Chryseobacterium meningosepticum is a ubiquitous Gram-negative bacillus historically associated with meningitis in premature neonates. We report 15 positive cultures and 6 cases of infection among immunocompromised adults at our institution over a 10-year period and review the English-language literature on C. meningosepticum. Excluding the present series, there are 308 reports of positive cultures in the literature, of which 59% were determined to represent true infections. Sixty-five percent of those infected were younger than 3 months of age. Meningitis was the most common infectious syndrome among neonates, seen in 84% of cases and associated with a 57% mortality rate. Less commonly reported infections among infants included sepsis (13%) and pneumonia (3%). Pneumonia was the most frequent infection among the postneonatal group, accounting for 40% of cases, followed by sepsis (24%), meningitis (18%), endocarditis (3%), cellulitis (3%), abdominal infections (3%), eye infections (3%), and single case reports of sinusitis, bronchitis, and epididymitis. The 6 cases in our series were all adults, with a mean age of 58.7 years. Sites of C. meningosepticum infection were limited to the lungs, bloodstream, and biliary tree. Infection in our series was associated with prolonged hospitalization, prior exposure to multiple antibiotics, and host immunocompromise, particularly neutropenia. C. meningosepticum is resistant to multiple antibiotics, and disk dilution is notoriously unreliable for antibiotic sensitivity testing. Sensitivity testing on the 15 isolates from our institution revealed the most efficacious antibiotics to be minocycline (100% sensitive), rifampin (93%), trimethoprim-sulfamethoxazole (67%), and ciprofloxacin (53%). In contrast to reports in the literature, the isolates in our series displayed widespread resistance to vancomycin (100% resistant or intermediately sensitive), erythromycin (100%), and clindamycin (86%). These findings have important implications for the clinician when choosing empiric antibiotic regimens for patients with risk factors for C. meningosepticum infection.

No abstract

We examined the effect of nucleic acid extraction methods on the analytic characteristics of a quantitative polymerase chain reaction (PCR) assay for cytomegalovirus (CMV). Human serum samples were extracted with 2 automated instruments (BioRobot EZ1 and QIAsymphony SP, QIAGEN, Valencia, CA) and CMV PCR results compared with those of pp65 antigenemia testing. Both extraction methods yielded results that were comparably linear and precise, whereas the QIAsymphony SP had a slightly lower limit of detection (1.92 log(10) copies/mL vs 2.26 log(10) copies/mL). In both cases, PCR was more sensitive than CMV antigen detection, detecting CMV viremia in 12% (EZ1) and 21% (QIAsymphony) of antigen-negative specimens. This study demonstrates the feasibility of using 2 different extraction techniques to yield results within 0.5 log(10) copies/mL of the mean value, a level that would allow for clinical comparison between different laboratory assays.

Vancomycin-intermediate Staphylococcus aureus (VISA) organisms have minimum inhibitory concentrations (MICs) of 4 to 8 microg/mL and are often associated with vancomycin treatment failure. Detection of VISA has remained problematic. A comparison of 4 methods to detect VISA was done. Of the 20 VISA isolates, the Clinical and Laboratory Standards Institute (CLSI) broth microdilution method yielded susceptible end points of 2 microg/mL for 7, MicroScan (Siemens Healthcare Diagnostics, West Sacramento, CA) for 2, Trek Sensititre (Trek Diagnostic Systems, Cleveland, OH) for 1, and Etest (AB Biodisk North America, Piscataway, NJ) for none. Comparison with the CLSI method showed essential agreement for 95% or more for the Etest, MicroScan, and Trek methods; categorical agreement was as follows: Etest, 60%; MicroScan, 65%; and Trek, 60%. Reliance on a single automated method for determining vancomycin MICs could lead to misclassification of some VISA isolates as vancomycin susceptible. At least 2 methods, including the Etest, should be used when confirming VISA because of slight differences in results from different methods around the end points of 2 and 4 microg/mL .

The VERSANT HCV Genotype 2.0 Assay (LiPA 2.0; Innogenetics, Ghent, Belgium; distributed by Siemens Medical Solutions Diagnostics, Tarrytown, NY) is a new-generation line-probe assay that simultaneously detects sequences in the 5' untranslated (5'UTR) and core regions to genotype and subtype hepatitis C virus (HCV). We tested 60 specimens of known genotype and subtype and 2 specimens with mixed infections with the LiPA 2.0 assay. After arbitration based on genotype and subtype determined by sequencing, there was concordance in 58 of 60 specimens (specificity, 96.7%). Computer-assisted typing yielded comparable results, but much more rapidly. Of 67 clinical specimens, 64 readily yielded genotype and subtype; 3 indeterminate specimens were typed by sequencing and were uncommon types not in the database. The newgeneration line-probe assay that detects the 5'UTR and core regions to genotype and subtype HCV is applicable to more than 95% of specimens. Interpretation is facilitated by computer-assisted analysis.