focusing overlay technique which simultaneously determined isoelectric points (pIs) and substrate or inhibitor profiles. Decreased susceptibility to cefotaxime, ceftazidime, and aztreonam (MIC range, 1 to 64 microg/ml) was detected and associated with resistance to gentamicin and trimethoprim-sulfamethoxazole. All strains produced an inducible Bush group 1 beta-lactamase (pI 83). Twenty-nine of the 31 isolates also produced an enzyme similar to SHV-4 (pI 7.8), while 1 isolate each produced an enzyme similar to SHV-3 (pI 6.9) and to SHV-5 (pI 8.2). The three different SHV-derived ESBLs were transferred by transconjugation to Escherichia coli C600N and amplified by PCR. Plasmid profiles of the clinical isolates showed a variety of different large plasmids. Because of the linkage of resistance to aminoglycosides and trimethoprim-sulfamethoxazole with ESBL production, it is possible that the usage of these drugs was responsible for selecting plasmid-mediated resistance to extended-spectrum cephalosporins in E. aerogenes. Furthermore, it is important that strains such as these be recognized, because they can be responsible for institutional spread of resistance genes.

Occurrence and detection of extended-spectrum beta-lactamases in members of the family Enterobacteriaceae at a veterans medical center: seek and you may find.

A total of 907 consecutive isolates of members of the family Enterobacteriaceae recovered during a 20-week period were tested for production of extended-spectrum beta-lactamases (ESBLs) by the double-disk (DD) potentiation method. Of 84 DD-positive isolates, 83 (9.2%) produced ESBLs based on isoelectric focusing. SHV-derived ESBLs and several TEM-derived ESBLs were present in nine species, including the first isolate of Citrobacter koserii and Morganella morganii known to harbor an SHV-derived ESBL. Results of testing 58 nonrepeat isolates for ESBL production by several recommended methods were as follows (percent detected in parentheses): DD method with aztreonam (95), ceftazidime (79), ceftriaxone (88), or cefpodoxime (90); broth microdilution method with ceftazidime (86) or cefotaxime (91) alone or in combination with clavulanate; and the standard disk diffusion method with new breakpoints and standard concentrations of aztreonam (78), ceftazidime (79), ceftriaxone (83), or cefpodoxime (98) or a novel concentration (5 microg) of ceftazidime (88). In three instances during an extended part of the study, an ESBL-producing isolate and a non-ESBL-producing isolate of the same species were recovered from a single blood culture bottle. These data indicate that ESBLs occur in several species of Enterobacteriaceae and at a relatively high incidence at our institution and that the standard disk diffusion method with cefpodoxime and the DD method with several beta-lactams are practical and cost-effective methods for detecting ESBL-producing isolates of Enterobacteriaceae.

Chromosomal DNA restriction endonuclease digestion patterns of beta-lactamase-producing Enterococcus faecalis isolates collected from a single hospital over a 7-year period.

Twenty-three beta-lactamase (beta-lac)-producing, highly gentamicin-resistant Enterococcus faecalis isolates collected over a 7-year period from the same hospital were examined by pulsed-field gel electrophoresis of SmaI-digested genomic DNA. The beta-lac+ isolates appeared to form a single clonal group, which had been previously designated the mid-Atlantic pattern. Eleven variations of the mid-Atlantic clone, differing by one to six bands, were identified; some of the changes were likely due to plasmid bands. However, a number of isolates had indistinguishable patterns, including some recovered over a 4-year period. There was a surprising lack of movement of the beta-lac determinant to other strains, although this trait was transferable in vitro by conjugation. We conclude that a single clone (the mid-Atlantic clone) of beta-lac+ E. faecalis has remained endemic in this hospital for at least 7 years. The reason(s) for the apparent lack of spread to other strains of E. faecalis is unknown.

Use of time-kill methodology to assess antimicrobial combinations against metronidazole-susceptible and metronidazole-resistant strains of Helicobacter pylori.

Optimal therapy for Helicobacter pylori infection to date, consists of metronidazole, bismuth, and tetracycline. This combination, however, is less effective against metronidazole-resistant organisms. We used a time-kill kinetic methodology to assess the bactericidal effects of selected agents, alone and in combination, to a metronidazole-susceptible and a metronidazole-resistant strain of H. pylori. single, double, and triple agents showed increasing bactericidal activity. The combination of metronidazole, bismuth, and tetracycline showed maximal killing effect (no detectable regrowth) against the susceptible strain, but against the resistant strain this combination showed less killing. The time-kill methodology may therefore offer an in vitro approach to the initial selection of agents to be evaluated for the treatment of H. pylori infections.

