Differences in Bacterial Transfer and Fluid Path Colonization through Needlefree Connector-Catheter Systems In Vitro

Background: The use of mechanical valve connectors, as opposed to split septum connectors, has been implicated as an infection risk. Differences exist in the transfer of bacteria through the various connectors (C) but the impact of repeated injection of bacteria into the bloodstream and the colonization of the catheter hub (CH) and internal lumen (IL) is unknown.

Objective: The primary objectives of this study were to evaluate differences in transfer of bacteria through the connector-catheter system and biofilm formation in the C, CH and IL.

Methods: Clearlink™, SmartSite®, InVision-Plus®, MaxPlusä, Q-Syteä and MicroCLAVE® connectors were evaluated using an in vitro model designed to simulate clinical use.  Connectors were evaluated in 3 groups of 3 experiments, with 2-3 connectors run side by side in each group. The connector septum was inoculated with 10⁶ CFU Staphylococcus aureus, dried for 30 minutes and then attached to a catheter and flushed with 3 ml sterile saline (SS) which was collected and plated.  The catheter-connector sets were flushed x2 and locked with sterile Brain Heart Infusion Broth (BHI) for 1 hour.  The sets were flushed again x3 with 3 ml SS. This was repeated again x3 for a total of 15 flushes and a BHI lock after the 1st, 3d, and 4th set of flushes for a total of 18 accesses. The last flush was collected and plated. The entire procedure was repeated x5 days. At 72 and 96 hours, 2 connector-catheter sets for each connector type were destructively sampled for bacterial counts and microscopy.

Results: The MicroCLAVE connector had a significantly smaller mean log density (MLD) of bacteria (p<0.0001) when averaged over all flushes, inoculations, days and runs than any other connector.  In contrast, the Q-Syte, a split-septum connector, had a significantly larger MLD (p<0.0001) than the other 5 connectors.  In side-by-side comparisons, MLD of the CH and IL was also significantly smaller for the MicroCLAVE than for the other connectors, except the Clearlink.  At the end of the experiments MicroCLAVE MLD on the CH was less than SmartSite (p=0.0042), InVision-Plus (p=0.001), MaxPlus (p=0.0258), and Q-Syte (p<0.0001).  The MLD on the CH for the Clearlink was less than SmartSite (p=0.0199), while the MaxPlus was less than Q-Syte (p=0.0002).  The MLD of bacteria on IL was also less for MicroCLAVE than SmartSite (p=0.0012), InVision-Plus (p=0.0031), MaxPlus (p=0.0196), and Q-Syte (p=0.0004). The MLD on the IL for the Clearlink was less than SmartSite (p=0.0012).

Conclusions: Significant differences among connectors were observed for bacterial transfer and colonization of the CH and IL.  In general, the split septum connector tested performed poorly compared to the other connectors while the MicroCLAVE performed better than the other connectors.  Overall, the results suggest that connector selection may impact both bacterial transfer to the bloodstream and bacterial colonization of CH and IL.