Molecular analysis of bacterial diversity on central venous catheters recovered from cancer patients

Abstract

Central venous catheters (CVC) are a known source of nosocomial blood stream infections. However, standard cultivation methods to identify causal pathogens of catheter-related blood stream infections (CRBSI) on catheters often fail. In this study, I compare the traditional cultivation method to the cultivation-independent PCR-DGGE method to examine bacterial colonization on central venous catheters retrieved from cancer patients. In order to study the biofilm communities that colonize the catheters, an optimized sonication protocol was developed to remove biofilm bacteria from their substratum. I showed that a sonicating power of 12 Watts for 5 minutes could remove bacteria, such as Pseudomonas putida and Staphylococcus epidermidis cells, from a glass surface without killing the bacteria. Twenty-four catheters were retrieved from cancer patients and used in this study. Five out of the 24 catheter samples (21%) showed growth in at least one of the culture media used. The isolates recovered from the catheters belonged to five bacterial species, including Staphylococcus aureus, S. epidermidis, S. hominis, a Staphylococcus sp., and Streptococcus agalactica. The PCR-DGGE method showed that 100% of the catheters were colonized by bacteria. Furthermore, unlike the cultivation assay demonstrated that contaminated catheters were colonized by only one or two bacterial species, DGGE analysis showed that all the catheters in this study possessed a mixed-bacterial community of at least 4 bacterial species. By analyzing the sequences of the DGGE bands, Enterococcus faecalis (100%), E. faecium (96%) and Roseomonas sp. (88%) were the most predominant species detected on the catheters. Staphylococcus epidermidis, S. hominis, S. aureus, Corynebacterium sp. and Serratia sp. were detected on 30-54% of the catheter samples. Other bacterial species, such as E. coli, Aeroccous sp.. Micrococcus sp. and alpha proteobacterium were detected on less than 30% of the catheters. In addition, scanning electron microscopy (SEM) confirmed the DGGE findings in that biofilms were found on all the catheter samples. Biofilms were found in the lumens of the catheters but rarely associate with the outer surface. The catheters were rinsed prior to the sonication process and the rinsing buffers were also analyzed by culturing and DGGE assays as described for the catheter samples. None of the buffer samples showed any positive growth in the growth media but all of them showed positive amplification by a pair of eubacterial universal primers and showed similar DGGE profiles as their respective catheter samples, showing that even gentle rinsing could remove enough biofilm cells for PCR-DGGE analysis. In conclusion, the PCR-DGGE method is a superior method in identifying biofilm microbial communities on catheters. This method could detect and identify “viable but non-culturable” bacteria that were missed by the conventional cultivation methods. Furthermore, this study also revealed that some bacterial species (such as Enterococcus, Roseomonas spp. and Serratia spp.), that have yet been recognized as a major cause of catheter related infections, can potentially be important catheter-associated pathogens.