Factors affecting growth and susceptibility testing of Helicobacter pylori in liquid media.

In order to increase the database for in vitro growth and/or susceptibility testing in liquid media, we evaluated the growth of Helicobacter pylori in broth media containing 5% sheep blood. We also compared the effect of bismuth on the growth of H. pylori in broth media containing 10% fetal calf serum with the effect on growth in media containing 0.5% starch. In contrast to the result seen with agar, we found that sheep blood, whether whole or laked, inhibited the growth of H. pylori in broth media. In addition, we found that bismuth inhibited growth in media with starch but that this inhibition was negated in media with serum.

Utilization of time-kill kinetic methodologies for assessing the bactericidal activities of ampicillin and bismuth, alone and in combination, against Helicobacter pylori in stationary and logarithmic growth phases.

Assessment of in vitro susceptibility testing of Helicobacter pylori is difficult because of the fastidious, slowly growing nature of this microorganism. The high rate of relapse observed clinically and a possible subpopulation of cells that are not actively replicating suggest the potential need for bactericidal therapy in order to eradicate H. pylori. We used modified time-kill kinetic methodology in order to evaluate the bactericidal activities of ampicillin and bismuth, alone and in combination, against three strains of H. pylori in both a stationary (slow) growth phase and a logarithmic (rapid) growth phase. We found that ampicillin produced a decrease in CFU per milliliter (2 to 4 log10 units) for three strains of H. pylori when tested in logarithmic growth phases but was less inhibitory (< 1-log10-unit decrease in CFU per milliliter) when tested in a stationary growth phase. In contrast, bismuth, when tested in a logarithmic growth phase, produced little inhibitory effect, as the CFU for all strains tested increased above the inoculum. However, when tested in a stationary growth phase, bismuth produced a decrease in CFU per milliliter of < 1 to > 3 log10 units). The activities of these two agents when combined mimicked the activity of the most active drug alone for that growth phase. We conclude that the clinical use of ampicillin combined with bismuth has been more effective than that of either agent used alone because ampicillin targets replicating cells, whereas bismuth targets cells that are not actively replicating.

Comparison of ampicillin-sulbactam with vancomycin for treatment of experimental endocarditis due to a beta-lactamase-producing, highly gentamicin-resistant isolate of Enterococcus faecalis.

Increasing antibiotic resistance in the enterococci, including the capacity for beta-lactamase production and the development of high-level aminoglycoside resistance, has complicated the treatment of serious enterococcal infections, which often require synergistic antibiotic combinations for cure. We utilized the rabbit model of aortic valve endocarditis to investigate the effects of various antibiotics, alone and in combination, against a multiply antibiotic-resistant isolate of Enterococcus faecalis. Female New Zealand White rabbits were infected with either a beta-lactamase-producing, gentamicin-resistant isolate of E. faecalis or a non-beta-lactamase-producing, aminoglycoside-susceptible isolate, and the mean log10 CFU per gram of vegetation were determined. The most active agents were low-dose ampicillin-sulbactam (200 mg/kg of body weight per day), high-dose ampicillin-sulbactam (400 mg/kg of body weight per day), and vancomycin (150 mg/kg of body weight per day), which reduced the titers of bacteria by 2.27, 2.76, and 2.85 log10 (CFU/g, respectively, compared with controls. While ampicillin-sulbactam and vancomycin were equally efficacious in reducing titers of bacteria in vegetations, no animals were cured (defined as < 2 log10 CFU/g of vegetation) by either agent, whether treatment was continued for 3 or 7 days. The addition of gentamicin was not associated with increased killing in rabbits infected with the aminoglycoside-resistant isolate. Both high-dose ampicillin-sulbactam and vancomycin regimens demonstrated significant, continued reduction in bacterial titers with the longer periods of treatment (P < or = 0.05); 7-day treatment with high-dose ampicillin-sulbactam produced a greater reduction in bacterial titers in vegetation than 7-day treatment with vancomycin (P < or = 0.05). We conclude that ampicillin-sulbactam and vancomycin are equally effective in the treatment of experimental endocarditis due to beta-lactamase-producing, highly gentamicin-resistant E. faecalis. The optimum therapy for such infections in humans is not known.

Molecular typing of ampicillin-resistant, non-beta-lactamase-producing Enterococcus faecium isolates from diverse geographic areas.

