In vitro activities of the oxazolidinone compounds linezolid (PNU-100766) and eperzolid (PNU-100592) against middle ear isolates of Streptococcus pneumoniae.

Two hundred and sixteen isolates of Streptococcus pneumoniae recovered between 1994 and 1996 from the middle ears of children with acute otitis media were tested for their susceptibility to penicillin, erythromycin, clindamycin and the oxazolidinones, linezolid (PNU-100766) and eperezolid (PNU-100592). There were 116 isolates from the Children's Hospital of Pittsburgh and 100 isolates from a national collection. Eighty percent of the local strains were susceptible to penicillin (MIC < 0.1 mg/l); 20% of the local strains and all of the national strains had reduced susceptibility to penicillin. All strains of S. pneumoniae tested had an MIC < 2.0 mg/l for both oxazolidinones. A regional difference was noted in the frequency of resistance to erythromycin with local isolates being more susceptible than isolates from the national collection. This difference was most pronounced among the high-level penicillin-resistant strains of S. pneumoniae.

A streptococcal score card revisited.

OBJECTIVE: To evaluate the utility of a simple scoring system as a predictor of obtaining a positive throat culture for group A streptococci (GAS). DESIGN: Prospective descriptive study. Scores were assigned prior to the availability of the results of throat cultures. SETTING: Emergency department and walk-in clinic of the Children's Hospital of Pittsburgh. PATIENTS: Patients were 365 children between the ages of two and 16 years with acute onset of sore throat and a history of or documentation of fever within the preceding 24 hours. INTERVENTIONS: A streptococcal score was assigned on the basis of a 6-point schema in which the features were 1) age; 2) season; 3) temperature of at least 38.3 degrees C; 4) adenopathy; 5) pharyngeal erythema, edema, or exudate; and 6) no symptoms of a viral upper respiratory infection (conjunctivitis, rhinorrhea, or cough). A throat culture was performed for the isolation of GAS. MAIN OUTCOME MEASURE: Positive predictive value of the streptococcal score in identifying children with a positive throat culture for GAS. RESULTS: A score of 5 or 6 predicted a positive culture for GAS in 59 and 75% of children, respectively. In patients with evidence of acute pharyngitis, the combination of age between five and 15 years, fever and absence of upper respiratory symptoms predicted a positive culture for GAS in 72% of patients. CONCLUSIONS: The score can be used to predict the likelihood that a throat culture will be positive for GAS.

Recovery of ceftazidime-resistant Klebsiella pneumoniae from pediatric liver and intestinal transplant recipients.

The purpose of this report was to determine the frequency of colonization and bacteremia with ceftazidime-resistant Klebsiella pneumoniae among pediatric candidates for and recipients of liver (LTx) and/or intestinal (ITx) transplantation. Between January and December 1994, surveillance stool cultures obtained from 51 children on the abdominal transplant service were planted on a selective medium containing 2 microg/ml of ceftazidime. Isolates of K. pneumoniae which grew on the selective medium were screened for extended-spectrum beta-lactamase (ESBL) production by the double-disk synergy test and underwent microdilutional susceptibility testing. Strain relatedness of ceftazidime-resistant K. pneumoniae was analyzed by field inversion gel electrophoresis (FIGE). Sixteen of 51 patients had > or = 1 positive stool cultures for ceftazidime-resistant K. pneumoniae; 9 out of 16 on the first culture obtained 1-23 days (median = 5) after admission. All 9 had been in hospital prior to this admission. Four others were positive on their first culture but were in our hospital for > 1 month at the onset of the study. Three patients became colonized after admission. Colonization with ceftazidime-resistant K. pneumoniae was more frequent among recipients of ITx (including combined LTx-ITx) compared to LTx alone (7/13 vs. 7/32, p<0.05). All of the ceftazidime-resistant isolates recovered from surveillance stool cultures had positive double-disk diffusion tests suggesting the presence of ESBL production as the mechanism of resistance. Ceftazidime resistance was identified in 7/8 episodes of bacteremia due to K. pneumoniae in patients on the abdominal transplant service compared with 0/17 episodes in other children in the Children's Hospital of Pittsburgh during the study. During this same time period, 69/312 clinical isolates of K. pneumoniae evaluated in the hospital laboratory were ceftazidime-resistant; 67/69 came from patients on this service. In none of the 312 isolates was resistance to cefotaxime found in the absence of ceftazidime resistance. Unique clones were identified for 10/19 isolates of ceftazidime-resistant strains of K. pneumoniae analyzed by FIGE. Colonization and bacteremia with ceftazidime-resistant K. pneumoniae were commonly identified among recipients of LTx & ITx at the Children's Hospital of Pittsburgh. The mechanism of resistance appeared to be due to the presence of ESBL production by resistant strains. Although resistant strains were frequently recovered from patients on the abdominal transplant service, recovery of ceftazidime-resistant strains from patients outside of this service was rare even in the intensive care setting.

