Russell Olmsted, MPH, CIC is an epidemiologist in Infection Control Services at Saint Joseph Mercy Health System in Ann Arbor, MI.

1. Has your institution/facility identified a facility-specific “bundle”(products, protocols) to reduce the occurrence of CABSI? What are the elements of that bundle?

• The Saint Joseph Mercy Health System (SJMHS) has employed a specific five-item bundle to be used within our system:
• hand antisepsis by the health professional inserting the central line prior to starting the procedure
• preparing the insertion site with chlorhexidine gluconate
• maximal sterile barrier precautions, including sterile gloves, cap, sterile gown, and sterile drape
• avoiding insertion into femoral veins, because of possible mechanical complications and increased risk for infection
• daily assessment of the continued need for the central line, with the idea that the line should be removed as soon as possible.

2. How did your facility determine this bundle, and how was it implemented?

Saint Joseph Mercy Health System, an integrated, regional healthcare provider, participated in a state-wide performance-improvement collaboration led by the Michigan Health & Hospital Association’s (MHA) Keystone Center for Patient Safety & Quality. The Keystone Center coordinated implementation of the central line associated bloodstream infection (CABSI) prevention bundle across a large number of hospitals with intensive care units (ICUs) in Michigan. SJMHS enrolled its four ICUs in this Project. The elements in this bundle were based on scientific evidence.

Our system was an enthusiastic participant in the MHA Keystone Project because we were determined to reduce device-associated risks and improve our patient care performance. That enthusiasm gained us the support of hospital leadership and ICU staff, and we had valuable infrastructure and networking from the MHA and the principal investigator, Dr. Peter Pronovost, who was awarded a grant from the federal Agency for Healthcare Research & Quality (AHRQ) to study the translation of scientific evidence to direct patient care.1 In addition, a coordinator helped each hospital’s panel of subject-matter experts (SMEs) to develop site-specific programs that were then adapted for individual units, such as ICUs, within each hospital.

Implementing the bundle began in February 2004 at each hospital’s pilot sites, and we were pleased to encounter deep and significant participation by direct-care personnel. The momentum of the experience in the ICUs has resulted in expansion of Keystone prevention initiatives in other areas such as surgery and medicine.

We can credit smooth implementation of the bundle to excellent teamwork, strong communication, and enhancing the culture of safety within our system. The focus is on patient safety, and everyone is involved because we have a common understanding of that goal.

3. What are the elements—products and procedures—of your organization’s bundle that have produced positive outcomes?

The Saint Joseph Mercy Health System requires that a designated person—usually a nurse—be responsible solely for ensuring that all elements of the bundle are employed each time that a central line is inserted, and for monitoring for continued need thereafter. A checklist is used during the insertion to ensure that no element of the bundle is forgotten. If there is any difficulty or uncertainty at any stage, the monitor can alert the clinicians involved and rapidly consult with a physician leader on the involved unit to resolve these efficiently. I think that this process is helped by the fact that critical-care nurses have a long history as patient advocates, so this is a comfortable role for them.

a) Can you tell us about site preparation, connectors, flushing protocols, etcetera?

Site preparation is the same whenever a line is installed: follow the checklist. We’ve found it helpful to use prepackaged central line kits that contain many of the necessary items such as drapes, chlorhexidine gluconate, catheter, and insertion device. In our ICUs we have catheter-insertion carts with all the necessary equipment.

Different types of lines call for different connectors or flushing and care protocols, and training is provided each time a new element is introduced. Of note: the CABSI prevention bundle is focused on technique of insertion, but there is also a need to assure good practices for care and maintenance of the vascular device thereafter. We’re in the process of exploring extension of the bundle elements to these aspects.

Manufacturer-prepared saline syringes are the most common method of flushing vascular access devices (VADs) at SJMHS. The sequence of steps for flushing VADs is based on the instructions for use from the manufacturer of the IV (intravenous) equipment.