Molecular typing methods were compared by using 66 ampicillin-resistant, non-beta-lactamase-producing Enterococcus faecium clinical isolates from diverse geographic areas. Whole-plasmid analysis, restriction enzyme analysis of plasmid DNA with EcoRI and HindIII, and contour-clamped homogeneous electric field electrophoresis with digestion by SmaI and ApaI were performed on all isolates. Whole-plasmid analysis identified 47 different groups. Restriction enzyme analysis of plasmid DNA identified 50 groups when EcoRI was used and 51 groups when HindIII was used. Results with EcoRI and HindIII differed in 9 of 66 isolates. Grouping results with whole-plasmid analysis differed from results of restriction enzyme analysis of plasmid DNA (combining EcoRI and HindIII) in 20 of 66 isolates. Contour-clamped homogeneous electric field electrophoresis identified 46 groups when SmaI was used and 44 groups when ApaI was used. Results with SmaI and ApaI differed in 3 of 66 isolates. Grouping results with contour-clamped homogeneous electric field electrophoresis (combining SmaI and ApaI) differed from results of restriction enzyme analysis of plasmid DNA (combining EcoRI and HindIII) in 17 of 66 isolates. The combined use of whole-plasmid analysis, restriction enzyme analysis of plasmid DNA with two enzymes, and contour-clamped homogeneous electric field electrophoresis with two restriction enzymes should be considered when E. faecium is typed for epidemiologic investigation.

Isolation of a beta-lactamase-producing, aminoglycoside-resistant strain of Enterococcus faecium.

Beta-Lactamase-producing, aminoglycoside-resistant strains of Enterococcus faecalis have been isolated from different geographic areas and are endemic at our institution. We report the isolation of a beta-lactamase-producing, aminoglycoside-resistant strain of E. faecium. The beta-lactamase was plasmid mediated and transferable with high frequency into a plasmid-free E. faecalis recipient strain. MICs suggested that the E. faecium strain also contained intrinsic (chromosomal) resistance to penicillins.

Infections due to beta-lactamase-producing, high-level gentamicin-resistant Enterococcus faecalis.

OBJECTIVE: To investigate the risk factors, clinical features, molecular epidemiology, and treatment outcomes associated with an outbreak of infections due to beta-lactamase-producing, high-level gentamicin-resistant Enterococcus faecalis. DESIGN: Case-control and molecular genetics study. SETTING: Tertiary care Veterans Affairs hospital. PATIENTS: Sixty-five patients infected or colonized with beta-lactamase-producing high-level gentamicin-resistant E. faecalis (case patients) were matched and compared with 65 randomly selected patients infected or colonized with beta-lactamase-negative, gentamicin-susceptible E. faecalis. MEASUREMENTS AND MAIN RESULTS: During the 20-month study period, 124 of 1506 isolates (8.2%) of E. faecalis from 70 patients were found to produce beta-lactamase. Univariate analysis showed older age, higher APACHE II score, nosocomial acquisition, recent surgical procedure, total parenteral nutrition, and antibiotic treatment to be significantly associated with the acquisition of beta-lactamase-producing, high-level gentamicin-resistant E. faecalis. A multivariate analysis, done using stepwise multiple logistic regressions, showed that only two variables remained significant: an APACHE II score greater than 6 (odds ratio, 9.5; 95% CI, 4.4 to 20.3) and antibiotic treatment (odds ratio, 10.2; CI, 4.5 to 23.2). The mortality rate for case patients was 7.7% (5 of 65 patients; CI, 1.2% to 14.2%); no control patient died. Of 23 infections occurring in case patients, 13 were treated with "inappropriate" antibiotics (regimens that included a beta-lactamase-unstable antibiotic); 2 patients improved and 11 had complete resolution of disease. "Appropriate" treatments (regimens that included a beta-lactamase-stable antibiotic) were used in 10 patients; 5 of 10 infections were fatal. Restriction enzyme digests of total chromosomal DNA showed nearly identical patterns for selected isolates of beta-lactamase-producing, high-level gentamicin-resistant E. faecalis, suggesting dissemination through the hospital of a single strain of E. faecalis. CONCLUSIONS: Fatal infections were observed despite treatment with beta-lactamase-stable antibiotics. The risk for infection or colonization with beta-lactamase-producing, high-level gentamicin-resistant E. faecalis was strongly associated with severe underlying disease (acute physiology and chronic health evaluation [APACHE] II score, greater than 6) and previous antibiotic treatment. These results may be useful in targeting high-risk patients for infection-control interventions.

Coryneform group A-4 endophthalmitis. An experimental animal model.

