This assignment is about providing recommendations on how to prevent central line associated blood-stream infections.
O R I G I N A L A R T I C L E
Reducing Bloodstream Infection Risk in Central and Peripheral Intravenous Lines: Initial Data on Passive Intravenous Connector Disinfection Michelle DeVries, MPH, CIC Patricia S. Mancos, BS, SM(ASCP), CIC Mary J. Valentine, MSN, RN, CNS, OCN
Methodist Hospitals, Gary, IN
Abstract Background: Although few facilities focus on it, bloodstream infection (BSI) risk from peripheral intravenous catheters (PIVs) may exceed central line-related risk. Over a 6-year period, Methodist Hospitals substantially reduced BSIs in
patients with central lines but not in patients with PIVs. A practice audit revealed deficiencies in manual disinfection of
intravenous connectors, thereby increasing BSI risk. Methodist thus sought an engineered approach to hub disinfection
that would compensate for variations in scrubbing technique.
Methods: Our institution involved bedside nurses in choosing new hub disinfection technology. They selected 2 devices to trial: a disinfection cap that passively disinfects hubs with isopropyl alcohol and a device that friction-scrubs with
isopropyl alcohol. After trying both, nurses selected the cap for use in the facility’s 3 intensive care units. After no BSIs
occurred during a 3-month span, we implemented the cap throughout the hospital for use on central venous catheters;
peripherally inserted central catheters; and peripheral lines, including tubing and Y-sites.
Results: Comparing the postintervention period (December 2011-August 2013) to the preintervention span (September 2009-May 2011), the BSI rate dropped 43% for PIVs, 50% for central lines, and 45% overall (PIVs þ central lines). The central line and overall results are statistically significant. The PIV BSI rate drop is attributable to cap use alone because
the cap was the only new intervention during the postimplementation period. The other infection reductions appear to be
at least partly due to cap use.
Conclusions: Our institution achieved substantial BSI reductions, some statistically significant, by applying a disinfection cap to both PIVs and central lines.
Keywords: disinfection cap, IV connector, bloodstream infection, CRBSI, CLABSI
Background
T he risk of catheter-related bloodstream infections (CRBSIs) associated with use of a central intravenous (IV) line or peripherally inserted central catheter
(PICC) is well known and well documented. Although there have been documented improvements, these infections are still estimated to affect 250,000 people annually in the United States.1 Because bloodstream infections (BSIs) related to
central lines/PICCs are so dangerous and widespread, they have been the subject of substantial public and private efforts to minimize them.2
Far less attention has been devoted to the BSI risk from periph- eralIVcatheters(PIVs),althoughthatriskmaybeatleastaslarge. The individual risk of a BSI occurring in a PIV is lower than in a central line/PICC.3 However, some 150 million PIVs are placed in the United States annually, a number that is much greater than the number of central lines used overall.3 PIVs have a lower infection rate (estimated at 0.5/1,000 peripheral line days), but when this is multiplied by the large number of patients receiving these lines, the total number of patients infected is seen in several studies to approach totals reported for central lines.4
Pujol et al5 examined 150 BSIs in 147 patients over an 18-month span to study the relative BSI risk associated with
Correspondence concerning this article should be addressed to mdevries@methodisthospitals.org http://dx.doi.org/10.1016/j.java.2014.02.002 Copyright � 2014, ASSOCIATION FOR VASCULAR ACCESS. Published by Elsevier Inc. All rights reserved.
2014 j Vol 19 No 2 j JAVA j 87
PIVs versus central lines/PICCs. They actually found slightly more total BSIs related to PIVs (77 BSIs; 51.3%) than to cen- tral venous catheters (73 BSIs; 48.7%), although the results are essentially equivalent. A more extensive study of the relative BSI risk between PIVs and central catheters encompassed 72 hospitals and 14,966 patients in Germany.6 On the day preva- lence was assessed, 5.1% of patients had central lines and 23.9% had noncentral lines. The device-associated incidence rate per 1,000 catheter days was 0.8 for central lines and 0.3 for noncentral lines. In other words, the rate for central lines was about 2.7 times the rate for noncentral linesdbut there were about 4.7 times more noncentral lines than central IVs.6 Thus there were actually more BSIs in noncentral lines than in central lines in that very large sample, although statis- tically these 2 figures should be viewed as fundamentally equivalent numbers.