Field inversion gel electrophoretic analysis of Legionella pneumophila strains associated with nosocomial legionellosis in children.

Two nosocomial cases of Legionnaires' disease occurred in children. Legionella pneumophila serogroup 1 was isolated from both patients and 30 of 39 plumbing system sites in the hospital. The patient and hospital environmental isolates yielded identical field inversion gel electrophoretic patterns which differed from patterns observed with epidemiologically unrelated strains.

Comparison of M and T type antigen testing to field inversion gel electrophoresis in the differentiation of strains of group A streptococcus.

Recent clusters of patients with acute rheumatic fever and invasive group A Streptococcus (GAS) have stimulated renewed interest in the epidemiology of streptococcal infections. We compared conventional serotyping for M and T antigens and serum opacity factor with field inversion gel electrophoresis (FIGE) for distinguishing among GAS. Fifteen pairs of throat isolates obtained from children positive for GAS before and after therapy were evaluated by conventional serotyping and by FIGE after SmaI digestion. Ten of the 15 pairs were identical by serotyping. FIGE correctly identified the 10 concordant and 5 discordant pairs. Individual clones were identified within each M type tested, including analysis performed on additional isolates of M1 and M3 obtained from the Centers for Disease Control and Prevention. This preliminary experience suggests that FIGE can successfully determine whether serial isolates from a given patient represent persistence of one strain or acquisition of a new strain of GAS and that this method might provide an alternative typing system for GAS.

Comparison of field inversion gel electrophoresis with contour-clamped homogeneous electric field electrophoresis as a typing method for Enterococcus faecium.

Direct comparisons between contour-clamped homogeneous electric field (CHEF) electrophoresis and field inversion gel electrophoresis (FIGE) to determine the epidemiology of antibiotic-resistant enterococci have not been previously published. Fifty non-beta-lactamase-producing, ampicillin-resistant Enterococcus faecium isolates and 10 vancomycin-resistant E. faecium strains collected from multiple centers were analyzed in a blinded fashion by CHEF electrophoresis and FIGE after digestion with SmaI. Isolates were considered clonally related if there was a difference of three of fewer bands between electrophoretic patterns. Agreement between CHEF electrophoresis and FIGE was seen for 12 of 13 identified groups of ampicillin-resistant E. faecium and 7 of 7 groups of vancomycin-resistant E. faecium. The lone discordance was accounted for by a fourth band difference between two strains recognized near 350 kb by CHEF electrophoresis but not by FIGE, placing them into different clonal groups. Better band separation was noted in the 50- to 200-kb range for FIGE, while CHEF electrophoresis revealed better resolution over 250 kb more reliably, including detection of some bands not seen on FIGE. Molecular epidemiologic investigations of E. faecium by either technique should provide comparable results.

Bacteremia due to vancomycin-dependent Enterococcus faecium.

