Tracheal colonization with Sphingomonas paucimobilis in mechanically ventilated neonates due to contaminated ventilator temperature probes.

Sphingomonas paucimobilis was isolated from tracheal secretions of a total of 85 mechanically ventilated babies in a neonatal intensive-care unit (NICU) during a two-year-period. None of the neonates developed pneumonia or sepsis. After each increase in the fluctuating number of S. paucimobilis isolates, extra attention was paid to hand hygiene and to the maintenance of the ventilation equipment. This resulted in a reduction of the frequency of isolation each time. Cultures of all liquids in use and of the ventilation equipment were negative on several occasions. Fifteen months after the start of the outbreak, the NICU was moved to another building, and some older ventilation equipment was abandoned. After a period of six weeks without problems, S. paucimobilis was isolated in association with at least four ventilators. A new investigation showed that the ventilator temperature probes were the source of contamination. Once effective sterilization procedures for the temperature probes were introduced no new cases appeared, until a spare ventilator with an unautoclaved temperature probe was accidentally used and this caused contamination of one child. After correction, no further cases have occurred to date. The clonal relatedness of the outbreak isolates from patients and from ventilator temperature probes was documented by fingerprinting with the arbitrarily primed polymerase chain reaction.

Case of recurrent Flavimonas oryzihabitans bacteremia associated with an implanted central venous catheter (Port-A-Cath): assessment of clonality by arbitrarily primed PCR.

Flavimonas oryzihabitans bacteremias, which occurred immediately after the flushing or use of an implanted central venous catheter (Port-A-Cath) in two patients at the same pediatric ward, were studied by arbitrarily primed PCR. We conclude that the colonization of the Port-A-Cath with F. oryzihabitans described here lasted for several months.

Comparison of arbitrarily primed polymerase chain reaction and cell envelope protein electrophoresis for analysis of Acinetobacter baumannii and A. junii outbreaks.

Two successive Acinetobacter outbreaks in a neonatal intensive care unit were studied with arbitrarily primed polymerase chain reaction (AP-PCR), cell envelope protein electrophoresis (protein fingerprinting) and antibiotic susceptibility testing. AP-PCR fingerprinting and protein fingerprinting yielded identical clustering of the isolates studied. Susceptibility test results were useful for rapid recognition of the outbreaks, but clustering of several isolates was different from the clustering obtained with AP-PCR fingerprinting and protein fingerprinting. Typing results indicated that the two outbreaks, which occurred at a three-month interval, were each caused by a single strain, and that both strains differed from the strains prevailing in the hospital. The strain of one outbreak was identified as A. junii, a species commonly not involved in outbreaks. A. baumannii isolates collected from different departments of this hospital during a period of four years clustered into only five different types. Moreover, strains from different departments of a second hospital belonged to the type prevailing in the first hospital, although there were no apparent connections between the two institutions. This may indicate that only a limited number of strains of the A. calcoaceticus-baumannii complex are involved in nosocomial outbreaks.

Identification of Acinetobacter genomic species by amplified ribosomal DNA restriction analysis.

A total of 53 field and reference strains, including the type strains of the seven named species (nomenspecies) and belonging to the 18 described genomic species (DNA groups) of the genus Acinetobacter, were studied by amplified ribosomal DNA restriction analysis (ARDRA). Restriction analysis with the enzymes AluI, CfoI, MboI, RsaI, and MspI of the enzymatically amplified 16S rRNA genes allowed us to identify all species except the genomic species 4 (Acinetobacter haemolyticus) and 7 (A. johnsonii), 5 (A. junii) and 17, and 10 and 11, which clustered pairwise in three respective groups. Further analysis with the enzyme HaeIII, HinfI, NciI, ScrFI, or TaqI did not allow us to differentiate the species within these three clusters. However, use of a few additional simple phenotypic tests (hemolysis, growth at 37 degrees C, production of acid from glucose, and gelatin hydrolysis) can be used to differentiate between the species within these clusters. ARDRA proved to be a rapid and reliable method for the identification of most of the Acinetobacter genomic species, including the closely related DNA groups 1 (A. calcoaceticus), 2 (A. baumannii), 3, and 13. The results of this study suggest that ARDRA can be used for the identification of Acinetobacter species and as such may help to elucidate the ecology and clinical significance of the different species of this genus. Since ARDRA uses universal 16S rRNA gene primers, it is expected to be applicable to the identification of most bacterial species. Furthermore, ARDRA is less prone to contamination problems than PCR for detection, since the use of cultured organisms results in a large initial quantity of target DNA.