Many hospitals are engaged in substantial efforts to lower CRBSI rates. Data such as those above suggest that a hospital trying to reduce the incidence and cost of BSIs should target PIVs with preventive efforts similar to those it aims at central lines. Although central lines present the greater risk on a per-line basis, the largest numerical risk is greater with PIVs.
The failure to focus BSI prevention efforts exclusively on central lines can also distort infection reporting. When submit- ting mandated infection reports to the Centers for Medicare and Medicaid Services and state departments of health, hospi- tals must use the Centers for Disease Control and Prevention National Healthcare Safety Network’s protocols for surveil- lance. The definitions for CLABSIs are quite broad and can overclassify infections that may not have a central line as the true source. Tip cultures or site cultures are not included in these federal definitions.
For these reasons, Methodist Hospitals (Gary, IN) has for the past 10 years closely examined CRBSI rates among both patients with central lines (central venous catheters and PICCs) and those with PIVs. Previous internal data based on a 6-year sample demonstrated that up to 21% of our facility’s hospital-acquired, laboratory-confirmed blood- stream infections were in patients with peripheral lines alone. Up to 47% of infections meeting the definition of central line-associated occurred in patients with multiple lines present; the majority of those central line-associated bloodstream infections (CLABSIs) with multiple lines involved a peripheral line as well. During that 6-year period the hospital achieved substantial reductions in CLABSIs throughout the institution, as did hospitals across the coun- try as evidence-based recommendations for risk reduction became more fully adopted. But the hospital did not see the same reduction in patients with peripheral lines only. There was not a focused campaign looking at that particular group during this period.7
A practice audit in 2011 revealed that manual disinfection of the hubs of IV needleless connectors was not being performed properly. The standard manual disinfection method (commonly called “scrub the hub”) requires wiping the hubdusing downward twisting pressuredwith a disinfectant
such as isopropyl alcohol (IPA) for a prescribed number of seconds and then waiting another prescribed number of sec- onds for the alcohol to dry before accessing the IV line. Compliance with the technique (which was mandated to be performed before all line accesses) was satisfactory at Meth- odist Hospitals, but virtually all bedside nurses varied from the technique in some way. Inadequate disinfection of connector hubs increases BSI risk because it enables microor- ganisms to gain entry to the intraluminal surfaces of the IV system and form infection-causing biofilm, which in turn pro- motes infections.8 Noncompliance and variation from proper technique are widely recognized and are not just issues at Methodist Hospitals.9
Methodist sought an engineered approach to disinfecting connector hubs that would overcome the problem with varia- tions in scrubbing technique, which were perceived to be a likely source of infection. This approach could be in the form of a device or devices that would address these lapses in connector disinfection practice and be applicable to both central lines and PIVs.
About the Hospital Methodist Hospitals is a not-for-profit, community-based
health care system with 2 full-service acute care facilities located 1 hour east of Chicago, Illinois. We serve a predomi- nantly urban, economically challenged population. Based on the 2012 annual report, the hospital system has a total of 634 beds, including 504 adult beds as well as beds in 3 adult intensive care units. Services provided include behavioral health, bloodless medicine, a breast care center, cardiovascular services, a diabetes center, emergency services, home health services, a neuroscience institute, an oncology institute, ortho- pedic and spine care services, rehabilitation services, surgical weight loss/bariatric services, wound care services, and women’s and children’s services. From 2003 through 2013, the hospital system experienced
multiple turnovers in senior leadership, a workforce reduction, and patient satisfaction scores below the 10th percentile. Physician and nursing trust reached an all-time low. After new senior leadership was in place, the hospital identified key improvements and implemented a turnaround plan that reinvested in improving the patient experience, updating equipment, and staff development. Nursing turnover during this time frame remained high. New graduate staff were brought on board and were a welcome addition, but skill devel- opment remained a challenge.
Methods Methodist Hospitals sought to involve bedside nurses in its
efforts to identify and implement preventive technology for 2 reasons. First, it was believed that nurses would be more accepting of a new device if their input was central to its selec- tion. Second, the nurses believed they were partly responsible for BSIs that may have resulted from their errors in executing the scrub-the-hub method. They wanted an opportunity to participate in formulating a less error-prone approach to connector hub disinfection.