Diphtheroids, members of the coryneform family of bacteria, increasingly have been recognized as the cause of serious ocular diseases. After isolation of coryneform group A-4 from two patients with delayed endophthalmitis after cataract extraction and intraocular lens implantation, 10(7) organisms were injected into the vitreous of seven New Zealand white rabbits, producing endophthalmitis in all eyes inoculated. Coryneform group A-4 subsequently was isolated in six of seven eyes receiving 10(7) organisms, proving Koch's postulates. Five of these seven eyes were treated with a single dose of intravitreal gentamicin, and three eyes remained culture positive. Eyes inoculated with 10(5) or 10(2) coryneform group A-4 organisms had transient anterior chamber and vitreal inflammation; all vitreous cultures were negative. These studies demonstrate that coryneform group A-4 endophthalmitis can be reproduced in an animal model and that gentamicin may not sterilize an eye infected with this organism. Future studies are needed to determine the optimum antibiotic regimen for treatment of this type of endophthalmitis.

Pneumonia and empyema infection associated with a Bacillus species that resembles B. alvei.

An organism resembling Bacillus alvei was isolated from the lung and pleural fluid of an immunocompetent patient. The isolate differed from the type strain of B. alvei in its ability to reduce nitrate and its inability to produce dihydroxyacetone and acetylmethylcarbinol. The isolate was resistant to ciprofloxacin and showed intermediate susceptibility to vancomycin.

Antimicrobial susceptibility and molecular epidemiology of beta-lactamase-producing, aminoglycoside-resistant isolates of Enterococcus faecalis.

beta-Lactamase-producing (BL+), aminoglycoside-resistant (AR) Enterococcus faecalis is endemic in our hospital, having caused widespread colonization and infection. Suitable therapy for infections caused by these organisms has been problematic. We compared the antimicrobial and bactericidal activities, by broth macrodilution and time-kill methods, of several antibiotics, alone and in combination, against BL+, AR isolates of E. faecalis and determined the transmissibility of antibiotic resistance markers. Ampicillin-sulbactam, imipenem, daptomycin, and ciprofloxacin were the most active antibiotics with MICs for 90% of isolates tested of 2, 1, 2, and 1 microgram/ml, respectively, against inocula of 10(3) and 10(5) CFU/ml. Little inoculum effect was noted with imipenem, vancomycin, daptomycin, or ciprofloxacin, while the addition of sulbactam to ampicillin partially inhibited the effect of the increased inoculum. Penicillin-sulbactam and ampicillin-sulbactam combinations in a 2:1 ratio were most frequently bactericidal (greater than or equal to 3-log10-unit decrease in bacterial titers at 24 h for 13 of 20 isolates), followed by daptomycin (8 of 20 isolates) and ciprofloxacin (2 of 20 isolates). Bactericidal activity was not demonstrated for imipenem or teicoplanin. beta-Lactamase production and aminoglycoside resistance were associated with a 60- to 65-MDa plasmid which was easily transferred to a plasmid-free E. faecalis recipient. The 840-bp beta-lactamase gene probe hybridized to purified plasmid DNA from BL+ donor isolates of E. faecalis and transconjugants but not from BL- isolates. Ampicillin-sulbactam and daptomycin (an investigational antibiotic) seem to be reasonable choices for the empiric therapy of presumed enterococcal infections in hospitals in which BL+, AR E. faecalis strains are isolated. Their use should ideally be supported by tests for bactericidal activity.

Coryneform endophthalmitis. Two case reports.

Recent clinical studies have emphasized the importance of diphtheroids, previously regarded as nonpathogenic bacteria or contaminants, as causes of ocular disease. We encountered two patients with endophthalmitis following cataract extraction and anterior chamber intraocular lens implantation. Both patients had previously been treated with subconjunctival and/or oral corticosteroids for presumed sterile endophthalmitis. Vitreous cultures in each case yielded pure growth of a diphtheroid that was subsequently identified as coryneform group A-4. The clinical response to standard intraocular therapy with gentamicin and cefazolin was delayed, although both patients eventually had restoration of functional vision. A comparison of the antibiotic minimum inhibitory and minimum bactericidal concentrations of the isolates may help to explain the delayed response to therapy seen in these two patients.

Listeria monocytogenes keratitis.

We treated a farmer who had Listeria monocytogenes bacterial keratitis. Therapy with topical antibiotics was unsuccessful; it was necessary to treat the patient with topical and systemic penicillin and gentamicin. To elucidate the pathogenesis of this infection, we developed a rabbit model. Using the patient's strain of L. monocytogenes, we determined that the severity of the rabbit infection was dose-related. If we used an inoculum of more than 10(7) organisms, many of the features of the human Listeria keratitis were mimicked. We also found that treatment with either penicillin or gentamicin did not control the infection as well as using both antibiotics simultaneously, a combination which resulted in relatively rapid resolution of infection and no corneal scarring. The human and animal data indicate that L. monocytogenes can be a virulent corneal pathogen. Listeria corneal infections must be treated aggressively with both penicillin and gentamicin to prevent permanent visual loss.