Understanding the diagnosis of catheter-related bloodstream infection: real-time monitoring of biofilm growth dynamics using time-lapse optical microscopy

Abstract

Background: The differential time to positivity (DTTP) technique is recommended for the conservative diagnosis of catheter-related bloodstream infection (C-RBSI). The technique is based on a 120-minute difference between microbial growth in blood drawn through the catheter and blood drawn through a peripheral vein. However, this cut-off has failed to confirm C-RBSI caused by Candida spp. and Staphylococcus aureus.

Objective: We hypothesized that the biofilm of both microorganisms disperses faster than that of other microorganisms and that microbial load is rapidly equalized between catheter and peripheral blood. Therefore, our aim was to compare the biofilm dynamics of various microorganisms.

Methods: Biofilm of ATCC strains of methicillin-resistant Staphylococcus epidermidis, methicillin-susceptible S. aureus, Enterococcus faecalis, Escherichia coli and Candida albicans was grown on silicon disks and analyzed using time-lapse optical microscopy. The time-lapse images of biofilms were processed using ImageJ2 software. Cell dispersal time and biofilm thickness were calculated.

Results: The mean (standard deviation) dispersal time in C. albicans and S. aureus biofilms was at least nearly 3 hours lower than in biofilm of S. epidermidis, and at least 15 minutes than in E. faecalis and E. coli biofilms.

Conclusion: Our findings could explain why early dissemination of cells in C. albicans and S. aureus prevents us from confirming or ruling out the catheter as the source of the bloodstream infection using the cut-off of 120 minutes in the DTTP technique. In addition, DTTP may not be sufficiently reliable for E. coli since their dispersion time is less than the cut-off of 120 minutes.

Keywords: biofilm; catheter-related bloodstream infections; differential time to positivity; growth; time-lapse optical microscopy.

Figure 1

Schematic representation of the in vitro biofilm formation procedure and time-lapse images analysis.

Figure 2

(A) Time-lapse images (63× magnification) showing the development of biofilms of Candida albicans ATCC14053, Staphylococcus aureus ATCC29213, Staphylococcus epidermidis ATCC35984, Enterococcus faecalis ATCC35186 and Escherichia coli ATCC25213. Growth dynamics of biofilm on silicon disks after 1 hour of incubation under orbital shaking shown as time-lapse images taken hourly during the 17-hour follow-up. (B) The moment of detachment of cell clusters and/or single cells (red arrows) from the biofilms indicating the dispersal phenomenon. The scale bar represents 50 µm to (C) albicans and 15 µm to S. aureus, S. epidermidis, (E) faecalis and (E) coli. (C) Graphical representation of the growth dynamics of Candida albicans biofilm, Staphylococcus aureus biofilm, Staphylococcus epidermidis biofilm, Enterococcus faecalis biofilm and Escherichia coli biofilm. Data shown biofilms on three different areas of silicon disks by hourly quantification of biofilm thickness in µm over 17 hours. Mean ± SEM is shown. The y-axis shows biofilm thickness in µm. The x-axis shows the monitoring time in hours. Significant differences respective to the first time point (*, p<0.05, Student´s t test). (D) Biofilm dispersal time. The graph shows the comparison of biofilm dispersal times for each strain studied in minutes. These are indicated in the figure legend and include Candida albicans, Staphylococcus aureus, Staphylococcus epidermidis, Enterococcus faecalis and Escherichia coli. The longest dispersal time was recorded for S. epidermidis, followed by E. faecalis (above 120 min in both cases). Data are shown as the mean ± SEM from three different areas of three silicon disk.