88 j JAVA j Vol 19 No 2 j 2014
The nurses suggested several device technologies that might address the issues with connector hub disinfection based on what they had used at other institutions and additional options provided by the materials management department. From those candidates, 2 were selected for testing in the health system’s 3 intensive care units. The 2 devices were a disinfection cap (SwabCap, Excelsior Medical, Neptune, NJ) that passively dis- infects the hub with IPA and a device that scrubs the connector hub with IPA while applying friction (Site-Scrub, Bard Access Systems, Salt Lake City, UT).
Following the trials, the nurses stated a preference for the disinfection cap (see Figure 1), so it was selected for continued use. Throughout this intervention, the policy remained consis- tent regarding peripheral line changes at 96 hours and the use of scrub-the-hub techniques when contact time between a disinfection cap and an access device/Y-site was <5 minutes. The cap was chosen because of multiple attributes that address the issues with manual hub disinfection:
d The cap tends to eliminate variation because it twists onto the connector/hub threads like a lid on a jar, and as such can only be readily applied 1 way.
d The orange color makes it easy to confirm compliance. That is, if a supervisor or bedside nurse observes that a connector has been capped with the device, she knows disinfection has occurred.
d The sterile-package cap supports nursing practice because it is much easier and quicker to apply than it is to perform the standard scrub-the-hub routine.
d The cap provides improved disinfection compared with the scrub-the-hub method because it continuously bathes the hub in IPA while it is attached (with a seal that pre- vents IPA from evaporating or leaking out).
d The cap is designed to be left in place between line ac- cesses. While it is attached to the hub, it is protecting the hub against touch and airborne contamination, which is a preventive attribute that even perfectly executed manual disinfection cannot provide.
After its selection by bedside nurses, the disinfection cap was used in each of Methodist’s 3 intensive care units for a 3-month span. No BSIs occurred during this period. Because of this record of successful use, the hospital implemented use the disinfection cap throughout the hospital system for application on central/PICC and peripheral lines, including tubing and Y-sites. There was significant discussion about which devices and access points should be protected with the cap. Ultimately the nursing staff, in conjunction with infec- tion control department staff, decided the best application was to be consistent across all devices and all potential access points. Selective use of other devices had not been a successful approach in the past. Since housewide implementation of the disinfection cap, an
ongoing prospective observational study has been conducted on the cap’s use. Regarding internal review board approval of this study, Methodists’ medical staff and board of directors are responsible for approving all process and outcome surveil- lance at the institution. The approval process takes place on an annual basis. The data reported are within the scope of this approval and did not require formal approval by the institu- tional review board.
Results The data reported compares the 21-month postintervention
period of December 2011-August 2013dwhen the cap was being used housewidedwith the preintervention period of September 2009-May 2011, which is also 21 months long. For PIVs alone, the BSI rate dropped 43%, comparing the pre- intervention rate of roughly 0.010/100 patient days to the rate of approximately 0.0059/100 patient days during the postinter- vention period (see Figure 2). Although not statistically signif- icant (P ¼ .078), this substantial change encompasses a total of 373,394 patient daysd185,950 during the preimplementation period and 187,444 during the postimplementation span. The decreased infections appear to be attributable to the influence of the cap alone, because unlike with central lines the cap was the only new intervention applied during the postimple- mentation period. The data regarding the effect of the cap on CLRBSIs show a
similar reduction. After SwabCap use was implemented, the BSI rate dropped 49.3% from the preintervention rate of about 0.075/100 patient days. The postintervention rate is roughly 0.038/100 patient days (See Figure 3). This result is statisti- cally significant (P < .00037). (Methodist uses central-line days as its denominator in its intensive care units. At the time of the above-reported data collection, it did not have that denominator data available for the other hospital units where its infection control professionals conduct surveillance. Therefore, the data was reported using the widely available “patient days” denominator.) Putting these 2 results together, the combined BSI rate for
PIVs and central lines decreased 50.0%. The rate dropped from the preintervention figure of approximately 0.086/100 patient days to the postintervention rate of about 0.043/100 patient days (see Figure 4). The combined rate reduction was statistically significant (P < .000001) and was at least partly
Figure 1. Disinfection cap (SwabCap; Excelsior Medical, Neptune, NJ).