There is increasing interest in maximizing use of single-dose vials following some reports of cross-transmission in ambulatory care settings. We do, as much as possible, emphasize use of single-dose vials but still have some multidose products in use. For the latter there is a comprehensive pharmacy-services policy that addresses use and dating of these medications.

b) Do you feel that a renewed emphasis on staff education may be necessary to reinforce good practice for antisepsis of valves? of any other elements?

I don’t think that renewed emphasis applies to our situation, because the use of monitors and checklists means that we have continual reinforcement on good practices. We do cover this in new-staff orientation, and existing staff make sure that newcomers implement the bundle plus other aspects of VAD site care and use. Ongoing education and reinforcement are necessary, however, with VADs as there is a multitude of different valve technologies on the market. We also work closely with other departments such as anesthesia and imaging services, since they insert various central lines also.

We have transitioned from a split-septum valve to Luer-activated needleless neutral-pressure valve technology. There are other valves such as Luer-activated positive-pressure devices that are available but we’ve had no direct experience with these. Some have been temporally associated with increased risk of infection, but the evidence is not definitive.

c) What methods have proven most beneficial for you in improving decontamination of IV ports?

We currently advocate using isopropyl alcohol. New research2 suggests that other methods—such as having an antimicrobial inside the valve or even incorporated into the valve—might help prevent contamination but we’ve not considered this further at the current time. However, it is important that providers disinfect the access point prior to each entry.

4. What would you consider the reasons for your success in reducing CABSI? What do you feel still needs to be done?

The Keystone Study says that a 66-percent reduction in incidence of CABSI is achieveable.1 Other large performance-improvement collaboratives have reported similar results.3 Even with implementing the bundle, the infection preventionist (IP) serves a key role of maintaining surveillance for CABSI, as it is an ongoing challenge to eliminate all of these but a goal that has become the goal. In general we’ve seen significant and sustained reductions; for example, one of our units has had no infections for four years. Having monitors and checklists ensures that the bundle is followed. We haven’t needed to use additional measures such as antimicrobial catheters or special dressings, but we have implemented a patient-cleansing process that uses chlorhexidine as an adjunct to the bundle.

Whenever a CABSI does occur, the bundle has altered the conversation wherein there is intense scrutiny of what happened, and there is open communication with the direct-care providers to determine what, if anything, should have been done differently.

5. Many think of blood reflux as a risk factor only for occluded lines. Do you feel enough attention has been paid to blood reflux and the role it can play in CABSI?

Not entirely. Reflux can play a role in CABSI. Our checklist doesn’t address this situation, and occluded lines are an area that can be a significant challenge. We do monitor VADs for patency.

Can you elaborate on the "procedural problem" aspect involved in the use of the devices? Exactly what human factors have you observed or suspect are to blame?

Human-factors engineering points to some of these issues. For example, the range of VAD valve design is extensive, and each specific valve requires slight variations in flush technique; with some, no flush is required. This complexity makes it difficult for the direct-care provider to be aware of which kinds of valves are in use and the correct methods of accessing the different types. A principle of human-factors design is to engineer products and establish environments that simplify care processes so that providers can focus on the needs of patients and not be distracted by an overly complex infusion system.

6. Is there anything else on this topic that you feel should be mentioned?

Clinicians commonly use the term catheter-related bloodstream infection (CABSI). However, the data that most IPs collect relate to CABSI. For inter-facility comparisons, therefore, one needs to be aware of definitions used and how rates of CABSI are being expressed. As with other fields, understanding terminology is important, especially with the growth in public reporting of healthcare-associated infection (HAI) data. Therefore I foresee only continued growth in the need for the IP to be a key member of an interdisciplinary team that is working to optimize patient safety.

References
1.   Pronovost P, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med. 2006 Dec 28;355(26):2725-32
2.   Menyhay SZ, Maki DG. Disinfection of needleless catheter connectors and access ports with alcohol may not prevent microbial entry: the promise of a novel antiseptic-barrier cap. Infect Control Hosp Epidemiol. 2006 Jan;27(1):23-7.
3.   CDC. Reduction in central line-associated bloodstream infections among patients in intensive care units--Pennsylvania, April 2001-March 2005. MMWR Morb Mortal Wkly Rep. 2005 Oct 14;54(40):1013-6.