A recipient of small-bowel and liver transplants developed recurrent fever and polymicrobial bacteremia due to multiply resistant Enterobacter cloacae and an inducible VanB strain of Enterococcus faecium while receiving therapy with amikacin, imipenem, and vancomycin. These organisms could not be subcultured onto blood agar but did grow around the vancomyin disk on a direct-susceptibility test plate. Additional testing confirmed the strain as E. faecium, which would not grow in the absence of vancomycin. Growth around a disk containing D-alanyl-D-alanine was demonstrated. Spontaneous vancomycin-independent revertants were obtained at a frequency of approximately 1 x 10(-6). Two classes of vancomycin-independent revertants were obtained: one that was constitutively vancomycin resistant and one that was nonconstitutively vancomycin resistant. We hypothesize that the normal D-ala ligase is not expressed in the vancomycin-dependent strain; thus survival of these strains is dependent on expression of the VanB ligase, which produces a depsipeptide precursor that is resistant to vancomycin binding. This is the second reported case involving a clinically important vancomycin-dependent enterococcal strain. Awareness of the existence of these strains is important, especially when clinical and microbiological data are consistent with infection due to a fastidious or nutritionally-deficient organism.

Simple test of synergy between ampicillin and vancomycin for resistant strains of Enterococcus faecium.

The combination of ampicillin and vancomycin kills some but not all strains of ampicillin- and vancomycin-resistant Enterococcus faecium. We compared a simple test for synergy utilizing a commercially available microdilution susceptibility system with time-kill studies and determined acceptable breakpoints for this test for 20 strains of ampicillin- and vancomycin-resistant E. faecium. The combination of ampicillin and vancomycin was tested for synergy by time-kill, broth macrodilution, and broth microdilution procedures. Repeat testing of isolates by macro- and microdilution synergy methods yielded MICs that were within one twofold dilution of each other for both intra- and intertest comparisons. Synergy was always detected by time-kill studies when the MIC of ampicillin in the combination synergy screen was < or = 8 micrograms/ml in the presence of vancomycin. No synergy was detected when the MIC was > 16 micrograms/ml in the combination microdilution synergy screen. The determination of the synergy by the broth microdilution procedure appears to be simple, convenient, and accurate.

Constitutively vancomycin-resistant Enterococcus faecium resistant to synergistic beta-lactam combinations.

Vancomycin resistance among enterococci has recently been recognized. Synergy between vancomycin and penicillin has been shown in vitro for isolates of Enterococcus faecium resistant to both of these antibiotics. We describe three isolates of vancomycin-resistant E. faecium which demonstrate unique phenotypic characteristics. The isolates exhibited high-level resistance to both vancomycin and teicoplanin, consistent with the VanA phenotype. However, resistance in these isolates could not be induced or cured, and mating experiments failed to detect a transfer of resistance. The combination of vancomycin and penicillin did not significantly change the MIC of penicillin for any of the three isolates. Immunoblotting with polyclonal anti-VanB antibody showed no reaction with the cellular proteins of these strains. Probing with a vanA oligonucleotide revealed hybridization with chromosomal but not plasmid DNA. The mechanism of constitutive resistance of those strains remains unclear. A second mutational change, perhaps involving PBP 5, may explain the presence of resistance to synergistic combination penicillin-vancomycin therapy. In vitro evaluation of penicillin-vancomycin should be carried out in all clinical cases where this therapeutic regimen is being considered.

Respiratory syncytial virus: a comparison of diagnostic modalities.