2014 j Vol 19 No 2 j JAVA j 89
due to the disinfection cap. The reduction cannot be entirely attributed to cap use because the cap was not the only interven- tion for central lines during this time. Methodist’s nurses who insert PICCs played an increased role in central line moni- toring and dressing changes, per the recommendation of a multidisciplinary team charged with improving prevention of CRBSIs. This may have played a role in the BSI rate reduction for central lines and thus would be a factor in the overall rate reduction, as well.
Discussion Although several device technologies are available to
address deficiencies of the scrub-the-hub technique, the nurses at our institution expressed a clear preference for a de- vice that appeared to comprehensively compensate for those deficiencies. They rejected a device that only addressed a few of the concerns about the method: first, nurses’ possible failure to apply appropriate pressure for the prescribed time while wiping; and second, inability to reach the recessed
Figure 3. Effect of disinfection cap on central line-associated bloodstream infection (BSI) rates.
Figure 2. Effect of disinfection cap on peripheral intravenous catheter-associated bloodstream infection (BSI) rates.
90 j JAVA j Vol 19 No 2 j 2014
surfaces of the connector. Unlike the friction scrub device, the disinfection cap does not assist nurses with applying proper pressure. But it continually bathes the connector hub in alcohol after scrub-the-hub technique has been executed. By also offering protections that the scrub-the- hub method alone cannot provide, the cap may offer an improvement over scrub-the-hub method alone in both worst-case (variation from and/or noncompliance with the scrub-the-hub routine) and best-case (no variation or noncompliance) scenarios.
The preimplementation shortcomings in connector disin- fection and the efficacy of the disinfection cap point to an interesting and important anomaly in public and private ef- forts to minimize CLABSIs. It is widely recognized that the scrub-the-hub method is problematic and thus may be a causative factor when an institution has an elevated CLABSI rate. But this issue is not addressed in many of the most widely promoted central line bundles (ie, groups of evidence-based practices). For example, it is not mentioned at all in the central line bundle of the Institute for Healthcare Improvement.10 In the central line practices recommended by the Society for Healthcare Epidemiology of America and Infectious Diseases Society of America, the scrub-the-hub technique is listed with no acknowledge- ment that the practice may be compromised by technique variation and noncompliance.11
Our data also show the importance of knowing one’s own institution’s data and implementing solutions that are the best fit to the current situation. The Centers for Disease Control and Prevention has a Category 1A recommendation to evaluate compliance with insertion and maintenance pro- tocols for all staff involved with intravascular devices. However, it is commonplace for hospitals to not perform such process monitoring on peripheral linesdparticularly
those outside the critical care setting. Although there is research showing that intraluminal colonization may be a bigger concern after the first week, our outcomes suggest that practices within a hospital can greatly affect this risk factor. Institutions will benefit from directly observing practices at the bedside to identify their unique opportu- nities for improvement. Bundles such as the Society for Healthcare Epidemiology of
America and Infectious Diseases Society of America bundle emphasize the scrub-the-hub method because infection control experts recognize that hub disinfection is crucial for preventing bacteria entry into the connector’s intraluminal space. But it seems increasingly likely that recommending this practice does not sufficiently address the BSI threat. Evidence for the efficacy of the disinfection cap solution has only recently started to appear in the peer-reviewed, published literature.12
Multiple institutions have reported on their successes using a disinfection cap to improve catheter maintenance and address CRBSIs.13-26
Finally, as demonstrated above, BSI risk for PIVs is sub- stantial and may sometimes even be comparable to the risk for central lines, especially when the sheer number of PIVs placed is considered. One advantage of both the disinfection cap and the friction scrub device is that the technologies can be equally applied to PIV and central lines.
Limitations Ours was a prospective observational study that followed an
intervention. It was not a randomized, controlled trial. Other limitations include the fact that the study took place at a single institution and the related fact that the sample size was small. In addition, the improved BSI rates we achieved follow-
ing the interventions occurred in an environment where defi- ciencies with manual disinfection of IV connector hubs had
Figure 4. Effect of disinfection cap on overall (central line þ peripheral intravenous catheter) bloodstream infection (BSI) rates.
