Infections caused by coryneform bacteria in pediatric oncology patients.

BACKGROUND: Invasive infections caused by coryneform bacteria are uncommon but have been reported with increasing frequency in recent decades, especially in immunocompromised persons. Because pediatric experience is limited, we examined the epidemiology and clinical characteristics of these infections in children undergoing cancer therapy. METHODS: Using strict case definitions, 17 coryneform bacterial infections were identified in 16 children during a 13-year period; there were 12 episodes of bacteremia and 5 skin or soft tissue infections. RESULTS: The median age of children with bloodstream infections was 11.2 years, and that of children with skin or soft tissue infections was 3.5 years. Most were receiving cancer therapy at the time of their infections, were outpatients at the onset of their infections, had central venous catheters, and were not neutropenic. No patient died as a result of infection and most had relatively mild signs and symptoms. All patients responded promptly to antimicrobial therapy and, although 3 infections relapsed, there was only 1 serious complication. The most common species isolated were Corynebacterium striatum, C. amycolatum, and Microbacterium species. CONCLUSIONS: The epidemiologic and clinical features of coryneform bacterial infections in immunocompromised children differ in several important respects from the previously reported characteristics of these infections in adults.

Identification of clinical coryneform bacterial isolates: comparison of biochemical methods and sequence analysis of 16S rRNA and rpoB genes.

We compared the relative levels of effectiveness of three commercial identification kits and three nucleic acid amplification tests for the identification of coryneform bacteria by testing 50 diverse isolates, including 12 well-characterized control strains and 38 organisms obtained from pediatric oncology patients at our institution. Between 33.3 and 75.0% of control strains were correctly identified to the species level by phenotypic systems or nucleic acid amplification assays. The most sensitive tests were the API Coryne system and amplification and sequencing of the 16S rRNA gene using primers optimized for coryneform bacteria, which correctly identified 9 of 12 control isolates to the species level, and all strains with a high-confidence call were correctly identified. Organisms not correctly identified were species not included in the test kit databases or not producing a pattern of reactions included in kit databases or which could not be differentiated among several genospecies based on reaction patterns. Nucleic acid amplification assays had limited abilities to identify some bacteria to the species level, and comparison of sequence homologies was complicated by the inclusion of allele sequences obtained from uncultivated and uncharacterized strains in databases. The utility of rpoB genotyping was limited by the small number of representative gene sequences that are currently available for comparison. The correlation between identifications produced by different classification systems was poor, particularly for clinical isolates.

Cross-reactivity of non-Aspergillus fungal species in the Aspergillus galactomannan enzyme immunoassay.

The Aspergillus galactomannan enzyme-linked immunosorbant assay (EIA) has been demonstrated to facilitate rapid and sensitive detection of invasive aspergillosis. However, test specificity has not been fully evaluated in non-Aspergillus fungal species. Of 53 fungal isolates, cross-reactivity was observed with 5 non-Aspergillus spp.: Blastomyces dermatitidis, Nigrospora oryzae, Paecilomyces lilacinus, Penicillium chrysogenum, and Trichothecium roseum.

Optimizing blood culture practices in pediatric immunocompromised patients: evaluation of media types and blood culture volume.

BACKGROUND: The optimal use of blood cultures to determine the etiology of febrile episodes in neutropenic children has not been well-defined. METHODS: Single volume blood cultures using the Pediatric ISOLATOR System (ISO), were compared with variable, weight-based culture volumes using the BACTEC 9240 Culture System (BAC). Additionally the value of routinely inoculating the BACTEC MYCO/F LYTIC culture vial (MFL) as well as the BACTEC AEROBIC/F culture vial (AF) was examined. RESULTS: A total of 2620 cultures had both ISO and BAC inoculated; 182 cultures were positive (7.0% of cultures); 97.8% of positive cultures were detected by the BAC (AF and/or MFL) vs.46.2% detected by the ISO. The advantage of the BAC over the ISO was statistically significant for overall recovery of isolates and bloodstream infections, including most individual organism categories. There were only two instances (one each of histoplasmosis and candidemia) in which a blood stream infection was detected by ISO only. All the isolates judged to be contaminants were recovered by BAC only. AF detected significantly more coagulase-negative Staphylococcus spp. than the MFL. Of the isolates 16%, representing 14% of the bloodstream infections (including Gram-negative infections), were detected by the MFL only. Infections were detected more quickly by BAC than by ISO (P < 0.0001). Among the BAC media types, AF was faster than MFL (P < 0.0001). CONCLUSIONS: Optimal yield of blood cultures in immunocompromised pediatric patients included the use of BAC with a weight-based, graduated volume of culture inoculation and routine use of both AF and MFL.