We report the first case of Weissella confusa bacteremia in an allogeneic hematopoietic stem cell transplant patient. After engraftment and discharge, the patient returned with fever and graft failure and was started on an empiric regimen of aztreonam and vancomycin. A blood culture grew an alpha-hemolytic, gram-positive coccus forming pairs and chains, originally thought to be a viridans Streptococcus and a skin contaminant. The isolation of the organism from multiple blood cultures, and the presence of vancomycin resistance prompted identification and additional susceptibility testing. The RapID(™) Str panel, which has W. confusa in its database, provided multiple incorrect identifications. The MicroScan WalkAway 96 SI, using PC-20 or -29 panels, also did not identify this bacterium, because it is not in their database. The organism was identified as W. confusa by 16S rDNA sequencing. Antibiotic susceptibility determination by Etest revealed vancomycin resistance and daptomycin susceptibility. Therapy was changed to daptomycin, and the infection resolved. Additionally, W. confusa sepsis, with multiple positive blood cultures, developed in a patient in the burn unit at our medical center. The patient's blood cultures remained positive until vancomycin was discontinued and daptomycin therapy initiated. Infections with vancomycin-resistant, gram-positive cocci are emerging among immuno compromised hosts. Under appropriate circumstances, clinicians need to request that the laboratory perform susceptibility testing and accurate identification, by nucleic acid sequencing if necessary. Sequencing of 16S rDNA is an important tool in the accurate identification of unusual pathogens.

Molecular method of 16S rRNA sequencing is reported to be helpful in the accurate identification of organisms with ambiguous phenotypic profiles. We analyzed the use of 16S rRNA sequencing method to identify clinically significant, "difficult-to-identify" bacteria recovered from clinical specimens, and evaluated its role in patient management and consequent clinical outcome. Among the 172 "difficult-to-identify" bacteria recovered over a 4-year period, 140 were gram-positive cocci or gram-negative bacilli; identification by 16S rRNA did not play a role in the management of patients infected with these bacteria. From 32 patients, 33 "difficult-to-identify" gram-positive bacilli were identified; the organisms were mycobacteria, Nocardia, Tsukamurella, Rhodococcus, and Gordonia. In 24 patients for whom clinical data were available, results from the 16S rRNA sequencing method led to treatment change in 14 immunocompromised patients (including 7 hematopoietic stem cell recipients and 1 liver transplant recipient). Therapy was modified in 9 patients, initiated in 3 patients, and discontinued in 2 patients. Most patients' therapy was switched to oral antibiotics with discontinuation of intravascular catheters, facilitating early hospital discharge. All 14 patients were alive 30 days after infection onset. The present study demonstrates the clinical application of 16S rRNA sequencing method to identify "difficult-to-identify" mycobacteria and other gram-positive bacilli in clinical specimens, particularly in immunocompromised hosts.

Although a trend toward less ventilator and hospital days in the ibuprofen group was observed in this pilot study, the differences were not statistically significant. The incidence of BPD was similar in both groups. In the study group, two infants developed gastrointestinal complications and a third infant experienced reversible renal failure. The plasma ibuprofen levels were low. Further studies are needed to assess the use of ibuprofen for the prevention and/or treatment of BPD in preterm infants.

No abstract