Comparison of Bloodstream Infection (BSI) Rates Associated with Split Septum, Positive and Negative Mechanical Valve Intravenous Connectors to a Zero Fluid Displacement Connector

Background: Needleless connectors (NC) differ in design features of the external surface, intraluminal fluid pathway and internal mechanisms for pressure at disconnect; these features affect BSI risk.  The SHEA/IDSA Compendium, CDC Draft IV Guideline and the FDA have raised concerns regarding BSI risk associated with positive pressure/displacement mechanical valve NCs.  Comparative studies evaluating the BSI risk associated with different NCs are urgently needed.

Objective: Multicenter comparison of BSI rates associated with use of split septum, positive or negative pressure mechanical valves to a zero fluid displacement intraluminal protection device (IPD).

Methods: Prospective study, 8 states (California, Colorado, Florida, Nevada, New York, Pennsylvania, Tennessee, Texas), 6 settings (ICU, MICU, SICU, long-term care, oncology and hospital-wide) comparing BSI rates associated with split septum (21,987 catheter days), positive (10,067 catheter days) or negative pressure mechanical valves (18,616 days) for a total of 50,670 catheter days to an IPD connector (68,752 catheter days).  BSI definitions were CDC NHSN criteria.  Paired t-tests were used to examine differences between catheter days and BSI rates before and after IPD connector adoption.  Statistical significance was assessed using an alpha level of 0.05.

 Results: Catheter days were similar before and after IPD adoption (Table 1).  There was a statistically significant higher BSI rate when split septum, positive or negative pressure mechanical valve NCs were used, p =0.0008.  Overall, a mean 85% BSI rate decrease was documented after changing to the IPD. 

Conclusions: We documented a significant decrease in BSI rate when split septum, negative or positive pressure mechanical valve NCs were changed to an IPD NC in multiple populations and settings. These data show that NC design does impact BSI rates.

Table 1: Bloodstream infection rates by connector type per 1000 catheter days.

CONNECTOR TYPE	PRE + POST- IPD MONTHS OF DATA COLLECTION	PRE # INFECTIONS	POST # INFECTIONS	PRE-IPD BSI RATE	POST-IPD BSI RATE	INFECTION DECREASE = PRE-POST	AVERAGE DECREASE PER CONNECTOR TYPE AFTER CHANGING TO IPD
Positive Pressure Mechanical Valve							5.16
ICU-CA	12 + 12	11	0	3.7	0.0	3.7
MICU-CO	24 + 24	3	1	4.2	0.8	3.4
MICU-PA	7 + 7	12	4	4.3	1.5	2.8
MICU-NV	12 + 12	28	2	11.4	1.4	10.0
SICU-CO	24 + 24	8	2	7.1	1.2	5.9
Split septum							3.75
ICU-TX	8 + 8	27	1	2.9	0.1	2.8
LONG-TERM-TN	32 + 32	77	5	5.2	0.5	4.7
Negative Pressure Mechanical Valve							2.70
Hospital wide-FL	12 + 12	53	24	3.3	1.2	2.1
Oncology-NY	6 + 6	9	1	3.7	0.4	3.3