This study compared prospectively viral culture for respiratory syncytial virus (RSV) with three rapid RSV antigen detection tests: RSV EIA and TestPack RSV (TP), and Directigen RSV (DIR). Additionally two methods of specimen collection were compared: nasopharyngeal rub (RUB) and nasopharyngeal wash (WASH). True positives were defined as positive RSV viral culture or at least two positive antigen tests. One hundred ninety-eight WASH specimens obtained from children 3 years of age or younger during the 1991 RSV winter epidemic were tested for RSV antigen. Sensitivity and specificity of WASH specimens were 59.5 and 100% (culture), 86.2 and 98.2% (RSV EIA), 91 and 96.3% (TP) and 83.5 and 94.3% (DIR). Concurrently obtained RUB samples from 124 children were tested by TP and/or DIR. Sensitivity of RUB specimens was significantly lower than that of WASH specimens; 64.9% (TP) and 43.6 (DIR). These easy to perform, rapid RSV antigen tests for WASH specimens provide timely diagnosis thereby facilitating management decisions and isolation efforts.

Recovery of vancomycin-resistant gram-positive cocci from pediatric liver transplant recipients.

Between November 1988 and October 1989, 49 first-time pediatric liver transplant recipients at the Children's Hospital of Pittsburgh were prospectively monitored for the presence of stool colonization and the development of disease caused by vancomycin-resistant gram-positive cocci (VRGPC). Quantitative stool culturing was done on a weekly basis, and cultures were planted onto a selective medium for VRGPC. Isolates for which the MIC was greater than or equal to 8 were considered resistant to vancomycin. Patients were monitored clinically for the development of infection, and their charts were systematically reviewed for the use of antibiotics. Eighty-six isolates were recovered from 36 of the 49 patients. Enterococcal species were isolated from 31 patients and included Enterococcus gallinarum (n = 28), E. casseliflavus (n = 14), E. faecium (n = 9), E. faecalis (n = 2), E. mundtii (n = 2), and E. durans (n = 1). Stool colonization with vancomycin-resistant enterococci was noted to increase steadily during the first month after transplantation. Only 9 of 31 patients demonstrated clearance of these organisms in serial repeat cultures. Additional isolates of VRGPC included Lactobacillus confusus (n = 13), Lactobacillus spp. (n = 12), and Pediococcus pentosaceus (n = 4). Infection due to VRGPC developed in three patients: a urinary tract infection in two and peritonitis in one. E. faecium was the pathogen in each of these cases. The ranges of MICs of vancomycin were 8 to 32 micrograms/ml for all enterococcal isolates and greater than 128 micrograms/ml for Lactobacillus and Pediococcus isolates. All Lactobacillus and Pediococcus isolates were resistant to teicoplanin, although they were susceptible to daptomycin. All other isolates were susceptible to both teicoplanin and daptomycin. This study demonstrates that stool colonization with VRGPC may be a common and early finding among pediatric liver transplant recipients. However, infection appears to be uncommon.

Recovery of vancomycin-resistant gram-positive cocci from children.

A cross-sectional survey of vancomycin-resistant gram-positive cocci (VRGPC) in the feces of children was initiated after several bacteremic infections with these organisms occurred at our hospital. A selective medium consisting of colistin-nalidixic acid agar, 5% sheep blood, vancomycin (5 mg/liter), and amphotericin B (8 mg/liter) was developed to isolate VRGPC. A single stool specimen submitted to the clinical microbiology laboratory from each of 48 patients was inoculated onto the medium. Plates were incubated at 35 degrees C with 5% carbon dioxide and examined at 24, 48, and 72 h. Susceptibilities were determined by broth microdilution. A total of 14 isolates from 11 of 48 (22%) children were recovered. The density of growth ranged from a single colony to 2+. The VRGPC were identified as Leuconostoc lactis (n = 2), Lactobacillus confusus (n = 4), Enterococcus species (n = 5), and Lactococcus lactis (n = 3). One strain of Lactobacillus confusus was recovered from both the stool and the blood of one of these patients. The MICs of vancomycin were 4 micrograms/ml for one of the isolates, 8 micrograms/ml for four of the isolates, and more than 16 micrograms/ml for the remaining eight isolates. All isolates were susceptible to both penicillin and ampicillin. Only 1 of the 11 children had received prior treatment with vancomycin. We conclude that low concentrations of VRGPC may be common in the gastrointestinal tracts of children.