2014 j Vol 19 No 2 j JAVA j 91
been identified and relatively high rates of BSIs had been recorded. Hospitals that have substantially lower rates of infec- tion and/or have confirmed that variance and noncompliance with their hub disinfection protocol are not an issue may not experience the same degree of improvement that we did. Finally, compliance with disinfection cap use was monitored intermittently, not continuously. Therefore, it is not possible to predict from our study the level of compliance associated with the reported outcomes.
Conclusions The large number of PIVs placed in the United States annu-
ally and the scant attention paid to the BSI risk associated with these lines constitute an urgent situation in IV therapy. Meth- odist Hospitals achieved statistically significant BSI reductions by applying a disinfection cap to both PIVs and central lines. These results point toward the cap as an effective means of mitigating that risk. Although the published evidence base for the efficacy of a disinfection cap is small, our study pro- duced results that are consistent with previously published research.
It should be noted that several institutions have reached their goal of eliminating CRBSIs after implementing the same disinfection cap used at Methodist. The fact that we have not yet completely eliminated CRBSIs is not a comment about the cap’s effectiveness. Instead, it suggests that although the cap appears to have addressed our issues with connector hub disinfection, we need to take further preven- tive steps to address issues that the cap is not able to influen- cedin particular, issues involving the extraluminal route to infection. The cap has been an important addition to our pro- tocols for preventing CRBSIs.
The promising outcomes we found need to be validated with data from other institutions. Should future studies yield similar data, a disinfection cap should be considered for inclusion in the recommended bundles of professional organizations and for use on PIVs, as well, particularly by institutions whose BSI rates are persistently above national benchmarks or that are otherwise having compliance or variance problems with connector hub disinfection.
Disclosures Excelsior Medical provided free disinfection caps for the
comparative product trials discussed in the article, as well as writing support for the preparation of the article. The authors are responsible for all content including the data.
References 1. O’Grady NP, Alexander M, Burns LA, et al. Guidelines for
the prevention of intravascular catheter-related infections. Clin Infect Dis. 2011;52(9):e162-e193.
2. Centers for Disease Control and Prevention. Vital signs: central lineeassociated blood stream infectionsdUnited States, 2001, 2008, and 2009. MMWR Morb Mortal Wkly Rep. 2011;60:1-6.
3. Mestre G, Berbel C, Tortajada P, Alarcia M, Coca R, Fernández MM, et al. Successful multifaceted intervention
aimed to reduce short peripheral venous catheter-related adverse events: a quasiexperimental cohort study. Am J Infect Control. 2013;41(6):520-526.
4. Maki DG, Kluger DM, Crnich CJ. The risk of bloodstream infection in adults with different intravascular devices: a systematic review of 200 published prospective studies. Mayo Clin Proc. 2006;81(9):1159-1171.
5. Pujol M, Hornero A, Saballs M, Argerich MJ, Verdaguer R, Cisnal M, et al. Clinical epidemiology and outcomes of peripheral venous catheter-related bloodstream infections at a university-affiliated hospital. J Hosp Infect. 2007; 67(1):22-29.
6. Wischnewski N, Kampf G, Gastmeier P, et al. Prevalence of primary bloodstream infections in representative German hospitals and their association with central and pe- ripheral vascular catheters. Zentralbl Bakteriol. 1998; 287(1-2):93-103.
7. DeVries MM, Mancos P. Non-central line related labora- tory confirmed bloodstream infection (LCBI). Scientific poster presented at: 2012 Interscience Conference on Anti- microbial Agents and Chemotherapy (IACC); September 9-12, 2012; San Francisco, California.
8. Ryder MA. Catheter-related infections: it’s all about bio- film. Top Advanced Pract Nurs eJ. 2005;3.
9. Moureau N, Dawson RB. Keeping needleless connectors clean, part 1. Nursing. 2010;40(5):18-19.
10. Implementing the IHI central line bundle. Institute for Healthcare Improvement Web site. http://www.ihi.org/ knowledge/Pages/Changes/ImplementtheCentralLineBundle .aspx. Accessed September 30, 2013.
11. Marschall J, Mermel LA, Classen D, et al. Strategies to prevent central lineeassociated bloodstream infections in acute care hospitals. Infect Control Hosp Epidemiol. 2008;29(Suppl 1):S22-S30.