Pseudomonas aeruginosa serotype O12 outbreak studied by arbitrary primer PCR.

A total of 16 colonizing and infecting ofloxacin-resistant Pseudomonas aeruginosa strains and two strains isolated from ventilation equipment fluids, all with similar colonial morphologies and with minor but distinct susceptibility differences, were suspected of belonging to a single outbreak and were studied by arbitrary primer (AP) PCR. Thirteen nonrelated strains were included to evaluate the discriminatory capacity of the technique. AP PCR fingerprinting was compared with serotyping, phage typing, and antibiotic susceptibility testing. AP PCR was performed independently with three different primers. The different AP PCR typing systems yielded almost identical patterns for the epidemic strains and enabled us to differentiate most of the nonrelated strains from each other and from the outbreak strains. The combination of AP PCR typing and the phenotyping techniques that we used enabled us to conclude that an outbreak was occurring. In general, the typeability of AP PCR was greater than those of phage typing and serotyping, while the discriminatory powers of the three methods were comparable.

Study of the influence of plasmids on the arbitrary primer polymerase chain reaction fingerprint of Escherichia coli strains.

To study the effect of plasmids on the arbitrary primer-polymerase chain reaction fingerprint of bacterial strains, the Escherichia coli strains DH5, Top10, and W3110 were transformed with plasmids of different sizes: respectively, pUC19, pCEP and two clinically important plasmids carrying resistance to several antibiotics. Total DNA, i.e. both chromosomal and plasmid DNA, was prepared from transformed cells by boiling the cell suspensions and by phenol-chloroform extraction; chromosomal DNA was prepared by the same methods from the non-transformed, plasmid-free strains; plasmid DNA of pUC19 was purchased; plasmid DNA of pCEP was purified from the transformed strains by caesium chloride density gradient centrifugation. Arbitrarily primed polymerase chain reaction was carried out for all of these preparations. Amplification carried out independently with three different primers resulted in similar patterns for the chromosomal preparations whether or not plasmid was present. Amplification of plasmid DNA gave different patterns, characterized by fragments larger than those obtained when total or chromosomal DNA were used as the target. These data illustrate that the plasmids studied here do not influence the chromosomal arbitrarily primed PCR fingerprint, although plasmids alone are amplified in the absence of chromosomal DNA. Experiments comparing different relative concentrations of plasmid and chromosomal DNA indicate that under natural conditions the amount of chromosomal DNA per cell is sufficient to inhibit observable amplification of the plasmid(s) present.

Identification of Mycobacterium species by using amplified ribosomal DNA restriction analysis.

A rapid procedure for the identification of cultured Mycobacterium isolates, based on the combination of enzymatic amplification and restriction analysis, is described. The 16S rRNA genes (rDNA) of 99 strains belonging to 18 different species of the genus Mycobacterium were enzymatically amplified. Amplified rDNA restriction analysis with the enzymes CfoI, MboI, and RsaI was carried out. The combination of the amplified rDNA restriction analysis patterns obtained after restriction with CfoI and MboI enabled differentiation between Mycobacterium asiaticum (number of strains = 4), M. avium (n = 22), M. chelonae (n = 5), M. flavescens (n = 1), M. fortuitum (n = 6), M. gordonae (n = 6), M. intracellulare (n = 13), M. marinum (n = 7), M. nonchromogenicum (n = 1), M. simiae (n = 5), M. terrae (n = 5), the M. tuberculosis complex (n = 11), and 2 of 4 strains of M. xenopi. Further restriction with RsaI was necessary to differentiate between the species M. kansasii (n = 5), M. scrofulaceum (n = 4), and the 2 other M. xenopi strains. The M. avium-M. intracellulare complex was characterized by a specific MboI pattern, and M. avium and M. intracellulare strains could further be differentiated by restriction with CfoI. The whole procedure, including sample preparation prior to the polymerase chain reaction, can be carried out within 8 h, starting from a pure culture.