12. Wright MO, Tropp J, Schora DM, et al. Continuous pas- sive disinfection of catheter hubs prevents contamination and bloodstream infection. Am J Infect Control. 2013; 41(1):33-38.
13. McMahon LA. The Magnet� effect: success in minimizing bloodstream infections. http://www.americannursetoday.com/ Article.aspx?id¼10756&fid¼10686. Accessed September 30, 2013.
14. Dobin A, Lee J. Disinfection cap for luer access valves: a key to preventing catheter infections. http://www .infectioncontroltoday.com/articles/2010/11/disinfection-cap -for-luer-access-valves-a-key-to-preventing-catheter-infections. aspx. Accessed September 30, 2013.
15. Leone M, Pratt M. Preventing central line-associated bloodstream infections in home infusion patients through the use of disinfection caps. Infusion. 2011;17(6):10-14.
16. Perry M, Goss L. IV Connector’s role in an initiative to reduce bloodstream infections among vulnerable oncology patients. Scientific Poster presented at: Association for Professionals in Infection Control and Epidemiology 40th Annual Conference; June 8-10, 2013; Ft Lauderdale, Florida.
92 j JAVA j Vol 19 No 2 j 2014
17. Pavia M. Testing elimination of an infection prevention device from catheter bundle, and potential effect on overall catheter bloodstream infection rate. Scientific Poster pre- sented at: Association for Professionals in Infection Control and Epidemiology 40th Annual Conference; June 8-10, 2013; Ft Lauderdale, Florida.
18. Posa P. Improving IV connector disinfection by using human factors engineering to identify effective, nurse-friendly solu- tions. Scientific Poster presented at: Association for Profes- sionals in Infection Control and Epidemiology 40th Annual Conference; June 8-10, 2013; Ft Lauderdale, Florida.
19. Stokes P. Evidence-based device for disinfecting IV con- nectors proves effective addition to anti-bloodstream infec- tion practices and education. Scientific Poster presented at: Association for Vascular Access 25th Annual Scientific Meeting; October 3-6 2011; San Jose, California.
20. Simpser EF. Supplementing central line bundle with evidence-based devices and protocol reduces bloodstream infections in vulnerable patient population. Scientific Pos- ter presented at: Association for Vascular Access 25th Annual Scientific Meeting; October 3-6, 2011; San Jose, California.
21. Bor B, Johnson CJ, Noble C, Miller BR. It takes a village to prevent central venous catheter infections and promote safety of patients. Scientific Poster presented at: Association for Vascular Access 26th Annual Scientific Meeting; October 16-19, 2012; San Antonio, Texas.
22. McCalla S, Greco J, Warren M, Byrne P, Bogetti J. Inte- grated delivery system of disinfection cap and flush sy- ringe, plus staff education, reduce bloodstream infections and treatment costs. Scientific Poster presented at: Associ- ation for Vascular Access 26th Annual Scientific Meeting; October 16-19, 2012; San Antonio, Texas.
23. Seiler RL, Meyer S. Improving infection control compli- ance using combined cap/flush syringe technology to reduce central line associated bloodstream infections. Scientific Poster presented at: Association for Vascular Access 27th Annual Scientific Meeting; September 20-23, 2013; Nashville, Tennessee.
24. Miller M, Haddad L. Approach improves IV line disinfec- tion, helps reduce bloodstream infections. Scientific Poster presented at: Infusion Nurses Society 2012 Annual Convention & Industrial Exhibition; April 28-May 3, 2012; Las Vegas, Nevada.
25. Lee J. Disinfection cap makes critical difference in central line bundle for reducing CLABSIs. Scientific Poster pre- sented at: Association for Professionals in Infection Con- trol and Epidemiology 38th Annual Conference; June 27-29, 2011; Baltimore, Maryland.
26. Farris P, Mosier W. Adoption of new best practices elim- inates CLABSI in bone marrow transplant unit. Scientific Poster presented at: Association for Professionals in Infec- tion Control and Epidemiology 38th Annual Conference; June 27-29, 2011; Baltimore, Maryland.
2014 j Vol 19 No 2 j JAVA j 93
Copyright of Journal of the Association for Vascular Access is the property of Association for Vascular Access and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.